Complete Guide to Microsoft Power Automate | Gregor Maric | Skillshare

Playback Speed

  • 0.5x
  • 1x (Normal)
  • 1.25x
  • 1.5x
  • 2x

Complete Guide to Microsoft Power Automate

teacher avatar Gregor Maric, Changing the world with automation

Watch this class and thousands more

Get unlimited access to every class
Taught by industry leaders & working professionals
Topics include illustration, design, photography, and more

Watch this class and thousands more

Get unlimited access to every class
Taught by industry leaders & working professionals
Topics include illustration, design, photography, and more

Lessons in This Class

36 Lessons (3h 54m)
    • 1. Class Introduction

    • 2. Lesson 1.1 - Introduction

    • 3. Lesson 1.2 What is RPA

    • 4. Lesson 1.3 - What is intelligent automation

    • 5. Lesson 1.4 - Flow Walkthrough

    • 6. Lesson 1.5 - Desktop Walkthrough

    • 7. Lesson 1.6 - Difference between power automate cloud and desktop

    • 8. Lesson 2.1 - Introduction

    • 9. Lesson 2.2 - Licencing

    • 10. Lesson 2.3 - Set Up Desktop

    • 11. Lesson 2.4 - Set Up Flow

    • 12. Lesson 2.5 - Set up virtual machine

    • 13. Lesson 3.1 - Introduction

    • 14. Lesson 3.2 Getting Started With Templates

    • 15. Lesson 3.3 - Instant Cloud Flows

    • 16. Lesson 3.4 - Instant Cloud Flows part 2

    • 17. Lesson 3.5 - Automated Flows

    • 18. Lesson 3.6 - Approval workflows

    • 19. Lesson 3.7 - AI Builder

    • 20. Lesson 3.8 - Sentiment Analysis

    • 21. Lesson 3.9 - Build Ai Flows from Template

    • 22. Lesson 3.10 - How to build a custom AI model

    • 23. Lesson 3.11 - How to use your AI model in Flows

    • 24. Lesson 4.1 - Introduction

    • 25. Lesson 4.2 - Getting Started

    • 26. Lesson 4.3 Desktop Automation

    • 27. Lesson 4.4 Web Automation

    • 28. Lesson 4.5 - Excel Automation

    • 29. Lesson 4.6 - Send Emails

    • 30. Lesson 4.7 - AI Services

    • 31. Lesson 4.8 - Building Strong Flows

    • 32. Lesson 5.1 - Lesson Overview

    • 33. Lesson 5.2 - Process Selection

    • 34. Lesson 5.3 - Desktop To Cloud

    • 35. Lesson 5.4 - Final Project

    • 36. Lesson 5.5 - What's Next

  • --
  • Beginner level
  • Intermediate level
  • Advanced level
  • All levels
  • Beg/Int level
  • Int/Adv level

Community Generated

The level is determined by a majority opinion of students who have reviewed this class. The teacher's recommendation is shown until at least 5 student responses are collected.





About This Class

Power Automate Flow and Desktop are two technologies provided by Microsoft that will allow you to save countless hours of time and gain a competitive advantage in the work environment.

Knowing how to automate repetitive processes and leverage the power of AI to automate cognitive tasks are fundamental skills to have in 2021 and that you will learn in this crash course.

If you have been struggling with learning automation technologies this video is meant to take you from zero to hero and while keeping you engaged, curious, and entertained. I have created this course in such a way that you will learn by example with exciting projects.

You will learn how to use Microsoft Power Automate Flow and Desktop to easily automate repetitive and boring tasks. With Power Automate you create automated workflows that can be triggered based on an event, a schedule, or manually triggered by you. With Desktop you create flows on your computer to automate web applications or desktop applications. You will also learn how to use AI builder to build some amazing automation that could potentially save your company or you countless hours of work.

Not only does Power Automate save time, but it can generate reports that help you and your team be more productive and streamline business processes. This Power Automate tutorial is great for beginners!


Meet Your Teacher

Teacher Profile Image

Gregor Maric

Changing the world with automation


Class Ratings

Expectations Met?
  • Exceeded!
  • Yes
  • Somewhat
  • Not really
Reviews Archive

In October 2018, we updated our review system to improve the way we collect feedback. Below are the reviews written before that update.

Why Join Skillshare?

Take award-winning Skillshare Original Classes

Each class has short lessons, hands-on projects

Your membership supports Skillshare teachers

Learn From Anywhere

Take classes on the go with the Skillshare app. Stream or download to watch on the plane, the subway, or wherever you learn best.


1. Class Introduction: Hello and welcome to this course. My name is Gregory and I had been working in automation for the past decade. I have worked too many consulting companies with many different clients and on many different projects. With all this experience, I had decided to create a hands-on training for anybody to start getting into automation. Automation is becoming fundamental skill that anybody has to have in the workplace. In addition to this, one very important thing that you should know and that is proof of this is that microsoft has decided to release the RPA or the automation technology on every single licenses of Windows 11. Therefore, this technology will be available on any machine for everybody to use. Therefore, it is crucial that you get into this right now and that you start learning this with the experience that I have. I have decided to cleave a five lesson calls with some additional Warner's classes. In this five lessons, we're going to go from 0 to hero. We're going to learn everything there is to know about automation. And we're also going to learn how to use the Microsoft technology called Microsoft Power, Automate Flow and desktop to automate any business task that we want. And therefore, same time, the course is going to be structured in five lessons. In Lesson 1, we will understand the overview and the general idea of what automation is and on what kind of processes we can apply it. What is intelligent automation and what are the different components of intelligent automation? In lesson 2, we're going to start looking at the two different technologies that we are going to use. We're going to have a short walk you through two different technologies. And also we're going to understand the pricing and how to set it up on our computer. In lesson 3, we're going to take a look at Flow. Microsoft Flow is a technology from Microsoft. The tool is going to allow us to automate any application that is on the Web, any application that has an API connection. And to do this, we're not going to use any code. In lesson 4. We're going to look at the other components of Microsoft called Microsoft Power Automate Desktop. This is the complimentary technology to the first one. This technology, the desktop Power Automate Desktop, is going to allow us to automate any application or any web page that we have on our browser or that we can visit. Power Automate Desktop is going to allow us with flow to automate anything that we want. We're also going to learn AI Builder. So we're going to also learn how to integrate AI components into our automation. In less than five, you are going to have and knowledge on how to build an end-to-end process. We will leverage everything that we have learned so far to create a full on automated end-to-end process. Once you have covered this class, you will be able to automate anything that you want. This class is intended for anybody that is interested in learning a new skill, that is automation. Anybody interested in learning how to save time by automating repetitive and manual tasks. And also for anybody that wants to discover these new technologies from Microsoft called Power Automate Flow and desktop. And this scale is going to give you a great boost your job career and for your skills as well. This class has been created for anybody. You don't need to have a coding background, you don't need to have a Microsoft background. This is for anybody. Just like I said previously, Microsoft has made this technology available for free for any body, and it will be available on Windows 11 by default, this means that anybody who has access to this. Now, I am really excited about this. I'm really excited about sharing my knowledge about automation, showing you how it's going to allow you to save time and giving you this great skills. Let's get started right away. 2. Lesson 1.1 - Introduction: Welcome to Lesson 1, understanding intelligent automation. In this lesson, we are going to put Power, Automate Flow and desktop into context. We are going to understand the RPA tools that Power Automate Flow and desktop are part of. We're going to understand how it fits into intelligent automation, why we're doing this and understanding this so that we can understand what kind of processes will be able to automate and what also the future holds for these technologies and also for your learning the skills. These skills are still very new in the marketplace and new by learning it right now you're gaining a competitive advantage compared to anybody that is not learning it right now. After that, we're gonna do a quick walk-through of Power Automate Flow and Power Automate desktop so that you get familiar with the tools before we start diving into them and building flows that are going to help you so much and help you save so much time. Once we have seen the two tools, once we have done a walk-through from two different tools, after that, we're going to look at differences between flow and desktop. We're going to do this so that we understand when we can use one and when we can use the other one. And also when we can use them together to create really great flows that can ultimately a lot of stuff. Excellent. At the end of the chapter, we're going to do a short Knowledge check. Just make sure that you remember everything that we talked about and that you pick up on a key concept? Let's dive right into this. 3. Lesson 1.2 What is RPA: So what is robotic process automation? This is important to understand what actually is before we start building our key processes and automating tasks with an RPA tool such as Microsoft Power Automate Flow slash desktop. So it is important that RPA simulates any action that an employee or anybody that is trying to automate a task. Thus, it can change password, it can send an email, it can check and access Excel file. It can do all of those activities that a human does. It is also important to understand that arche doesn't change any of your underlying systems. So if you have an application or a web page or an Excel sheet, RPA just sits on top of it and spice different elements in. You don't have to change anything inside of the code of any of these components for RP to work. It is a very lightweight application, or it is a very lightweight software that is very easy to setup and very easy to get started with exactly for this reason. It is also important to note that RPA is a software and that set, it means that there is over 200 RPA vendors and solution in the marketplace. Microsoft Power, Automate Flow and desktop. That is just one of these 200 socials. That is solutions for all kinds of budgets and all kinds of requirements that you need to do with RP. Therefore, there is really, it is a very, it has been already a democratized technology and many people have access to it. Therefore, you shouldn't stop wasting time and you should start learning it immediately and grade that you have picked up this course and decided to understand how to automate with Microsoft Power, Automate and vessel. Now that we have understood what RPA is, let's put Microsoft Power Automate into context. Now there is something that is called a Gartner Magic. And Gartner is a technology company that does research on different software vendors. And they have placed all of the, well, the majority of the leading vendors inside of a Magic Quadrant 2. So which ones, in their opinion are the ones that are leaders, which ones are niche players, challenges, and visionaries? Microsoft, we can see that is right here with the leaders. And the leader is UiPath and Automation Anywhere, this magic quadrant changes pretty much every year and the competition changes and companies change as well. Microsoft wasn't, he wasn't present on this list. It was present, I believe, only in 2020 and 2021 since they acquired Microsoft Power Automate Desktop, which previously was a different technology. Now, with this, with this in mind, it is important for you to understand that there's other RPA solutions out there in the marketplace if you decide to learn them. Now that's all great. Understanding the context of all of these applications and understanding that all of these applications are there. That is great. But what can we do with our key technologies? Practically, what can we do? How can it help us automate a time and save time? So Microsoft Power Automate can help us with different things. It can help us automate business applications such as Dynamics, 365, common data services, applications such as Salesforce or warm day. Other applications like that, we can automate anything that we're doing in those applications. We can also automate anything that is regarding our modern workplace, such as Office applications like SharePoint, Microsoft Teams, OneDrive, PowerApps, SharePoint, and so on the list, just continuous, we can automate any activity that we're doing on those applications. We can also automate things regarding applications and infrastructure. So for example, we can automate updating Oracle or managing IoT devices, collecting data from them, storing it, and so on. We can also manage data and AI, like with Power BI and SQL and Excel, we can create reports automatically and on the fly, we can check that certain data pipelines are running in the air and there are operational without us having to login into the application. We can also do much on social media such as create, posts, do Reich's, and check different things, whatever comes to mind, we can pay them much automated with Power, Automate. And in this course we're going to do so many different little processes and complete processes that we're going to see example in each one of these categories, we're going to cover everything from business applications to social media. My favorites are social media because nobody ever really thinks that there is actually a robot behind certain account. But as I will show you in this training, it is so easy to create robots that act as the grow different accounts on social media that you will be quite shocked. And also you will look at social media maybe from a different aspect or light. Now, with all that said and done, we are ready to pass on to the next chapter. And that is intelligent automation. 4. Lesson 1.3 - What is intelligent automation: All right, We have understood what RPA is and how we can use it to automate different processes and different tasks and how we can use it to save loads and loads of time and make our life a lot easier. But how does it all fit? Intelligent automation, and why is this relevant for you? That is, learning Power Automate right now? Well, first of all, it's relevant for you because you are going to be understanding how we can use in this course, we're going to learn how we can use AI together with RPA as well. Because Microsoft Power Automate offers the ability to use AI to automate really end-to-end processes, but more about that in a second. So Let's understand intelligent automation. Intelligent automation is automating the process from the beginning until the end. A process is usually made out of different components. It's made out of parts where somebody is maybe creating reports, are doing manual and repetitive things like checking, creating report, extracting data. So things that don't often changed. After that part of the process is done, somebody usually makes the decision, extract some information maybe from a PDF or makes a, maybe a cognitive decision regarding a process before doing some more repetitive tasks. Now, with Power Automate and when RPA we can automate the active and the manual. As we can see in this slide, we automate the manual and repetitive tasks, but with AI, we automate the cognitive tasks. Now, what does that mean? That means that we can automate a process or more of the same process, make it better, enhance it, and also automate processes. And 12. Now, that is very important to understand because right now we are learning just to automate things that are archived later. However, we're going to see a few examples of what we're going to automate, also a related processes. Well actually we're going to see a lot of examples where we're going to automate a eyelid processes. We're going to do sentiment analysis, extracting data from invoices and the seeds. We're going to do some cognitive machine learning and much, much more. Let's take a look and let's just take a look at how RP starts and it, how it evolves. It is important to understand that RPA is just a beginning of evolution chain to get to artificial intelligence or two end-to-end automation, as I like to call it. So usually with the automating the manual and effective with Mark process automation, we get 60% of the activities done. Now. Then, once we add a little bit of cognitive RPA with some machine learning or some natural language processing, some chatbots, action. We automate. We start automating more and more until we reach artificial intelligence. And we have automated pretty much the entire, the entire process. This shows a breakdown of how different technologies combined with RPA can help you achieve end-to-end automation. There is also a lot of intelligent automation tools that do all kinds of different things that help us automate processes. And 21, now if we put all of these, whether we are able to automate a process and one. Now this is important because Microsoft Power Automate Flow and desktop tried to do everything that all of these tools, tools do together. Well, that might be a little bit of an overstatement because all of these tools have excellent functionalities in many of the things that they do. But the idea behind what Microsoft is doing, that they want to provide a service that is going to cover everything inside of the whole automation chain, meaning automating the process. And so that is why it is also important to learn Microsoft Power Automate Flow and rest of. It is also important to understand that power Automate is part of the Power Platform from Microsoft. And the Power Platform gives you the capabilities to create chatbots, to do analytics, to automate things, to create low-code apps with Power Apps and childhood through virtual agents and reports and analytics would be AI and automation with Power Automate. Now with all of that in mind, power Microsoft with also the Machine Learning services provided on Azure, is trying to really be a complete solution to help you automate processes. And 20, now this is not a promotional message. This is just to say that power Automate is integrated inside of a suite of other components that eventually, and also right now, can help you automate not just things with RPA, but also using intelligent automation. We're not going to see how we are going to use Power Apps, Power BI, and Power Virtual Agents. But we're going to use AI inside of our automation projects in this course. Now, with all of that in mind, what is going to happen with the work as we know it today. Work is we know it today focused around going processes inside of companies that are usually manual and repetitive in nature because they are required by the business to operate on a daily basis. Processes like payroll, like accounts receivable, accounts payable, onboarding an employee, of boarding an employee, and processes like this. All of these processes are very repetitive in nature. And if we manage to automate the first one, RPA, initially enhance them. And then with automation, then I have a question. What do you think is going to happen with inside of our workplace? We going crews many jobs, or is there going to be a creation of different jobs? There is different theories about this. I leave this for you to decide. 5. Lesson 1.4 - Flow Walkthrough: Welcome to the Power Automate, walked through. In this video, we are going to get familiar with Microsoft Power Automate Flow. And this is the Cloud version of the Microsoft Power Automate solution. Power Automate flow is going to allow us to automate API applications and ultimate brand new applications that are available and it has APIs. Now, let's get familiar with the interface so that we know where the different things are. And so that once we start creating flows, we already have a little bit of familiarity. Once again, we are directly to our homepage. On our homepage, I will see approvals pending from processes, any processes that I have, and I will see them on the homepage so I can approve or reject that process so that the next steps might happen. I also have some templates for me, start quickly and also some connectors. Now what our templates and what are connectors will get those in just a second. Now, right here on the left side of the menu, I have a menu that is a summary of everything that I need to know with Power Automate Flow. So that is my photos or all of the flows. And I have, I have the ability to create a new folder. I can choose flows from templates. I can create a flow from a different connector. I can manage different data options and data capabilities. I can monitor my Flows and see how they are acting and not just my flows, but also the flows from my desktop installation. I can see my desktop flows here as well. Then I have the AI Builder, and we're going to see this later on the course. This is going to be a lot of fun. We are going to use AI built, create some really fun processes. So after that we have process advisor. We're going to take a peek into process advisor as well. Process advisor is going to allow us to recall processes on our desktop and share them with colleagues, have them recorded as well. See how we do the same process and try to find an automation automatically. Like that. We are going to have solutions. And solutions are once we package different components from the Power Platform like apps, tables, automations, we can all package all of these inside of solutions and deploy them. So that's where solutions are and a very useful resource, which is to learn this, this part right here is going to allow you to give, get a lot of knowledge and love training from Microsoft directly about max of Power, Automate Flow, and less of any other technologies that they have. Now, let's take a closer look at the Create. Now, what I want to create a flow, I am presented with different flows to create. We will see these in more details, but these are the all of the flows that I can create in the cloud. Automated flows instruct closed scheduled flows. A desktop flow that I create my desktop and business process flow. A quick overview of each one of these automated flows are flows that run without you having to do anything that trigger happens. Somebody uploads a file, you get an email and this Flows trigger, meaning maybe you get an email. You get an email, the flow is triggered. It extracts information from the e-mail and inserted inside of an Excel example. In some cloud flows are flows that are triggered when a pattern or a click or an event that happens. So for example, I click the button on a page, or I click a button inside of my, my control panel right here. And the Florence and executes different things. Scheduled flows, as the word says, it's flows that are scheduled to run on certain events, dates, and times. So I schedule it to run every two hours or schedule to run every Sunday, first Sunday of the month. For example, desktop flows. Both are flows that we are going to see later in the desktop, we're going to build to automate legacy applications and desktop applications and business process flows. Business process flows are going to allow us to create integration with a user inside of our flow. So let's say that a user needs to submit some information to us, like fill out the form. For example, fill out a form to request form for TBI, a certain order. We can solve this for one side of the business, process flow and create a process out of it are all automated inside of Power Automate, which is really great and great functionality. This can help you save so much time. Now, that is from the create action. Now also I escaped the My flows. And here I can just have a look at all the different flows. I have. I have quite a lot of different flows, but these are all of the flows that I have on my Cloud. Now on the desktop, I can also see here from my control panel or from the Cloud version, I can also see all of the different flows that I have on my desktop as well. I can also see all of the business process flows that I have. And I can also see flaws that I have been shared with me or that I have shared with us. Now, I can also create a flow from a template. And this can be very interesting and also can help you save some time. Now, if you know already what you are after and what you're trying. By going to the templates, you can search for different templates and already create a process that has been put inside of a template and just find unit or sometimes I'll even find you, just use it out of the box. And there is hundreds, if not thousands, of different templates from which trues and which that you can use to start creating automations automatically. We're going to see these more detail later on in the course. Now connect us. On the other hand, connectors are connections with different applications. They are API connections, but let's not focus on complex words. Let's just call them connectors. And this allows us basically just connected Twitter for example. It allows us to connect to Twitter and do different actions inside of goodwill. So we can pause. We can check if, if a certain tag has been posted or we can see how many followers a certain user has different actions that the API or the connector allows us to do. So these are connectors. And if you find a connector for the application that you want, that's great because you can start using it immediately. If you don't, you have the ability to create your own Connect. Does obviously you have to know the API interface of the application that you are wanting to connect. But if u naught, you can build your own connector and put it in a market-based and sell it to others. Or you can use it just for yourself. Now, I'll take a closer, I'll show you also a closer look at the monitor. Why is this important? Because this is going to allow you to see details inside of the rounds of every single one of your processes. Now I can see that here I had some processes that have been running 11 minutes ago. One of these processes, some of these processes field, some of them were successful. I can also see any notifications that I don't have. And also I can focus on all of the failures that we're all processes that will fail. So and I can look into the details and understand the details wide has failed and where it has failed. Now we'll go into more details in later on once we start building processes and understanding exception handling. But for now, it is just good that you understand a walk-through and get an idea of how the platform is a sculpture where we can find different things and the capabilities that it offers. Now, why am I doing this? Because when I started learning this, I wish that I had learned this before I started using it because it saved me see a lot of time being able to know where to find the different components and what the platform has to offer. These are, on the other hand, all of my desktop flows and all of the different plans on my desktop. I can also see all different machines that I have. And this means that you can have many different robots. So you can install Power, Automate Desktop on many different computers, and have all those computers perform different actions and add those machines right here. For example, even have a virtual machine or you can have many virtual machines where you install Microsoft Power, Automate Desktop, and build yourself an army of robots that do different things for you. Ai Builder. Let's take just a quick look at AI Builder. Ai Builder allows us to build tried out of the box different automations with AI. So it helps us to build different automations right out of the box. Now, we can go straight into productivity, like they say idea and create one of these processes such as C processing. And we're going to do this later on in the course. But you can come here and just take them and start using them inside of your flows automatically. On the other hand, you can start from a model or from not trained algorithm, but you can train the algorithm and you can customize the algorithm for your needs. So for example, if you wanted to predict if a bottle of on a production line is half-full or half-empty, you could come here and you can train it on a prediction. On a prediction. You can also train a classification category for understanding taxed on, understand different lawyer types of contracts on information that you can have it extract and thes from packs and you can customize all of these models. It is pretty straightforward. There is no need to be a machine learning expert. All of this can be, all of these can be used without any knowledge in coding and in machine learning. Obviously, if you have knowledge encoding, it would be handy because you'll be able to do more things because both show and desktop allow you the ability to use cold, take some shortcuts. Now we also have process and wiser, and I will just give you a quick peek inside of this. This is going to show you all of the processes that you would want to automate in the future. So you would recall certain processes such as this. You will recall the processes, what the user is going. You will get the recording, a colleague of yours who do the same, you guys would compare it. You will upload it here. And 20 would compare the difference steps and create a map out of this, out of this, out of this process. As a result, you'll be able to understand which process or which steps are same in all of the cases that you submitted inside of the process analyzer and therefore understand which steps should be automated. Now, let's take a look at solutions very quickly. And Microsoft has improved or changed, is doing a lot of changes to the Power Automate flow and desktop constantly, they're constantly evolving and new features are being added. Therefore, it might look in my, look differently in your, in your version. Now, these are all different solutions I have, and this is where you would create solutions. You would also export your flows and important flows from one version of Power Automate flow to the next, or one environment to the next. Now, that is quite everything that I wanted to show inside of Power Automate Flow. It should give you quite a good idea of everything that the tool has to offer. It should give you some familiarity with everything. We're all different components are. So right now we are ready to take a look at desktop and then start building our flows. 6. Lesson 1.5 - Desktop Walkthrough: Welcome. In this video, we are going to take a look and walk, true, true max of Power Automate desktop. We're going to get familiar with the mean features and mean application capabilities so that when we start using it, we feel more comfortable and we can follow along much easier. Now this is the interface. Now, obviously I have loads and loads of processes that I have created. And you're probably if you have just install it, you will not have any processes. You will just have the ability to create a new process. And once you create your processes, you will see them. Here. It is very basic and simple interface. It is not very complicated, but what I want to bring your attention to is right here or to the settings. Here in the settings, we can update our version of Microsoft Power Automate. So you should always check that you are running the latest version and you can also update also from here. Another very useful piece of information is that when you're running your processes, you can stop them with the hotkey. And here this is where you would configure that hockey. Let's say that a process goals want and doesn't behave like it should be. And you cannot stop it anymore because it is blocking the mouse and locking your computer and so on. Well, you can stop it with this Hopkins. That is there is also other information but we'll get into it when the time, when the time comes. Now, let's take a look at what it looks like to create a float. So we'll just create a basic four and go through the interface, get familiar with every new flow that we create, creates a blank canvas with all the components, building blocks that we need. This is a low-code tool and low code platform. Therefore, we don't need to know any code or any or any advanced knowledge in coding or development. However, as I always say, it helps because it could make sales and steps and makes things easier. Alright, so this is our interface for building robots. It is pretty easy. It is a drag and drop interface, will look for different actions that we need to automate all do right here under the action. So let's say we want to open a web browser. We would go to the web application. We will look for what we need to do, like launch new Chrome. We will drag and drop it. We would fill out configuration parameters like the URL where we want to go, and so on. A lot of different parameters, not that many, pretty simple, and it will be added here, now, added here, but we'll get narrow because we don't have a URL. So this is what it will start looking at. We would add another action, another action, and our robot would start taking shape and form one action under another action. On the right side. On the other hand, you are going to have three things. You're going to have your variables. This is containers for your data. So right now when we created a new browser, we also produce a variable. So the instance of the browser, what does that mean? That means that if we want to attach to this browser, loosen actions, we're going to use this instance that is created right here and that is stored inside of the variable hold browser. And this variable, we can see it right here. Now that those are variables, we can also see input and output variables. So this is one process, but maybe we can have two or three processes. So we can pass information from one process and another process. And we do it here with input and output violence. But we'll see this when the time comes, when we build the processes right now I just have awareness of where different things are and what we have available. Now UI elements, UI elements are the selectors. So we need to select a different button on a webpage or we need to click on another web page. All we need to extract the link or inputs and texts. All of these are going to appear right here. These are UI elements and those are the things that we are going to interact and SPI and CPI. And finally, we also have images. Now we're going to use images to take screenshots, to maybe do some image recording certain parts where it gets hard to spy certain elements. So all the images are going to appear right here. So in a nutshell, this is our interface. This is what we use to run robots. This form right here, we would use to run our robot to recall different actions on the web, record, different actions on desktop. Right here on the bottom, we have summary and we also have the speed of which our robot runs. Now, this is by default set to 100. I always just that you should increase. This may be pulled for a 100 or 200. And if your robot is not doing what it should be because it's gone way, way, way too fast. All right, so that is in a nutshell, what it looks like. Microsoft Power Automate Desktop. 7. Lesson 1.6 - Difference between power automate cloud and desktop: Understanding the differences between a flow and desktop is crucial for building good flows and knowing how to leverage technology over the other. And one, we should use flow and when we should use desktop, that's why in this video, we are going to go over some basics differences that will help us understand why use one and why use the other. And will also give us a more complete understanding of Power, Automate Flow and decile. All right, Let's start with the basics. First of all, Power Automate Flow is a Cloud application. While Power Automate Desktop is obviously like the word says, desktop application, which means you download it installed on your computer and run it locally on your machine. While flow, you login into the Cloud and you set flows inside of the cloud. It can be from any computer, even from your phone. The pricing, the pricing model, we'll get into it in more detail. But basically, Power Automate Desktop is entirely fee, while flow has a small license fee attached to it. Building flows in desktop is drag and drop. While building flows in the Cloud is more adding actions and connect us. So while both of the technologies are low code, there is little differences, are significant differences in how you build flows in one technology and how you build flows in the other technology. The data connectors or the connections with AI and APIs are much easier and much more integrated inside of the cloud. The cloud version of flow is meant for API and integrating modern applications that have APIs. While desktop is used for legacy applications and it's hard to build APIs or integrations with machine learning. It is possible and we will see it in this course. But it is not out-of-the-box and it's not as easy as it is with the Cloud. These are some of the major takeaways for understanding the differences between the two technologies that we can go into many more detail. But these are just the main reasons why one technology is different and why the other technology is different. With this knowledge, we should be able to start exploring the two technologies and start building our flows. Let's go into the next chapter where we will see how we can install it to technologies and how we can start building flows right away. 8. Lesson 2.1 - Introduction: Lesson 2, setup in pricing. This is going to be a sharp lesson, but important. Nevertheless, we are going to understand the pricing model and structure behind Power Automate Flow and Power Automate Desktop. After that, we are going to get ready to start building flows. We're going to set up our ultimate flow in the Cloud. And we're also going to install Power Automate Desktop on our computer and get all of the components like plug-ins for browsers and things ready for us to work. After that, we're gonna do a short Knowledge Check to understand, to make sure that we understand everything in this lesson and that we are ready to start building our automations lesson. We're also going to have our first assignment. This assignment is going to be for you to set up your account to install Power Automate desktop on your computer, and to be ready to follow along together with me. I'm excited to get started. Let's start immediately. 9. Lesson 2.2 - Licencing: Power Automate Christ. Now before understanding Power Automate and how to build flows, we have to know if this technology is for us and if we can use it in our business, if the ROI is going to make sense for us using this technology. So let's understand the licensing behind Power Automate. Now, Power Automate Desktop is entirely free. Now Power Automate Desktop, we can download it from my flows, from the flow platform. We can go to My flows and we can install it arrived here, and it's entirely free. So that takes care of Power Automate desktop. However, Power Automate Flow has a little bit more of a complex pricing structure that is also evolving over time because they're adding more add-on suite and they're making it more competitive and more in line with other RP tools in the market. The basic account is $15 per month, and this allows you to build unlimited flows. Well, not really unlimited flows. You have a 600 flow limit on one account. And you can build up to 600 and flows for that one account. And then you will have to probably delete some of your flows or create a new account. There is another plan, $40 per month, and this gives you access to AI Builder and credits with AI Builder. Much of the power of pop, of flow comes from AI Builder and the simplicity with which you can integrate machine learning and complex algorithms inside of your processes to really automate really complex and really great things, which might also seem simple, but they are complex, Ultimate. So that licensing plan is $40 amount. There is also a Power Flow Plan which is 500 dollars. Therefore, you can build a specific plan or a solution or overflow packaged into solution and license that for $500 a month is also different components and add-ons. It is not just Power Automate flow that you are getting. You're also getting Process Mining and desktop and Cloud automation. Now we can see that there is different components for each one of these. There is also different add-ons that can be added. So with a 500-dollar add-on with AI Builder, we can get much more access to AI Builder and also support and different features as well. We can also add a 100, $150 per Baht per month for unattended RPA. And that is just one additional robot doing our processes. It has AI included and everything that we acquire as well. That is the licensing for Power Automate. I will include the links for the update licensing because the licensing for Mike's about ultimate is evolving and changing. I will leave the link down in the description before, but the key takeaways are that there is going to be a 90 day free trial. The free trial usually gets extended. Therefore, we'll probably keep on using it after the 90 days. That is also $50 for a basic user. That is not that much to invest for the power and for the time that you can help us save inside of our processes. On the other side, Power Automate Desktop is entirely free and we can use it for ever, for free. Right now. We can use it forever for free without any limitations on the usage. Thank you so much for watching. And I will see you in the next video. 10. Lesson 2.3 - Set Up Desktop : All right, Now that we have seen how to set up Power Automate Flow, now we're going to set up Power Automate Desktop, Power Automate Desktop we have installed on our computer. So the process is a little bit longer than just setting up an online account. Now, Power Automate Desktop, I do have a little preference for Power Automate Desktop, but that is just my feeling. All right, to setup Power Automate Desktop, we login Power Automate flow from the homepage. We click on My Flows and here we have two things to install. We can install Power, Automate Desktop, and on-premise data gateway. Now let's click on both of these and let's install both of these. Once I click on the icon, the installation is brought up from downstairs. And now I will do the same for the on-premise gateway. What is the on-premise gateway? The on-premise gateway will help us connect Power Automate Flow and Power Automate Desktop. There is other ways of connecting them, not with the Gateway, but we will see those later on in the videos. Right now we are going to setup the gateway and setup Microsoft Power Automate so that we can communicate with flow and so that we can have all of our environment ready and set up and ready to go. Once Microsoft Power Automate Desktop has finished installing, we will click onto it and launch the installation file. The installation file is pretty straight forward. We can just click Next and agree to the Microsoft Teams and click install the current version. I already have the existing current version, therefore, I will not be installing this, but once you click install this will install it on your computer and you will have access to Microsoft Power Automate desktop that will look something like that. Always make sure that you are always up to date inside of your Power Automate and see what version you're running. Now that is for the installation of Microsoft Power Automate. Obviously once you once you launch Microsoft Power Automate, it will ask you to login inside of your Power Automate, use the same credentials that you have used with Microsoft Power Automate Flow. So the same credentials you will use in Microsoft Power Automate Flow and desktop. Now after that, the next thing that we have to do is set up the data gateway. Now, I have double-clicked on the installation file. This is also a pretty straight forward process. Now in this case, I will update the the on-premise data gateway and I will show you a walk-through of what is required. Once data gateway is all ready and has been installed, will have the ability to launch inside of our data gateway. Now, I will use the same credentials that I have used with Power Automate flow. Once you have logged inside of Data Gateway and you will see different configuration for your data gateway. Everything should be configured right out of the box. And now with data gateway connected, now you can connect Power Automate Flow and Power Automate Desktop. So let's check if our Gateway has been correctly installed or not. So let's create a quick, quick flow from an insulin Council. This is just going to be a fast no need to worry too much about this process. We're going to see everything in detail. We're just checking to see if a cloud and flow, desktop and Cloud can communicate right now. So we're going to add an action that is going to run a flow from Power Automate Desktop. Let's select that and you will probably have the ability here to create a new connection. So I already have a connection, but in your case, you would select on-premise data gateway. You would put in the data gateway name that you're using. You'll put in the domain and username. You could find those from your control panel. So you can go into the control panel and you can type in. So this is your, this will be your domain. So we can select this. We will put this right here and then we would insert the password to our account. Now that will connect to our data gateway and I already have a connection. So I will connect to my Data Gateway and see if the connection is working. So the connection is working. Why? Because I can see all of the different desktop processes that I have. Now I can now select one of these desktop processes, and I can also select, choose to run it in an attendant or unintended way. So our Automate Desktop has successfully been installed. We can go to Power Automate Desktop. Once it has been installed, all of the plug-ins should be installed as well. Otherwise, if the pagans haven't been installed yet, we can install them simply by going to power. We can go to the Power Automate plugins and we can find them into different stores for Mozilla, for Google, for, for Edge as well. So those are all of the requirements that we need to get Power, Automate Desktop flow configured, ready to, ready to rock and roll and build some really cool automations. Let's kick things off right away. 11. Lesson 2.4 - Set Up Flow: Alright, let's get started with Power Automate Flow. Let's set it up and we're going to start from Google. And we're gonna see every step of the way. So all we have to do is type into Google, Microsoft Power Automate, and click on the first link. We're going to click on Start for free. We will be redirected to this Power Platform capabilities. So these are the different components of the Power Platform that we've seen in the previous chapters. We have Power Apps, Power BI, power Apps to build dynamic applications with no code. Power BI to visualize reports and do analytics. Power Automate that we will be using and Power Virtual Agents to build chatbots. Now, it is important to say that all of these technologies work best when combined together so that we can complete and create a really end-to-end applications. So with Power Apps, we could build an application that, for example, takes pictures of invoices or medical receipts. We can automate the extraction with Power, Automate, we can communicate the results with virtual agents and we can visualize all of the flow and all of the data in Power BI, that is amazing, That is great, and that is surely the power of automation and Power Platform. Now let's dive into Power Automate and see how we can set up our free account. Now, what is a requirement for Power Automate Flow is that we set up a Microsoft account. So we have a Microsoft account either from our company or we set up one from our self. If we decide to use a different account, such as for personal domain, like dot whatever, process lands dot or something like that. We can, but we will have a limited functionalities and capabilities such as we won't be, we won't have access to OneDrive or any of the Microsoft services out there. So with that in mind, let's login and let's dive into Power Automate flow. Once you have created your account and log, then this will be the page that you are presented with. You will not have any of the flows that I have here. This is what you will have. Now. This is everything that is required to set up Microsoft Power Automate flow. Right now we can start building our flows immediately. 12. Lesson 2.5 - Set up virtual machine : Have you ever wondered where your robot can live? Providing a home for your robot is very important, especially if it's supposed to do automated tasks and run by itself. So it is not always the best solution to have your robot run on top of your computer. Why? Because you might be doing certain things on top of your computer. Therefore, it might be an interference with you and your robot walking at the same time. So what is a good solution if you don't want to install the robot on your computer. So if you don't want to install Power Automate desktop on your computer, what you can do is set up a virtual machine now, completely up to you and entirely up to you. You can continue learning Power, Automate Desktop and flow without absolutely no virtual machines. However, if you would like to have maybe ten robots doing different things, you would like to set up 10 Virtual Machines. It will be much cheaper than buying 10 different computers. So in this short video, I will show you how to set up a virtual machine that you require for Microsoft Power Automate. So now setting up a virtual machine, I will be doing it from Microsoft Azure. Once we have logged into Azure, I will go to virtual machines. I can also simply search for it and I will create a new virtual machine. There is just some basic configuration that we have setup. We have to create a new resource group if we haven't done one already. And we also have to select the type of virtual machine that we want to set up. Now the only virtual machine or the only operating system supported by Power Automate Desktop is Windows can Pro. So let's select that. And that is going to be the Virtual Machine that we are going to set up. We are also going to set up a username and account for our machine. So we will call this. All right, after that, we can basically set up our, we can set up our instance. Let's make sure that we set a Spot Instance so that the price is going to be lower for our machine. All, we can leave all of the other configuration as we want. Let's just set up a name for our virtual machine. And let's call it RPA champion. And we'll call it one. Obviously we cannot contain under spaces and our virtual machine is ready to go. Now, all options are, have been configured. Now let's select the disk size. So let's select a standard as DD, standard encryption. And let's attach an existing disk. So let's go on to networking. We can leave all of the settings as they are. We are going to enable auto shutdown of our machines at 07:00 PM in case we do not do it ourself. And let's configure advanced configuration. We do not need anything in the advanced configuration. We can choose to add some tags. We can review and create our virtual machine. Now, this is how much our virtual machine is going to cost. It is going to cost around 0.390 US dollars per hour. Now, once this is configured, we are going to be able to launch the creation process that is going to create our virtual machine. So all of the validation part have been passed correctly. And we can create our Virtual Machine. Now, create a virtual machine. All I have to do is click on Create, and this will create ourselves the virtual machine. Now, what is cool about this? That if there is anything about this that I do not like or about this virtual machine I'm not satisfied with. I can cancel it and deleted without any cost our charges. So I will be only charged if I use my virtual machine and if I log into it and I have performing different actions, otherwise the virtual machine will not be, will not be, will not charge me or cost me anything. Right now the virtual machine is being deployed. Deployed. Once it's deployed, I will receive instructions on how to log in, and I will be able to login and use this virtual machine. Once the virtual machine has finished the deployment process, we can go to the resource and access our virtual machines. So let's see how we can do that. We will click on the Go to resource. We will have all of the details that we require now will be shutting down this machine after this experiment. So there is no need to copy any of the information that I'm sharing right here. On the left side, we have all of the different information regarding our virtual machine. But all we have to do is click on Connect, select RDP. That is going to be the way that I will be connecting select on RDP. And after that, I can download the RDP file. Once the RDP file is downloaded, I can just click on the RDP file and this will be presented for me. Now, once I click on Connect, I will be have the ability to insert the credentials that I have created previously. So RPA champion and the password that I have used once, once, all of that goes, once the obviously the password is accepted and and everything works, we can log inside of our virtual machines. So we are logging in inside of our virtual machine. So guess what we are going to do inside of this virtual machine, we are going to set up Microsoft Power Automate flow desktop, the gateway, so that we can have another robot that is ready for us to utilize and to work with. Once the window is, once the virtual machine has finished all of the configuration, we can see it right here. So the first thing that I will be doing, a couple of things on this machine to make it ready for me to use. I will be installing a couple of plug-ins that are going to make browsing faster and smoother. I will be installing Microsoft Google Chrome. So I will also be installing an ad block functionality for Google Chrome to prevent chronic pages load faster and just be more reliable over all. So obviously the first thing that we're going to download is Microsoft Power, Automate Flow and desktop. And we have seen how to do those previously. So there is, ladies and gentlemen, this is how we would create a new home for our robot instead, if we want to run it on our computer and on our machine. 13. Lesson 3.1 - Introduction: Welcome to Lesson 3. Finally, building cloud flows. In this lesson, we are going to get our hands dirty and we're going to start building some automations. We're going to have a lot of fun because we're going to create some amazing processes and amazing automations. First we're going to get started and we're going to discover Microsoft Power Automate Flow. And we're going to create a process from a template to get a feeling for it. After that, we're creating, we're going to recreate the process that we created from a template with a manual flow. After that, we'll just continue creating different processes that help us save time. We're going to look into Excel, sending emails, into connecting with different triggers, posting on Twitter. And after we have covered all kinds of different flows, including also approval flows that are going to give you the ability to really automate pretty much any situations that you have out there. After that, we are going to take a look at AI Builder. Ai Builder. We're going to look at some practical examples. We're going to take first an overview of what AI Builder is and its capabilities. And then we're going to build two or three examples of processes with AI Builder. We're going to build a sentiment analysis process, a card reader process. And then we're going to train a custom AI model on some custom invoices. And then we're going to include this model inside of one of our flow. So we will build a process that extracts from a folder all of the invoices. It does it reads an extract all of the information from the invoices and puts this information inside of a sheet document or an Excel document. Now with all of that, you should be able to completely master and completely used well, Power Automate Flow. Now there's many other concepts in Power Automate Flow that are more complex, such as manipulating data or data and other, and other concepts that we are not going to be covering in this flow because they're advanced concepts that you don't really need to get started with this knowledge, we'll be able to automate pretty much all of the requirements that you have. So let's dive right into it and see how this process are built. 14. Lesson 3.2 Getting Started With Templates: The easiest way to get started with Power Automate Flow is with a template. As we've seen before, templates are comfortable ways of getting started with processes that are already built, ended are already walking out of the box. So I will choose a template just to show you how we easily we can get started with templates. Templates are divided into different categories from which we can search for the exactly the kind of templates that we want. Or we can just search for the template that we want. I will choose a template that I wanted to use quite some time ago and that is saved Gmail attachments to my Google Drive. So let's see how this works. So I have selected the template and now I will connect my tool connectors. I will be using a Gmail connector and a Google Drive connector. So I will make sure that the accounts are connected. If not, I can add a new account right here. This will just allow me to sign into my Google account. The same thing goes for Google Drive. I will click on the three dots and I will select the account that I want to use for my Google Drive. Now, let's create this flow and see how it works. Once the flow has been created, I am presented with this panel. This is my control panel for my flow. On the top side of the panel, I have all of the actions that I can use with my flow so I can edit or change my flow. I can share it with my colleagues. I can copy or clone it or save it, save as I can delete it. I can send or share a copy of this flow. I can submit it as a template that we use for creating other flows. I can export it and import it into another instance of Power Automate. I can visualize the analytics. I can turn this a flow off if it's walking and I can get some tips and insight on this. Here, I will see the details of my flow. So here I will see what it was created, a short description, and the owner of the flow. I will, in the bottom part, I will also see all of the times that this flow was on. We've just created this flow. So right now there is nothing here, but here I will see all of the previous grants. And on the right side we have the connectors that we're using in this flow. So this is very important because, for example, in this flow we are using two connectors, the Gmail and Google Drive connector. We can also see everybody that is shared in this flow right here. So in this case, I am the owner and nobody else is using this float. So let's peek inside and see what this flow looks like. So we can see that this flow is a connector that says inbox. And then that is something that is looks like a loop and a Create File actual from this connector from Google Drive. Without diving too much into this, let's just go back to our flow and let's just try out as it is. Let's click on. This call is already running, so I will turn it off. And now I can turn it on. But let's see where we are in Google Drive. This is being saved. So let's go back into our flow and let's see where it's being saved. It saved in the folder attachments. So if I go right here, and I see the account, this account in the folder attachments. So let's go into this folder and see if this flow is actually Wharton. So right now it hasn't, it hasn't worked. But now I will send an e-mail with an attachment and we will see if it is going to work. So this foal is currently running. So here I have prepared an e-mail I'm going to send right now to the crypto boss or the RPA champion. So let's send this email and see if this e-mail is going to get intercepted. And if it is going to get saved into the folder that we've select. Sort here is the email that I sent to Microsoft Power Automate Flow. And here is the run of the process. We can click inside and see the details. And here in the attachments we can see that the file has been uploaded automatically. So right now I have created a process that is going to monitor my inbox constantly. And as soon as I receive a file has an attachment, it is going to extract that file from an e-mail and save it inside of the folder that I have specified here in Google Drive. Let's just take a closer look at one of the actions. So without looking into the details of everything that was created, we can see that all of these things, they don't require any code. We can, absolutely. We can see the code into different actions. So we can peek inside of the code and we can see how the code is done. And if we want to, we can modify. But to achieve all of this, no code was required whatsoever. 15. Lesson 3.3 - Instant Cloud Flows: Now that we know how to create automated cloud flows, now let's create an instance cloud flow. This will allow us to create flows that are fingered with an action. So previously we had flows that were triggered automatically when something happens. Now we're going to create flows that we are going to click on something on a button, on a webpage or maybe a button on our mobile device, or a No button on Power Automate, or a button on teams and so forth. So these are flows that we want to control by triggering them with an action would click usually. And let's build a process that is going to manually write an Excel sheet, a list of the newest tweets that were tweeted regarding a hashtag. Let's create meaningful name like always. And let's create this process. Now we see that the process is a little bit different than the initially the automated flows. So here we have as our initial trigger or what starts our process, we have a manually triggered for now, I can add different inputs to this. I can add text inputs. So maybe I wanted to submit as an input, a name, or an email, or a hashtag. So let's, in this case, let's select the input field and let's type in the input which is going to be RPA. So that is going to be our input. Now when our process starts, it's always going to ask us to input the input field before our process can walk. All right, so this is going to be the start of our process. Now the next thing that we have to do is we have to search for Twitter posts. So let's add a connector. Let's add the Twitter Connector. Let's click on it and let's see all the different actions that we have. Below it. We need to search for tweets and let's assign into the Twitter application. So to sign in, in this case, I'm not connected to this connector. I will connect to try it now. Once we have signed in and it's very easy to sign in, now, I will be I can click on the three dots right here on the side. And I can see that I am connected to a Twitter account. So when I click right here, I will automatically get a list of possible inputs that I can use. So right here, I am using an input field. I have named it also input. So let's go right here and let's search for something that is called input. So here we go, Right here is input. That means that we're going to take this text and we're going to put it right here, and we're going to search for these tweets. So far, so good, It's always good practice to our processes after we complete a couple of steps. So let's test our process and see what we get as results. Let's start this manually and let's click on facts test. So we are connected to Twitter and we can start with the process all at, before testing the process, it is asking us to insert some inputs from. We can insert RPA or we can insert whatever we would like. So let's insert our automation. And let's run this flow. So our process should search for tweets that are automation and it should return us the tweets. Let's see what we get as a result. So the flow has run successfully. And we can see, and we can see the results right here. So this is the input that was sent via text. It was sent right here. And right here we have the search text which was automation. And here are a couple of tweets that it returned. Now you can notice that there is a format, a strange format to the results, because different data associated with the results were not just receiving the text, but we're also seeing an ID. We're also receiving loss of information that are all grouped in something that is called JSON adjacent string. So with this Jason's thing, we would like to write these results inside of an Excel sheet. But without modifying any code, we would like to do it just by dragging and dropping or without knowing much gold. So let's do that. Let's go back to our process. Let's edit it. And now that we have extracted all of the tweets and we're going to extract only 20, 20 tweets. We can further customize it and extract as many as one. Now what we wanna do is we want to write an Excel sheet or a Google Drive sheet. So what do we need to do is let's search for sheets. And we have the connector sheets, the character sheets has all of these different actions inside. It has get rows and throw the lead pro, ghetto gets sheets and update rule. So we will have to insert a new row for each tweet that we collect right here. Before doing that, let's create a Google Sheet document where we are going to be saving our tweets. Now that we have created our sheet tweet, we can put in a column, we can name this column. We can also add a user who tweeted this, and we can call this miscellaneous, miscellaneous 2 and miscellaneous. So we can put some other information in this columns. All right, now let's go back to flow and let's complete this process. So we need to insert a row. And when I click on Insert Row, I will have to login to my Google Sheets. And now when that is selected, I can select Automatically. I'm connected now to Google Sheets. I am connected to Google Drive and I can see all of the different folders where I can select file. So I've selected the tweet file. I will pick the first worksheet and I will insert the rows. Now automatically, I am going to get a column names or out of my Excel sheet, and this will automatically be put solid. I can insert automatically whatever information from the tweets automatically into this south. So let's see how that would work. So here I have the tweet and I will click here. And dynamically, I will see in the right side, let me disappear from the screen for a second. So right here in the right side will see automatically the tax that is generally so tweet text. Let's add a tweet tax. And when we ended we tax, since we're extracting 20 tweets, this is automatically pushed for us inside of a loop or an applied to each. This means that for each tweet that is here, we are going to insert a new row. Now that is amazing, that is great because Power Automate has just, for us generated a whole lot of code and whole lot of activities that if we were a developer and using normal developing language, it'll take us much more time. So let's go back to inserting the row. So here we haven't tweet tax the user. And now again, automatically we can see the different fields that have been enabled. So I can add the name of the user location, username. So let's add the username and let's send some miscellaneous fields, such as There's really a lot of information. So let's see if there is a lot of a retweet on this. And let's also see if we can get the number of likes on this, on this, on this tweet. So let's search for something that says, well, let's get the France Count. Elijah's get one more piece of information that we'll just use it as a description and see what we got. All right, now, theoretically we should extract, when we click on this process, we should search for whatever, whatever we put here inside of the input. Search, whatever we put inside search for 20 tweets. Let's make it more. Let's make it 32 is such for 2030 tweets and then copy and write those tweets inside of this Excel sheet or this Google Sheet document. Let's test our flow and see if it's going to work. So let's start this manually, save and test. And we are going to be searching again for automation. Now before every flow runs, we have a we have a little dialogue that tells us that these two are the connectors that we are going to be using. Why? Because usually this is a lot of action. So it's always good to make sure that all of the connectors are working properly. So let's write down automation. And let's run this flow and see all the tweets that we find. Alright, we're already getting some tweets. The Google Sheet document is getting populated. We are writing information already in here, and these are our tweets. So in a couple of minutes, we were able to create a robot that queries twitter. It searches for a specific tweet that we want, and it writes an Excel sheet or a Google sheet with all of that information. Now let's go back to our process. Let's click done. And now for my flows, let's say we wanted to trigger this process, get Twitter hashtags. Now if I click Run, I will be, I will not run the process automatically. I will have to input an input that I want to write. So let's say I don't want to search for automation anymore. I want to search for a dot AI. So let's run this flow again and see what happens. So we are running the flow and we can see that new tweets are added to our sheet. So we have collected the first 30 sheets that were from RPA and now the following ones are from AI. Let's make this process a little bit better, and let's add in one more column. And in this 11 additional column, we can change and know which word we're searching for. 16. Lesson 3.4 - Instant Cloud Flows part 2: So let's go back to our process. And very quickly let's edit our process. And let's go to apply for each insert rows. So right here in the miscellaneous, I will not put friends count, I will put something else. So if we go down here and we check out the different fields that we have, I want to put the input that I'm using. So we will use the input that is inserted right here and we will know exactly what kind of text is being created. So just, let's clean this up a little bit less clean out all of these tweets. And Power Automate automatically adds an additional column to your Google Sheets or Excel sheets, where it stores an ID so that you can reference whatever it has inserted and it knows where to where it left off so that it can continue from that. In this case, we are going to be refreshing everything. Let's save this and let's start this process one more time and see if this time it is going to insert the search them automatically in the column so that we have a little bit of a cleaner looking document. Let's search for actually RPA. That is perfectly fine. Let's run the process and let's look at our Google Sheet. Alright, so we are searching for RPA. This has been retweeted 41 times. There's actually a couple of Twitter bots as we have created one. Also other people have created Twitter bots as well, which kind of defines also the purpose of social media. Now let's do one final search and see if we can search for automation. Let's run this process from here, and let's insert automation and see how the Sheet document is going to look like right now, anytime that I wanted to make a search for a couple of tweets in a specific area, I could use my automation and prepare a report on the fly without having to do it manually myself. All right, so now we have created a process that instance automatically tweets inside of our Google Sheets document. Now let's send an email. Let's learn how we can send an email. Let's go back to our Flows and let's create a new flow. So we're going to create a trigger flow as well, and we're going to call this send emails. Let's create the flow and we're going to just leave it as it is. So when we click a button here, this flow is going to be triggered. What is flow is going to do is going to read a list of different emails and is going to send an e-mail to each one of these people. So let's go to our flow and what we need to do, we need to get all of the content of the sheet or Excel document that we have. So we have to search for Google Sheets. So let's search for Google Sheets. And we've selected the Google Sheet connected. Now let's select an action that is available for us. Let's use the action get girls. So let's use the action get rows. Now, let's pick a, let's pick a file that we want to attach. All that we want to get the data from. So we're going to use this email list. We're also going to use Sheet1. And we're gonna save this. Now, let's test our process and see if we're getting the results that we should be getting. So let's continue and let's run this flow. Or flow has successfully run. We can download the results and we can see that it has, it has got all of the results with its own, it wouldn't own format, but we can use this results now inside of our flows. And I'll show you right now how we can do that. So let's send an e-mail to each one of those, to each one of these people inside of this list. To do that, let's search for the connector and you're getting a handle this, I'm sure by now. So let's search for Google Sheets. Well, not Google Sheets. Let's search for a Gmail connector. So let's search for Gmail and also remember to use always Version 2 when you see a different connectors. So reply to an email or send an email, we have to send an e-mail. So let's try to send an e-mail. And here let's click into the field. And also let's click here in the dynamic field. Let me just disappear from the screen for a second. Let's click right here in the name field. And we're going to see this option right here. Email it says, so Power Automate was smart enough to identify that this is the field that we can insert right here, and this is also the title right here, emails. So if we go back here, we can see emails. It found that title and now it's going to send an e-mail to each one of those rows inside of that document. And that's why it also has created and applied to each. So this is a loop that it has created automatically for us. So let's add a subject here. This is, and let's generate this. And let's test this out and see if our process works. So let's save this. Let's test our process manually. And we can see that the process has successfully sent all four emails. So it has sent e-mail. One, we can cycle through different emails to the Send Email tool. And it has also use email three. We can also see the row inputs that were used right here. So it has sent for emails. I could have also customize this a little bit better, just make it a little bit more understanding. So right here in the, in the subject, I will add a name. Name is also the other column that I have right here. So lets out automatically just on the fly, center, save this and send for more tests, e-mails, and see what happens with that. So like this, we have learned how we can create and send. I'm e-mail. So let's say we have a list of orders of customers, orders that we need to send to different customers. Each customer has a different delivery date, different the new date, different ID order. And we want to summarize all of those inside of an e-mail non ether item individually. We can write them, we can we can complete them with Power, Automate and send them automatically. So let's check out also sound the emails that I have been receiving in the past few minutes. And here we have we have the latest emails that I have received just now that we have sent. So these are the four emails that we've just sent with different names. They were all sent and they were all received successfully. So that is how we can create an automated flow that is triggered. So if we go now back to my photos, I can trigger this flow automatically from here and just run it. Well, that is it. We have learned how to create two kinds of incidents, cloud flows. One that writes inside of an Excel sheet and one that extracts from an Excel sheet, one that automatically collects tweets from Twitter, and one that automatically sends emails. So far you have learned how to create automated cloud flows and instant cloud flows. Now, let's create a scheduled flow. 17. Lesson 3.5 - Automated Flows: Now let's create an automated flow. Well, actually we're going to be creating two flows to automated flows. The first outmoded Cloud flow is we're going to be breaking down what we have just created from a template. So we created a template, a flow that when we receive a attachment in our inbox and e-mail with an attachment, we save that attachment to Google Drive. So we created that from a template. A lot, lots of things were not, lots of things. A few things were created for us. Now let's try to replicate that from scratch by building it ourselves. It's always good to have meaningful names in our processes. We don't lose them. So let's search for the trigger that we are looking for. We can we can find all kinds of different triggers. I am looking for the Gmail triggers. So when I receive an e-mail to my G-mail account, something will happen. So we can see that automatically this sugar was created, and now I have the ability to create new actions or new connectors, connect new things. Let's take a closer look at this. We can see that I am using one of my connections and I can also change the folder from which I did. I'm checking right now we're emails RSE. So let's also look at the different details. We can specify the information that from the e-mail that arrives. So we can get really specific about the type of email that we are looking for. So we can configure the different fields from which person we are expecting to receive the e-mail, what could be in the subject. We can also use regular expressions in all of these so that our query would include other things. We will just select the inbox and leave all of the other configuration like that. Let's add a new step. So the new step is going to be to save a document in Google Drive. So let's search for the Google Drive connector. And right here, I am looking for different connectors. Slash here we have the connectors and here we have what? These connectors blue. So I'm looking for a Google Drive. So maybe let's search for drive. And here we have the connector. Now when I click on the connector, I have all of the different actions that are created. So let's look for one that says Create file. Fortunately is the first one on the, in the list. Let's select it. And so far, so good, it has been pretty easy. And now it's giving us a couple of fields. So instead of using coal to configure any of this, we're going to just use click. We're just going to click around and configure this connector. So create a file. It's asking for a folder for us to tools. So we're going to go to our Google Drive. We're going to look at the different folders that I have in Google Drive. We're going to see the attachments folder. I'm going to be using this attachment and then the filename that is going to be created. So the file name, I'm going to make it into a the title from the e-mail, maybe the subject. So let's, let's make it dynamic is going to be the subject and the file content is going to be the attachments content. And we can see that a couple of things created automatically, a loop was created automatically for us. So this means that like we have seen automatically, it is going to go through all of the attachments inside of an email. And it's going to save all of those attachments with the filename of the subject of this of the email. So we have a couple of minutes. We have already replicated the template that we have created and we have already created an automation that does a couple of things. Before running this, let's make one little adjustment to the subject and let's put the attachment name. So let's search for attachment, or let's search here in the dynamic content. And let's put the attachment name. Okay, our process should be ready and now we can test our process and see if it works. Now, let's create an e-mail, send it to ourselves. But first, let's save our process. And let's click on Test. Now that now that we are enabling this, we can send ourselves and email and see if our process is working. So let's create an e-mail. Let's write it to ourselves. And let's also add a couple of pictures. So to see if they will be saved as well. Let's add these pictures and let's send them to ourselves. All right. Apparently the process has run successfully. So if we track in our Google Drive in a couple of minutes, we're going to see that the files that we've sent have been uploaded successfully. Now that we have replicated the template that we have created with an automated flow. Let's create another automated flow from scratch to get a feeling of how Kinda flows are created. So what I will be doing this fall, I will be automatically posting to Twitter every time I upload a YouTube video, as I have told you already in the introduction, I have a YouTube channel, so I often upload to this YouTube channel and I would like to automatically make posts on Twitter when I upload a new YouTube to a new video onto YouTube. So let's call this Upload. And let's search for a trigger. So in this case, the trigger will be coming from YouTube. Here it is exactly what I'm looking for when I upload a video. And we can see that we have also other sugars as well. One of video is uploaded by another channel when a new video matches a search. So we could create a flow that checks when one of our interests, that channel is uploading a video and sends us a reminder or an e-mail or makes a post about it, or we can do pretty much anything that we can think about. So I will select this and I will create the flow. And just as previously, a trigger is created right here. So I am, are also connected already to my YouTube channel. And therefore they either need to set up this connector anymore, but it's just, uh, adding a new connection is just signing into you. So this is our trigger for our automated flow. Now let's create a new action. Now, what do we wanna do? We wanna post to Twitter. So I will search for Twitter plug-in, and here there's Twitter action. So post to Twitter again is the first action that we want. But we can see that we can get many different actions like get user, get the number of followers, will get my followers, get followers from another account. This could be actually pretty interesting, and so on. All right, so now that we have this, let's post a tweet. So this is why we will put the text that we want to be Twitter. So for example, we can write in. And after that, I want to bring your attention to this part right here. So this part right here is going to allow you to automatically insert dynamic content Power Automate Flow is great because it automatically looks for all of the types of APIs and connections that this connector provides. And it allows us to use them just by clicking without, basically without much configuration, just by using this dynamic content. So for example, if I wanted the title of this video, I will just put in the video, the video summary title. And in here, maybe I wanted to put the description of the video while I actually wanted to put the link, and I'll put the link to the video right here. So automatically I am getting from this connector right here. I'm getting the link to the video, and I am also automatically getting the title without having to write any code. Now, that is that is it. I can save my flow and I can test my, and I confessed my process. So once I click Tasks, this process is going to be enabled. And when we apple now something to YouTube, we will have a video that is automatically posted to Twitter. Now let me show you an example of how this works. I already have a process that does this. So when we go to this Twitter page and we go down here, we have a robot that uploaded a video for hours and go automatically when I uploaded something to want an apple it something to you. So that is how we create automated flows. 18. Lesson 3.6 - Approval workflows: Let's build a approval flow. And approval flow can be created inside an automated flow and Stein and insulin Cloud for all scheduled flow as well. Inside of any one of these, we can put inside. We can put an approval workflow inside. What approval is going to make us do is going to allow us to interrupt the process. Wait for approval. Once the approval has been collected and achieved, then it will continue with the process. So let's start with an incident Cloud for, for, as an example. Let's call this approval workflow. And we will trigger this process manually. So what we want in this process is we want to send an email. So when we trigger this process, we want an approval and B to be generated. And then if the approval is okay, we are going to send an e-mail. So let's create this process very quickly. So let's make it a little bit more interesting and let's replicate the weather report that we had in the previous example. So let's do this very quickly. So let's get the forecast for tomorrow. And let's get it in Montreal. And after that, we will add a new action. We're going to say if we're going to wait for an approval. So let's search for a pool. And here we have the pools. So let's select this approval and we have created an approval, start and wait for an approval, wait for our pool. So in this example, we are starting and waiting for an approval both at the same time. So we're going to start the approval process. We're going to approve and reject. We're going to call this approve whether report. We will assign it to myself. And we can add further description and items such as, for example, information that needs to be improved and description, let's say, Do you want to send this e-mail? So let's save this. And right here we are creating an approval. Now, in the next step, let's send an email. So let's go back to Outlook. Let's not select office, let's select Outlook connector. So we could use the office as well. But let's, let's go with the output selector. And let's send an email. So we have to sign in again. Let's sign in and connect our connector. Now, I'll connector is connected. Let's send an email to ourselves. And let's add some information, weather report. And let's add here a. Let's, let me show you what I'm looking at. And let's add here some dynamic content, such as the conditions. All right, let's save this process. And this process now should not send the email, it should interrupt, it should wait for an approval and we get an approval in our home or in our action items and approval. So let's see, let's save this and let's test this process and see if we get an approval for water. So let's test this and see what happens. We're using all of these three different connectors. So we're using approvals, weather and Outlook. So let's run this flow. And once we run this flow, we can check, then it's has stopped right here. So if I go to my inbox right now where I send, I can see that I have received a email and email that says approved this report or no. Now, if I approve it, the process should continue. Now let's go back to our process and let's see it in execution. So this was our approval process. This is still running. So it's still running and let's approve it. So let's approve it from right here. We are sent right here inside of our approvals in the action items, and we have the ability to approve our process. So let's write a description and send email and let's confirm this. So once this is confirmed, the process will continue with the different steps. Now this process has been approved. Now we should, in a little bit, we should receive an email. Now let's go back to our processes and see what is happening with it. So this is a summary of everything that happened inside of the approval. So this is the approver who who approved this process. There is a response that was collected which is to approve. It stores the date. I can also use all of this information further on and say also here in the e-mail, who is the one that approved the process out? Different people. And this also gives me the ability to build elaborate and complex processes that require also human input inside of the process. So that there is a human that is controlling different chunks of different checkpoints within the automation that is happening. 19. Lesson 3.7 - AI Builder: Ai Builder. Now, this is one of my favorite parts of Power Automate, and it definitely helps save so much time because it helps integrate AI inside of our automations. Remember when we're talking about intelligent automation and we send that spark of the automation is done in a repeated way or in a manual way. And the other part is automated in a cognitive way. And if we automate both of the two things, we can truly automated processes in its entirety. So AI capabilities have been included inside of Power Automate flow, and that is found inside of the AI Builder. If we go to the AI Builder and we go inside of the build capabilities, we have ability to use right out of the box. Already, a couple of great AI tools that can help us save a lot, a lot of time. So let's go through a couple of these and understand how we can utilize them. So we have the Business Card Reader. This is going to read the context of a business card and it will interpret the results and give them to us via a text inside of a tax so that we can scan if we wanted all of the business cards that we have, we have a category classification. So this is going to allow us to classify, for example, feedback from users and group it inside of positive feedback and negative feedback. So we could also do it on, we could also do it on customer reviews. So we can group a bunch of customer reviews and we can run this model on top of them and understand what are the different customer reviews saying about our product and group the good ones into good and the bad ones in combat, we have entity extractions. So this if we have pieces of text that we have or e-mail or whatever content, we can run it directly through this and we can extract automatically people, places, locations, and so on. And that is great if we're trying to find certain details and information inside of documents. Identity document reader, let's say we need a authentication to our website. By the, we need to read a passport or an ID from a customer. We can scan a hundreds of documents on the fly and extract the information from these documents using the identity document reader, the invoice processing. And we're going to see this in this process. This is going to allow us to extract information from invoices. So let's say we receive different invoices from our suppliers every month or every week or whenever. And we need to copy this information inside of our systems. We can use the invoice processing. To do that. We can extract that information, automatically, save it inside of an Excel sheets, send emails, notifications up the data, and so on. Power Automate flows truly great for automation. Key phrase extraction. So this will extract key, key moments or key entities inside of phrases so that we understand more or less the context of what is happening. Let Language Detection. And this is useful when we try to understand what kind of, when we try to group user feedback, for example, of coming from different countries and we want to tag the text so we can use language detection also if we want to make sure before using, for example, a language translation option, we can use language detection to make sure which language we are translating. And I see processing. This is an amazing feature. This allows us to extract information from scanned receipts and put them, use them inside of our flows. We're going to see this example as well. We have sentiment analysis and pretty much everybody knows what Sentiment Analysis is, but we're going to use this as well. We also have tax recognition, which is a great feature, and this can help us extract text from different images that we have. Now, all of this can be integrated with a click of a button inside of our close just by using it inside of a floor or using it inside of an app on the box up on the upper side of the screen, we have the ability to customize and build custom models. So if we wanted to build a custom category classification and the form processing or prediction custom model. We can, we don't need the one straight out of the box that has been trained already by Microsoft. But we can customize one ourselves. So we're going to see how we can do this for form processing, for example, we're going to train it on different invoices and extract different information from different invoices. So that's how we can use these different category classification, entity extraction, form, processing, and prediction. All of this is utilized and deployed automatically or Microsoft Power Automate environment. Therefore, you don't need any hosting options, any code. You don't need to be a machine learning expert. You just need to know how the methodology of using this tool works and what the best practices and approach to using it is. So in a nutshell, that what Microsoft Power Automate AI Builder is, and that's what's going to allow us to look. Now, let's jump right into it and let's build a AI powered flow process. 20. Lesson 3.8 - Sentiment Analysis: All right, Now that we know how we can use different AI models, keep in mind that these preconfigured an already existing models. You can use them straight out of the box if you have access to AI Builder while building your own custom AI model will require you access to database and you will have to create custom table. We will see all of that. Now let's create a simple AI process that is going to do some sentiment analysis on some tweets. So an update, the excel sheet that we have created previously. So let's create a new process from scratch. This is going to be an instant flow and we're going to call it sentiment analysis. It will be a manually trigger a flow. And let's add the new stuff. New step first, we have to connect to Twitter and we have to get all of the tweets that we want to do sentiment analysis on. So let's get some tweets. So we will search for tweets and we will search for AKI. So let's save it. After that, we will run sentiment analysis. So we will search for the AI Builder and we will look for the sentiment analysis option, analyze positive or negative sentiment tax. So we'll select this. And we will also select language. And here in the text we will insert the tax. So the tax that we want to insert is the the textContent of the tweet. So let me show you what I am looking at right here. So what we want to insert right here is the dynamic content is the tweet text. So let's insert that right there. So this created automatically as always, a loop that is going to search for the, let's see how many, 20, less good, just for five tweets. It's going to do it for five tweets, and it's going to give us the sentiment for these five weeks. So now we have the sentiment analysis being run on each tweet. So the next thing that we need to do is we need to insert this information inside of Google Sheets. So if you see that, I can add an action inside of this loop right here, or I can add an action outside of the loop. I will add an action inside because it's inside that I want to insert inside of Google Sheets. Let's search for sheets. And let's insert a new row right here. Insert a new row. We will pick the file that we want to insert. So we will insert this inside of attachments, inside of the tweet file. We will select the worksheet that we want to use. And we will also insert the information that tweet. And let's also insert this, the sentiment, sentiment analysis. So let's insert this and this should insert all of our. Now when we inserted the sentiment sentences and new loop has been inserted, created automatically here for us. So inside the first we're inserting the tweet, the user. We can insert also the username. So let's search for the username. And the username is. So let's use the user ID that is tweeted by. And let's add a little bit more sentiment assessment so we can say overall text sentiment. Let's see what that is. And let's also add the score. Alright, let's try out our process and see if it works. So to recap, we have just built a process that is going to be manually triggered, is going to search for some tweets about RPA. Let's change it to love so that we have a more, maybe a more positive keywords we are going to test do and sentiment analysis on each one of these tweets. And after that, we will insert the inside of this Google sheet. We will insert the tweet, the user who posted it and some sentiment analysis results that we're going to extract from our flow. All right, let's save this. Let's test it and see the results. So let's start this manually. Let's run the test, run the flow. The process has done successfully, so it is taking its time right here. Let's take a look at here. So this is going actually pretty fast. And these are the results that we got so far. Now that is how easy it is to use Power Automate and AI Builder together to build complex flows that do very useful. Things that otherwise would code or with other services would take a lot more time. 21. Lesson 3.9 - Build Ai Flows from Template: All right, Now, since building AI processes, It's so much fun. Let's create a Business Card Reader. But this time instead of creating a AI process from the create and creating it from scratch, I'm going to create it directly from this panel right here. So what this is going to allow me to do is I can click on it. I can also use it in map and using the flow, if I click on User Flow, it will take me to the templates page and it will show me all of the processes in templates that I can use it right out of the box. I can choose one of these. I will select this one. This is the process that I want to use. So onClick, we're going to check a business card. We're going to copy the information from the business card and inserted automatically inside of a table, inside the end digitalized that information. So in this example, it is creating a database connection automatically for us and is building the entire process automatically for us. So this is the process that has been generated. Everything has been already filled out, and I am basically ready to go. I could take my time and review the different, the different details of this. So this is basically it is taking, it is creating a manual trigger that is an image. So add input file. So they selected, they created a file and then they use the predict option of analytics of AI builder on the business card model is extracted all of this information created dynamic data, and it allows it to insert it right here inside creating a new record inside of a dataverse table. So in this table I have all of these columns and I can insert inside of all of those columns. If I click here, I have access to all of the parameters that are available, that are extracted here from the prediction. But let's save this and let's test this and see if this is working. So this is right out of the box. This manually, and let's test this. So I have found already a business card, which is this. But let's try using it on this business card. So let's go right here and let's import this business card that I have saved right here. So let's run this flow. The flow has run successfully, so we should get some results. So the prediction has run successfully. It is using the business card model and these are the name. So it is extracting it has found name, it has found the city is found the e-mail, the address, and it can automatically insert all of this information, but not only inserted, we can also send an e-mail right after this with all of this information to ourselves or do whatever kind of action, add an approval. Do we want to approve to insert it inside or no. So let's also check the bottom that is created a new record inside of a table. So a new a new record was created with all of this information. So that is how easy and simple it is to use AI Builder, either by creating new flows are using AI builder. Now let's look at a, another example of AI Builder, but this time we're going to create our own model from scratch. 22. Lesson 3.10 - How to build a custom AI model: All right, Now let's build an invoice processing or invoice extraction model. This is going to be a lot of fun and this is, we are going to be building and training our own model in this video. And let me remind you we are doing automation course and we're going to learn how to use AI to automate more of the things that we want to do. So we can start with an invoice processing for, and I can use it directly here in a flow. I can see and try it out how it works. So what is going to do it is going to automatically try to find different values inside of, inside of the forms. But this is not what my form looks like. Let's say that I have different forms and I would like to train this model so that it recognizes forms from my suppliers may be my suppliers have a specific type of form that looks completely different from anybody else's form. So I want to start from scratch. I want to create a new model, but I'm not just going to create new model, I'm going to build on top of Microsoft model. So what I need is 50 invoices from my supplier or five invoices that I want to analyze and extract information from. So let's get started and let's name this AI model. So we're going to be building a new AI model. And I'm going to call this model invoices, invoiced AI model. So let's create this model. This will take a couple of minutes to create a model. So what we are doing right now, we are asking Microsoft Power Automate to create a custom model that we can train with information. Microsoft Power Automate is going to take care of everything from feigning book hosting to deploying and accessing, giving us access to our model. So let's stop. So the first thing that we need to do is we need to add a couple of fields. So we need to add what kind of fields we want to extract from our from our invoice. Let me show you the invoice that we are going to be using. This is one of the invoices that we're gonna be using. So we want to extract probably the date. We want to extract the invoice number, we want to extract the address. And some of these amounts may be total, and that's and the customer ID. Alright, so let's insert this information right here. So we need a field. So let's add the fields. So we want invoice number. Let's add a field. All right, let's add another field. So what do we want? We want a address. So how about location? Let's add a customer ID. We can also extract a table. So let's try to extract a table as well from an invoice so that we can have a summary. So let's select this table. Let's take a look at the invoice one more time and see what it looks like. So we have a description and a price. So let's extract those two fields, a description and a price. So this is going to be a description. And right here we are going to have the price. All right, Now we have everything. We can move on to training our model or giving it some examples. So let's create a new collection. And let's add some invoices for collection so that we can sort of Power Automate, can start learning how to train. So let's add some documents and the minimum amount of documents that we need is five documents. So let me go to my folder and I will select training documents. I have this five training documents, but I will upload inside of Power Automate. So let's go back to our flow. And let's upload these documents. So let's find where we are saving them right here. Alright, we have uploaded these documents. These documents are going to be used by Microsoft Power Automate to train and to learn where the data is located. So that one we'll see with different invoice from this is Supplier 1. A different invoice that looks completely different than these? Well, not completely different, but it has different structure, more lines, different locations of different things. It is still going to be able to recognize, recognize the text. Let's add another supplier. So let's add one more supplier. So in this case we are adding Supplier 2. Let's add Supplier 2 with this five invoices. So again, we're adding five documents, which is the minimum amount of documents for Power Automate to learn where the different elements are on the webpage. So let's close this. Let's rename this. Supply up to and let's proceed. Let's go on to the next step. Now we have to train our documents. So let's click on Supplier 1, and now let's flag the different elements on the page where they are. So the Power Automate can identify them automatically. So let's select the invoice number. Alright, so invoice number is right here. And we're going to select this where we're going to call it like this. And we're going to select invoice number next, the location. So we're going to get all of this. How about this is the location? So this is going to be the location and then the date is going to be this is going to be the date, the total, we have it right here. So this is going to be the total and this table that we have right here. So let's select these two. This is going to be our summary table. That is perfect. So right here, this is going to be in the description, and this is going to be the price. This is going to be in the description, and this is going to be the price x. And let's click on Save. And like this, we have saved a table as well. All right, now let's repeat the same process for the next document. So we have to do the same thing for the next document. So we have to this and tell it that it's date. Tell it that this is a location. This is a invoice number and this right here is the total. This is the table, the summary table. So let's add a description. And let's add one more description. So it is pretty fast to do this. And I just have good on five documents. After I've done it on five documents, I can already utilize it inside of my flows and it can be considered that it has been trained already. So this is the date hours late. And now I have completed with all of the fields. Now that I have done this, I can go back. And I can repeat the same thing with Supplier 2. So let's February months late voices. So let's select the last piece of information, which is the invoice number, and we are done. We can go on to the next step. The next step is going to be who train our model, and then we will be able to use it. So this will take a couple of minutes. So let's take a couple of minutes and wait for our model to train. All right, The model has finished training now we can use it inside of our flows or inside of our apps. All right, this is the model that has been created for us and has been deployed. Now the next thing that we should do is we should publish our model so we can publish it so that once we publish it, we can use it inside of Power Automate and we can use it inside of Power Apps. So we will be able to create a new flow and utilize our model instantaneously inside of our flows. Let's run a quick test and see if the train, if we can use a test image and extract the information. So let's select this image, let's copy and paste it. Now, the image is running on top of our model and it's extracting the information. So it has found this information as the location, this as the date. This is invoice number, this as the summary. And this as a total. We also see a confidence score of almost 99 percent, which is pretty short. And we can also check another another. We can also make another test from the other supplier that we have. So we have this other supplier right here. So we can test then as well. So let's test their invoice. This inverse is a handwritten scan. Invoice sold should be more complicated to extract information. And let's see if this is going to function and extract all the information. So it found this 91 percent sure, it found this 99 percent sure. And the total in the end and the table as well. So that is how easy it is to create a custom Flow that we can utilize immediately inside of our process. 23. Lesson 3.11 - How to use your AI model in Flows: All right, Now let's build the flow that is going to allow us to extract information from different invoices with the model that we have just created. So I'm going to create a manually triggered Flow and call it invoice extract. And you should be quite familiar with this interface and with the whole process that we're doing right now, We're make this flow so that when we trigger this flow is activated. And the next step that we do is we need to get the content of a folder that we have. So in this folder right here, I have different files that I want to analyze. So let me create actually a new folder right here. And let's call this test. And let's move this untitled spreadsheet right there. So let's move this spreadsheet called. Let's see if the slide here is, is that alright, so we have one folder and we have some files in here. Inside of the invoice folder, we have a date and invoice number and amount and the location. And this files right here are the invoices that we had previously. So let's say we either received by email and with the flow, we save these documents inside of, inside of drive. And right now we want to create a flow that is going to extract the information automatically from there and put it inside of a Google Sheet document. Let's extract all the files. So let's go to Google Drive and let's look for and let's look for the action that we need. So in this case we were going to need a get all files. So we're going to list all files inside of a folder. Let's select the folder that we would like to use. So it's going to be inside of Google try instead of attachments. And it's going to be this folder right here. Save. And then the next thing that we need to do is we need to get the content of each file. So that is probably going to create a loop. So let's search for DR Again, Google Drive, and now let's get the file content using the ID. So here we are getting a list of different files. And right here let's get the content using the ID. So here we are getting the content and now we need to use our model to extract the different data. So let's use AI Builder and let's look for a for the model that we have just created. So we are going to add a custom AI capability, which is going to be two. And extract from custom forms, extract information from forms. So we're going to look for the model that we have trained in, voice, AI model, that is the model that we are trained now we have access to it. We are going to select this PDF documents or forms times from PDF. And we're also going to pull the file contents. So the form file content that we're getting right here. So everything right here in the set, we should be able to extract all of the information. All right, now we want to insert everything inside of Google Sheets. So let's search for sheets. And let's insert a numeral. Let's pick the file that we want to select. So it's inside of Google Drive, it's inside of attachments, it's inside the train and then it's inside of test. So this is the file we want to use. We're going to select Sheet 1, and then we are going to select a date or an invoice number. So let's see what we have. Total value. Let's put the total value right there. So let's look for the total value is right, that, all right, now let's try our process like this and see if it's working. So let's save this. Let's check. Our checkout sheet is currently is empty. Let's test our process. Let's manually test it. So what it should do, it should go to the Google Drive folder. It should extract the all of the files that are inside of that folder, all the PDF files, it should run them through the model that we have just created. And it should extract the value and write it inside of this file right here under the amount column. So it is working. The first invoice that it has checked is 81. Let's see if it's able to find the same for all of the other invoices. At the same time, I can check inside of the folder that I have inside of a different invoices. So this is currently a 150. Let's see the next one. This is 9270. So let's see this invoices to a 100, 17090. And let's see if we have a hundred and fifty and two hundred and seventy seven. So let's see. This is probably this invoice right here, 170. So this is it has managed to extract all of the information and it successfully wrote it down here. Now I can find you on this and invoice number, add more information, do more steps, but you get the idea, this is how you can build complex flows that require cognitive capabilities, such as using your eyes to extract information from this PDF. Let's say you have to extract this, you'll have to copy and paste in this example, I can't even copy and paste it, so I would have to manually update it and just close and open different screens. And that'll be very time-consuming. Alright, thank you so much for watching. I hope that you have enjoyed creating processes with flow. 24. Lesson 4.1 - Introduction: Welcome to lesson 4, building that's the flows. This lesson is going to be amazing. We are going to learn how to use Power Automate vessel. This is an amazing technology that can help us automate pretty much anything. Well, actually anything Power Automate Flow is great, but it helps us connect only APIs. On the other hand, Microsoft Power Automate can help us connect anything, even APIs. It just is very complicated. Connect with APIs and things, but it's possible, but it can help us automate pretty much anything. So I have structured this lesson so that in the beginning of the lesson, start by learning the main actions and the main building blocks of Power Automate. So we learn concepts like variables, loops, if conditions, flows, and concepts like that. After that, we will take a look at Excel automation. We'll take a look at a couple of examples of acceleration. We'll take a look at a couple of examples of e-mail actions and email automations, sending e-mail, retrieving email, and so on. We will look at exception handling. Exception handling is very important because exception handling is going to allow us to build strong flows. We're going to take a look at web automation, web automation and spying web pages and getting elements from web pages. After that, we'll take a look at desktop applications and how we can interact and spy applications like the calculator will do small in a car or any other application we have installed on our computer. After that, we will take again another look at some basic tips and suggestions on how we can build strong flows the don't fail. After that, we will concentrate and build our skills on a I also on Power Automate Desktop, we will use different services that they have to offer like OCR and different cognitive services. And at the end of that, we will do a knowledge check and an assigned. This is going to be an amazing class. Get ready to learn a lot. 25. Lesson 4.2 - Getting Started: To help us save some time, I have already prepared process with a couple of actions that will help us understand variables in a nutshell. So there's basically two kinds of main variables. That is, occasions where you have just one variable meaning for example, just the name of a person like arche champion. Or you have variables or data tables or tables where you have many variables or many, like a list. For example, you are exporting a list of connections and you are getting a list of different names or a table of different names. So we have to learn how we can manage a single variables and multiple variables. So let's take a look at that. So right here, I have added in action that is going to get anything that I have copied inside of my clipboard or Control C. And it's going to produce a variable out of it, is going to create this variable right here, and it is going to store the action right there. After that, I have added a display message. And display messages are great when you have debugging. So if we want to know what is inside of a variable, we can add display messages and display messages will pop up and show us what is inside of a viable. So to add display messages, just, we just search and we find display messages. All right, so let's run this first part of the process and see how this work. Let's also take a look at our variables right here. So I'll just run this. So the empty, we just received this viable in the clipboard. So this is a link I have copied and it's right here. It got inserted inside of the clipboard text, and now it's inside of this variable, and I can use it inside of my flow. I can open a web browser now with this variable and I can do different things. So that's the first part. Now, let's see how we can work with data with more data. So for example, I want to get all of the files inside of this folder. I have right here my downloads folder. I want to get all of the files and I want to know how many files I have inside. So let's see what happens then. So I will continue this run. And I'm getting another message that is telling me that I have 332 files and the name of the first file is processed lands partnership program. So after that we will continue, will continue. And now let's, let's go over a couple of these variables. So a couple of things happened right here. So when we are getting the one, we are getting the get files and folder action. We are saving all of the files that we get inside of this file variable. This is not just a simple variable. We can see it here. And if we click inside, we can see that is a list. It says right here, list of files. So I can have data tables and I can have lists. So right here we can, we can see that we can access different, different information from this list. I can access the detail of the full name or the name or the extension of this file. So I can exactly know. I can get, I can break down this into its components. After that, I can see that there's also many different files. That is a number right next to each one of them. And I can create a loop on this, so I can iterate on that. And we will see examples of this later on. For right now it is important that you understand there's different kinds of variables. And this is one kind of variable that is automatically produced by months of Power Automate later on, we're going to see data tables as well, but it's very, very simple. Let's take a closer look at the display message. So display message, I am calling, I want to bring your attention to this. I am calling the files. So with this x right here, we have access to all the variables that we have created. Now with this arrow right here, I can further see what kind of methods or what kind of options I have available to do. So if I put files dot count as a variable, I can select it automatically, right here, files and count. I don't have to write a line of code. It will automatically tell me the number of files that are inside of my folder. In addition to that, if I use files and squared parenthesis 0 and then the name of the file. Let's see what that is going to give us. If we go here and we see here is 0. And then we click on More and we go to name. We're going to get exactly this name right here. If I selected the extension out and gotten the extension, if I selected a different number, I would have gotten a different number. So I have full access to all of the data. Inside of the list, that is very important and it's going to help us do so many automations and manage data relatively, quite easily. After that, just one final note and observations on variables. We can also do different things with variables. We can set variables and this is going to be important for us when we're working with different processes to set the variable to a number, to a link, to some texts that we're going to require later on. So we can set different variables. We can also do multiplications and mathematical additions on variables. And we can also set variables too, DO conditions like true or false. So is this going to be true or false? So in a nutshell, this is what variables, this is how variables behave. This is a overview of variables. It is important that we get a grasp of what variables are. Now, let's take a look at loops. Loops are another very important component that is going to help us build many different processes and many different flows. So there are basically three kinds of loops that you are going to be using in every, in every situation of pretty much in any process that you want to repeat a certain action. So loops are accessed from here, and these are the different kinds of loops that we have. So let's go over very quickly what they are. The first kind of loop is just a simple loop. It says start at 0 or at one or two, and then end at five and increment by one. That means do an action five times this case, 1, 2, 3, 4, 5, and then it's going to stop. That is the most basic kind of loop. Let's say we want to send five emails to the same person. It is going to send as many times as we want inside of the loop. So that is the most basic kind of glue. The next kind of loop is a loop with the condition. This is a loop that checks if something is compliance with a condition, and if the condition is true, it exits the loop until the condition becomes true, the loop continues to work. So in this case, we have as, as long as this number is smaller or equal to 70, you are going to continue adding numbers until you reach the number 70, until new variable 2 is going to be smaller or equal to number 70. So that is the second kind of flu. And finally, the third kind of loop. Here, as we've seen in the previous example, I'm extracting all of the files. So I have 244 files. If we if we can remember from us example, now, I don't know how many files I can have. I can have many different kinds of numbers of files. Every follower can have a different number. So this kind of loop is going to be dynamic. It is going to check those in Excel or is God is going to track the number of files and it's going to execute for each number of rows or number of files that we have. So in this example, for each number of files we have inside of the pictures folder, this is going to execute this loop right here. And that's how basically loops and structure. We're going to see them in more detail later on when we start building our processes. However, before we start building the processes, good, an idea that we know what are the different kinds of loops so that we know how we can also fine tune and best create our processes. The next part is if conditions. Now this is a, also an important part of flows because it's going to control the way that our flow walks behaves. So if conditions, again, they are just actions and they're accessed from the panel. I can look for conditions right here, and I can see all different kinds of if conditions that we have right here. So just to go over a couple of basic conditions. So if, if we insert an if right here, so for the if, we have a first operator, which is an initial condition, meaning one, for example, or true. And then the operator is, can be different operators. It can be greater, toe can be less than new Numeric operators, or it can be empty. Maybe like if there is an empty string or if it contains a specific packs or specific, a specific key or something like that, we can check those things. And if those, if the condition is true, then the, then the process will continue and also pay attention that we can use also variables in each one of these. So we can have dynamic Fs. So that is how EFS walk. We can add another thing to f's and F's are useful for tracking different parts of our process to make it do different conditions. So for example, if a user, if a user has more than 34 years, then do a certain number of actions. If the user has less than four years than those specific other actions, then we have else if, as if is basically what if the user has 25 years? In that case? Else, if user has 25 years, Buddhist, other kinds of conditions. And we can see that our flows are divided into blocks. So for each one of these conditions, I will put a, the actions that I want the process to execute. Finally, the last conditions else, meaning if this doesn't execute and this doesn't execute, in that case, the else condition is going to execute. So that is loops. In a nutshell, there is nothing that is called a switch and switches just like an elaborate loop, you can see each one of these cases as an else. If so, if else-if Sunday, LC4 case Monday. So if it's Sunday, then do this. If it's Monday, do this. If it's Thursday or Wednesday or something else, do something else. But then the default cases, if it's any other day, then do something else. So that is loops. And one final component that we should take a look at is flow control. We will see all of these later on, but flow control basically gives us, we've seen it a little bit throughout the process, but it basically gives us the ability to control outflows. And here are some of the main flow controls that we have. Labeled, Run flow and display message, and also go to saw labels. We use labels to set different points inside of outflow. Then we would use the go-to, to go back to those labels so that we can jump around in our flow however we want. After that, we have some flows so that not everything has to be on the same page inside of our flow, but we can create different subflows and samples are created right here. So we can see the different subflows and different subflows and we can also create a new cell flow as well. So that is how subflows our work and are created. And then finally in flow control, something that is useful is also to get the display message to print for us all of our variables. And that's in a nutshell. All of the actions and all of the knowledge that you're going to need to start working with Power Automate. In the next chapters, we are going to dive into the different infinite detail of everything that we've seen just now. And we're going to start building processes on the web and on the desktop. 26. Lesson 4.3 Desktop Automation: I usually recommend to start desktop applications by running your application. So you would search for R1, then you will such for the system and run application, you can start the service just like this or any other kind of window activity, task, service, et cetera. So would run application, what are we going to? Which out of which application are we going to automate here in the application part, I'm going to write and put where and how I run my application. So if I wanted to run, for example, the calculator right here. And I wanted to launch it with my, with my power ultimate tool. I will put in the path to the execution folder of the calculator. So this is a Windows system that two calculator execution executable. I can also add any command line arguments while adding this a working folder. And I can also determine the window. Do I want it normal? Do I want to maximize, et cetera? In this case, I will leave it as normal. And after launching the application, we can have different actions. So we can wait for the application to low and wait for the application to complete. So this will allow different applications flowed all the different elements that they need to load and complete doing any, any initial activities that they do once the application is started. This comes with more advanced applications. So if we wanted to open, for example, photoshop or DaVinci Resolve, which are applications that flowed a lot of components. We will have to wait for them to load. Right now with the calculator, we can continue immediately. So let's save this. Now. After that, let's try to close our calculator, and let's run this and see if it opens our calculator. Alright, great. It is opening our calculator. Now, I can continue adding actions here like spy UI elements. Like for example, right here we can go and we can click on UI element inside of a window, or we can send different other actions as well. But I want to take a shortcut and I will use the desktop recorder. Now, with the desktop recorder, we have two capabilities. If the desktop recorder, it is not able to recognize the different images, like right here is able to recognize the button. It is able to recognize all of the different buttons so we can interact with this application. If it's not able to recognize the different buttons, we can use the image recorder right here and SPI, the different images and Power Automate will create the images for us. This is the image recording is only useful when the element does not change on the webpage. And one is going to always be in the same exact place. Alright, Let's do a base calculation and see if we can automate it. So we're going to click on number five, and then we're going to click on this button right here. We're going to click on the three. And then we're going to click on the equal sign. After that, we're going to try to get this text right here. So we're going to right click and we are going to try to get the text. So we're going to get the text. In this example, we are going to, let's see if we can get that. So let's maximize this. And let's see if we can select anything that is maybe better. So right here, we can get 15. So let's get this text right there. All right, Let's finish our process. We see all the different actions that were generated like press button in window get details of a UI element. And all of these actions are right here. If I search for press button, I will find the paths button and the UI automation. So let's leave the comments that were created for us. So let's run this process and see what happens. All right, it is opening the calculator. It is clicking. It is repeating all of the clicks that I have performed. Now if we minimize this, we can also see that we have managed to extract all of the information that we are required as well. Now, there is different ways that we can achieve different things. We could have done the same thing with image recording. We could have maybe done the same thing by clicking on the calculator and then sending keys like tab, tab, tab, and then enter that, that tab and then enter all we could have sent keys like for example, like multiply or add. And we could have created the automation like that. That is always different ways to create an automation and everything can be automated. There's just different ways of doing it. So that is, in a nutshell, how we do, That's the automation. Just keep in mind. Another easy suggestion for you is that if you know how to start a specific application, you can pin it down right here in the menu, and then you can use the desktop recorder and you can load should automatically from right here. So if you are having a hard time identifying where the application is, you can select it right here and you can launch it from right here. So I click right here, and let's finish this. And let's test this out as well. So we can delete all of these steps. And that would, for example, open the folder. Or if I had a different application that I wanted to use and utilize, I could have pin that down to my, my desktop and I can launch it directly from that. So let's wait for this process to attach itself to Windows. So it has open, it has opened on my other screen. So this is the File Explorer. All right, ladies and gentleman, thank you so much for watching, and I will see you in the next video. 27. Lesson 4.4 Web Automation: Welcome to web automation. Web automation is super important because it's going to allow us to automate any activity that we do on pretty much any webpage and any web technology that is out there. So that is why web automation is super-important when learning how to Automate and Power Automate Desktop provides us two ways to building our processes would web automation. It allows us to drag and drop activities from our actions panel right into our flow. So for example, launch new Chrome, click on a web element, extract some information from webpage and so on. But at the same time, if this is might be a little bit time-consuming, we can also use the web recorder right here. The web recorder is going to automatically help us record all the steps that we do on a page. So for example, let's say that we need to insert different data inside of a different account, data regarding different profiles. So maybe that addresses, the addresses are saved in an Excel sheet. And now we have to insert all of this information inside, inside of a webpage. Let's see how we could do that. We can do that with web. Recall that I would click on the web recorder and in this example, I will be using the web page yay challenge. So RPA Challenge has provided a form that we can use simply just simulate some texts. However, keep in mind that everyone pages built differently. So one way would walk one page, one way worked with another page. It is the understanding of different actions that we have and how we can use them to figure out how we can work on different pages. As you build your knowledge with Power, Automate desktop and web automation, you will understand what use and how to behave with technologies and different situations. Right now we are starting from the basics and seeing the simplest kind of form later on we'll see more complex form, especially used on social media that use all kinds of different chunks and how we can simulate actually human inputting into the different forms. All right, let's see how we can automate this webpage. So right here, I can select Chrome. This is the browser that I'll be using. I have my plugin installed and running. I also have a ad block capability that is going to block all ads and make the browser more responsive, faster and easier also to do what we're about to do, that we're going to select on advanced and make sure that we have the right tab selected here. So we are going to be integrating with the RPA challenge. So let's click Next. Once we click Next, this one, recall that appears. Now, this one we call that once we click on the record, it is going to capture everything that we need to do. And it's going to create a skeleton robot for us. So let's see, we want to submit some information. So let's hover our mouse. Everyone where we hover our mouse, red line with a description appears. Now this red line is telling us what power Automate is seeing and it can interact with. The text above tells us what kind of element it is. So this is the HTML element. This is an input text. So in this case, since I want to put some text inside of this, I'm going to be looking for the input facts. Not always is the case that we can find the input text. If we can find the input text, we're going to try to find the closest thing to it, or tried to automate the process may be in a different, more creative way, maybe sending keys or tabs or something along those lines. But in this example, it is walking. So let's click on right, on the right mouse-click. And here we have different capabilities that this input text enables us to do. So we can add the text with, I mean, so this is what we're going to do. We're going to add some tags. So this is a phone number. So let's add a phone number. Grade, this is a last name. So let's repeat thing. Let's add a last name. And he has. Great. Now let's input this information. So let's click on Submit, submit grade. Now, another page with more information appears. Now let's imagine that we have all this information for a list of Excel contexts. It will take us forever. So let's just do the first part or let's just insert some random information as well here. So let's add some text to me and let's add an address. I'm not very creative. So n, Let's submit this as well. So again, the same form is formatted in a different way, so we don't need to do anything else. We can also, we can also write now, we have created all of the actions that we need. We can click on Finish and see what happens. Once we click on Finish, we can see that a lot of actions were created. Most of them are populated texts on the field. That's what we were doing and that's what the tool has created for us. Now let's run this process and see if it can work right out of the box. So we created the two recorded. We didn't add a line of code, and now we can see that it is working pretty well. It went to the web page that we selected, the RPA challenge, and it's inserting the different information it is submitting. And it has inserted everything. Great, That is fantastic. We have just automated something that could be a very manual, repetitive task, just like we can insert information, we can also extract information from a webpage. Now, let me show you how we can do that. What we are going to be doing now, we are going to be extracting the cryptocurrencies from this webpage. So just now we saw how we can insert data inside of web page. Now we're going to see how we can pull data outside of webpage so that we can work it extracted, send it to somebody, or we create reports. So if we go back to our Power Automate, I have created a new sub flow and I have moved all of the information inside of this new soft glow. Now, in my main, I'm going to start with Google Chrome. I'm going to insert link that I want to navigate to, and I'm going to maximize the page when I stop this. Before launching this, I'm going to go to one we call this since I'm on this page county select Chrome like before, and make sure I'm on the right page. Once that is done, we are going to have the web the corner open up and we can start the recording process. Now we're going to select a couple of things from this table. So we can see that right now we are, we want to try to select the td. Td is the table cell. So in this example, let's get this value right here, which is Bitcoin. And let's get also this value that Let's see if we can get it. So we can get it either like this or we can select the parent element. In this case, the parent element is not that we can go up above. And here we have the TDI. So let's see if we can select the value. Alright. Now, when we go to the next, the next row and I select a theorem, we can see that automatically it has selected the first two rows that we have identified. So I identify the first row and second row, and the tool has identified all of the rows inside of the table. Now, it is going to extract automatically the name and the price. Now let's extract as well the change. Maybe I want to filter by change. So I would go to the cell, I would write, and I will go to extract element, and I will select what I need to extract. So let's extract the text as well. So again, automatically, I can see that all of the elements are created and extracted and extracted here. Now there's one more thing that we can do, and that is if we select right here and we go to set element as pager. Now this is going to tell Power Automate that this table has many different pages inside of it. And we can go through it and extract all of the different pages until it goes to the end. So with that, now we can extract all of the information. This doesn't work on all different tables because every table in a web page can be made differently. So it doesn't always work and it requires some fine tuning. So let's click on Finish and let's see what we get. So we just have a few actions. So let's delete all of the actions that we don't need. Like going to navigate a webpage and the comment that was generated. And we just have this extract data from webpage. Now, if I double-click on this, I can see that I will be extracting all the available rows. I'll be using the browser to instance, I will be storing it inside of a variable. I can also store it automatically inside of an Excel spreadsheet. Let's start with a variable. I will also output this variable right here, so maybe I can use it inside of my flow. I can also select how many, how many rows and how many pages I want to insert. So for right now, I will select all available. Let's see what happens. So we are opening the page of cryptocurrencies and we are right here, and we can see that it is cycling all of the different pages, all of different pages automatically for us, and extracting the information from all the different pages. All right, so let's see what the results were. So if we click right here, we can see that we extracted quite a lot of cryptocurrencies. We extracted the first value and the third value. There was an issue with the second value, but we will try to spine high to extract and fine tune our process. This is, however, the way that we would extract different information from a webpage. We can extract single information as well and tables like this. Now in this example, we extracted the information inside of a variable, but very easily we can change this and extract information inside of an Excel sheet so that we can use it automatically. And that is, we use what? Automation and different web elements to integrate into different pages in the following exercises and the end of the discourse, you're going to have a great chance of practicing on different webpages to fine tune your skills and to learn how to automate different pages. 28. Lesson 4.5 - Excel Automation: Let's dive into Excel automation. Excel automation is probably going to be one of the main blocks that you do in pretty much any process because you either have to read data or write data probably in the Excel sheet or a database or somewhere like that. But not all processes like that. However, it is a main important part. In this chapter, we are going to learn how to read data from Excel. We're going to read how, we're going to learn how to ride it a toxin. And we're going to look at some advanced features. So we're also going to try to do something useful. So we're going to try to get all of the files that we have in certain folder and write them down in Excel. So to do that, the first thing that we have to do is we have to look for different actions. So we have to look for an action that gets us all the files. We've seen those previously, but we've never used them right now. We're going to use a get files. So let's just search for this and let's get files folder and see how simple it is and powerful to use. So let's get all of the files that I have in my downloads folder. So let's get all of these files right here and we can apply different. So I can search maybe four different documents or PDFs or specific files in this example, I'm just getting all of the different files. I can also show them if I wanted to, in the output. I don't want to sort anything. I'm just going to leave it as standard like this. So the next thing that we wanna do is we want to try this files inside of an Excel document. So let's write, let's create an Excel document. So let's open the Excel Options, and here we see different actions. Let's drag and drop the launch itself. And right here we can either write a blank document or we can open a new document that is that we already created. Select it from apart. We're just going to create a new document and save, and save it and started like this. So right now we are starting a new document is being opened. The next thing that we do is we have to copy all the files and put it inside of the Excel sheet. So we have all of the files right here. Now let's write this files to the Excel sheet. So write books out worksheet. So like this, we can write all of the documents. So do I know what I want to write value 2, right? Yes. I know what I wanted, right? I will not write list of files. This right here. So we write this. Where are we going to write this on a specified cell or an currently active. So we are going to put it on the currently active cell because that is going to be the initial cell inside far Excel sheet. So let's save this and that is pretty much everything. Let's run our process and see if it works until this point. So it is reading and it has wrote down all of the files that we have inside of a folder. It has written down how many files? That is, a hundred and three hundred and twenty two files in just a second. Now let's take this a few steps further, and let's see how we can improve this. So let's add a clause. So let's save this document. We're going to save this document, and we're going to also save it with a current name. So let's save the document as, let's choose a path where we're going to save this document. So we're going to save it right here, and we're going to call it list. And let's say, all right, now let's make this a little bit better as well. So after closing the Excel sheet and saving it with this name, we want to, we want to change this so we want to get the name. So to get the name only, we're going to do something like this. This is going to get us the first name. However, we have we have quite a few names. We have let's let's get the count and see how many how many documents we have here. So let's add a message or let's set a variable instead. So let's set a variable right here. And this variable, we are going to call it number. We are going to store inside of this variable, we're going to score the count and we're going to call it counts of files. So we know how many files we have. So right here we're going to put the number of files and Let's run our process and see how many files we have. So we had 322 files. So now that we have seen how to write to an Excel sheet, Let's see how we can read from the Excel sheet. So let's read from the same excel sheet that we have right here. So let's open again an Excel sheet. So let's look for Excel. And let's open a new Excel. So let's open the same Excel that we have seen just now. And let's read from that Excel sheet. So we're going to read the list of files. And right here we have our Excel sheet. So what do we need? Do we need to read from our Excel sheet. So pretty much similar concept as writing for an Excel sheet. We can do it dynamically or we can do it statically, meaning that we can tell it which Excel, which range of cells from one until sell 48. Or we can select the cells dynamically. So let's select the cells dynamically because the other way is pretty straight forward. The values. So the values of a selection, the values from a range of cells. So for example, starting at column a and then add a row 45 ending at column B and row 45. So this will be a static way. We can select all the values from the worksheet, but let's select the values in a dynamic way so that we don't know when our Excel sheet is going to add. We know that everything is in column a, but we don't know when it is going to add. We don't know how many rows we have inside of our folder. So let's figure that, Let's figure that out. We can figure it out in different ways. We can get a count of the folders here, and we can just count for 233 rows and we can add that dynamically. But let's use a smarter way. So right here we are launching our Excel sheet. But opening the document that we have just saved previously. So after we open the document, we're going to get the first free column or row in our Excel sheet. So what does that mean? That means that it's going to go to the end of the worksheet and it's going to tell us which Excel, which row inside of our cell is free. So in the Excel that I'm opening right now is going to go all the way until the end and it's going to figure that out. So right here, I'm going to start at column a, start at row 1. I know that is rho L naught, which row is going to end. So I will add the last first three are o. However, the first three row. If we open our Excel, the first video is going to be this one right here. I need to get the second to the first few roll. So I'm going to do first three rho minus one. Alright, so let's select the instance is going to be the second Excel instance. So this one is producing a first Excel instance, and this one is producing a second Excel instance. And let's also close the Excel file. And we don't need to see anything and we can close it like this. So this should output the data. She produced a Beta variable called Excel data with all of the files that we have inside of that Excel sheet. So let's see if this is going to work and what kind of data we get. So it's launching first Excel sheet. It has copied all of the results. However, it has failed to save the document. Let us see what is the issue. So it cannot access this file. So let's create a problem because maybe there is a same file with the same name. So to solve this kind of issue, we can create an unique, a unique name. So let's, let's add a date and time, and date and time is always going to be unique. So let's get the current date. The current date, and we are going to save it to a variable called current date time. After that, we are going to convert this date into text so that we can use it. We can't use it once a day. So we're going to select here, right here, we're going to go to convert date and time standard, and we're going to select a format that we want. So let's select a format that might work with Excel. So I think this, this format right here might work with Excel. All right, so let's add this right here. And when we are saving our Excel file, let's save it with that date format as well. So let's go right here, where we close our Excel sheet. And let's add a variable right here in between. So let's add the, the time, the formatted date and time. However, we also have to remember that we have to also when we are opening, we also have to open the same file. So we have to add a variable here as well. And is the format the date and time. So one little fine tune to our process, converting it to long date so that exceptional have any issues saving it. Let's launch our process and see if it is going to run. So it has successfully managed to get the candidate converted to time and texts, get all of the files inside of this folder, save them, open an Excel, right? All these documents inside of an Excel sheet, close the Excel and save it with the current date and time. Are we going to see what that looks like in a second? Then it launches again the same Excel file and just saved, it gets the first three column and reads all of the content inside of that file, and then it closes the file again. So let's see what that looks like. So this is our file that it has just created. Let's see its content. So we have all of the files that we have inside of 322 files. So that is exactly what we have right here. We have 322 files. So that is in a nutshell how we can use Microsoft Power, Automate Desktop, and excel in tandem to do different things. Now there is a lot of advanced. So advanced Excel options that are pretty straightforward how to use. So I will just give me, show you at a glance different actions when we need to perform them in Excel, we can copy these different actions. So for example, if we need to insert a row inside of our worksheet, we will copy this. We will select Excel incidence that we are working on, since we're opening two instances here, we're going to select which one we're working on. And then we will put row index. What is row index? We would usually go to this information and we will see that the index of the row to add a new row above. So what would we put? We put the number starts from one, so I will put maybe 34 if that is where I would like to insert a new row. So all of these are pretty straight forward now that you understand how to utilize excel at the high level, you can dive into these and discover them for your specific use cases and your specific processes. 29. Lesson 4.6 - Send Emails: E-mail automation. In this chapter, we are going to see how we can automate and send emails with Power Automate Flow. This is going to come in handy in many processes that we're going to build. So that's why we're going to take a closer look at the different actions and the key concepts so that when we start building the processes, we know what we need to use and how it is going to behave. All right, so to automate email, I have prepared a couple of actions, but I'm going to go over these actions when you. So for e-mail, we have three boxes in Power Automate or three set of actions. We have email, we have Exchange, Exchange Server. So if we have an Exchange server inside of our company, we need to connect to the Exchange Server. We can use this action side here to connect, retrieve emails and send emails via exchange. We can also use Outlook. So if we want to use the algorithm that is installed on our computer and we want to open it just like we did with Excel previously, and need to process emails, respond to emails, and so on. We can use this actions as well. However, the simplest and quickest way to get started with with email is by using retrieve email process email and send e-mail. So let's start with these. Let's learn how we can retrieve our inbox or the e-mails that we have sought to that we will. And this action right here, and you see it right here, retrieve email messages. Now there is a couple of things that are sending and receiving email messages requires. So it requires a server address and the port, Gmail, Outlook, Yahoo, and all of the e-mail providers. They will provide you with a address if you type in into Google and you search for the e-mail server for Google, for Gmail, you will get different adjusts. The next thing that you need to do is you need the port which is going to connect. Usually the port is going to be provided with the server. Adjust. After that, you require your username, password, now you are ready to connect and retrieve emails. Another observation or another comment is that e-mail servers are used to retrieve email. Why SMTP servers are used to send emails. So these are two different connections. So if you hear SMTP or image, just know that SMTP is to send an email, is to retrieve. All right, so here we have the action to retrieve email messages. I have already inserted this, and it's also good practice that we put this inside of a variable. So in this case, I would put it inside the image variable so that we, if we use it in several parts of our process, we don't have to write it over and over again. We also have this useful thing that is called email filter. And that is going to allow us to be specific on exactly the formulas that we want to achieve. So in this example, for example, I want to retrieve all the unheard email messages that are currently in my inbox. I can also be more specific and specify the subject, the body, the two, and so on. We can also be specific with the Apache. So let's save this and let's try this process and see you can go into the, retrieve the email messages. One more observation is that the variables that are produced are here, retrieve email messages. So everything that is retrieved will be saved if that fiber called retrieves e-mail messages. All right, so we have seen that it has extracted one email item. So inside of this email item, since this is a list, I can click on More and I can access different things so I can access the attachment if I wanted to. I can access the body of the e-mail, the body text of the message, the date from who it is I sent, and information that I can also further go into beta. So for example, here inside of the body texts, I can see actually what the body text looks like. Alright, so this is how we would achieve email messages. Let's do one more modification to this, and let's retrieve all of the eBucks that I have inside of my email. So to do that, we would go to email filters. We would select in Box folder, and we would retrieve all of the messages. Now there is quite a few messages, so this will take a couple of seconds to complete. So I will be back in a second. All right. It has finished with fearing all of our email messages. And let's see what we have. So right here when I open my variable, I can see that I have quite a lot of email messages. I have actually a 147 and I can get the detail of each one of these messages as well. So that is how we can achieve all of the messages and how we can filter them if we're looking for a specific message. Now the next part of understanding how we work with email is to send an email message. Now sending an e-mail message is equivalently easy. We drag and drop this activity right there, and then we configure the details of the action. So we insert first of all, the S and P p parameters. We enabled SSL and will also enable SMTP server needs authentication. This will allow us to login side of our Outlook account and be able to send emails. Now the server port for SMTP is 25, It's not 99, something like four EMA. So keep, keep keep track of that. It might be also different for every provider. So after we've configured our SMTP connections down here at the bottom, we need to configure the email that we want to send. So from who we are sending it from. So we are sending it from this account and we can specify whatever we want. Display message that is going to be sent to address a CCC, a bcc, and any other information that that that is standard for any kind of e-mail. So once we save this, we can move this above and we can send this message and see if we can automatically send e-mail messages from Power Automate. Now imagine how powerful this would be if you put it inside of, in a loop that has an Excel sheet and inside of the Excel sheet you have list of different emails that you need to send to different people, maybe hundreds and hundreds of different emails. You could all do it with Power, Automate. And we will see later on how we do an example like that. So right now I have sent an e-mail and if I go to my inbox right here, I can see that I have received an e-mail to me, hello from Power Automate. I have sent to email one was received 0 minutes ago and 122 minutes ago. Alright, so that is in a nutshell, how we send emails from Power Automate Desktop. 30. Lesson 4.7 - AI Services: All right, now let's create a automation that uses AI inside of Power Automate Desktop. Now this is not going to be as easy and as straightforward as in Power Automate flow. But sometimes you cannot use Power Automate Flow. You have to use Power Automate Desktop. And desktop is a solution for pretty much anything. So what are we going to do? We're going to analyze. We're going to first extract some news titles, and then we're going to do some sentiment analysis on this news titles. So to start, let's open our blank canvas and let's add the actions that we need. So we need the web browser. First of all, we need to launch the web browser. We need to launch Google. And we need to paste in the link of the webpage that we are. Ultimately. We're also going to select maximized so that we can see what we're working on. After that, we are ready to use the web recorder. And to use the web recorder, all we do is we have to just extract and make sure that we're on the latest news, extract the titles of these pages. So let's record, Let's get this title right here. And let's get the following type of as walked. And that is all that we need right now. So let's finish this. And let's wait for our robot here. Here phase it has created everything that we require. Let's delete all the actions that we don't need. All right, Now that we have everything inside of this variable, now we need to do some sentiment analysis. Now there's different ways that we can do sentiment analysis. We can do a custom API or we can go to a web page and just integrate web page and shy child like that. Or we can use some out-of-the-box capabilities that are available from, from, in the cognitive section, from Microsoft, from IBM, and from Google. So here we have different APIs that they, they have already built that work already. And we're going to use the ones from Microsoft. So we're going to use computer vision, we're going to use text, and we're going to use sentiment. So all we have to do is drag and drop the sentiment analysis here before adding the subscription key. Right here, we are going to add the text that we are going to analyze. So before also doing that, we are going to have to add a loop. So right here we're going to add a loop. This loop is going to iterate on all of the data that we get inside of the output data. Then we're going to put in the sentiment inside and 4D text. We are going to use the current item, this item right here, the English is the language, and that is all of the information that we need to set so we can save this. We are missing a key and I will show you write in a second how we can get the key for this. Let's jump into Microsoft Azure, where you have created a portal and login. After that, you will create Text Analytics Services. So you will search for text Analytics and you would select this service right here. Now, I have already created it, and I will click on this service. Once the service is ready, you need to get your API keys right here. So once you click on your API keys right here, you will be able to copy and paste them inside of Power Automate. So if we go back to Power Automate and to our process, now we can add our subscription key. With this subscription key, now we can run the sentiment analysis on this text. So let's run the process and see if it works. Remember also to make sure that you set your region in the sentiment is to the right region. And we can see in the result of the JSON response, then the score of the analyzed news is 0.5. So that means that the score is between 0 and 0, 1. We can see that right now the sentiment is no charge for this news. So that is how easy it is to extract data and do sentiment analysis on it. We have also the ability to do other things like Key Phrase Extraction, detecting the language. We can do some computer vision like analyzing images, describing what is in an image, reading the text with OCR that is inside of an image or tagging images. We can also do a spell check, which is a great functionality that I use often on different documents. We can also use Google and IBM provider. So if want use Watson from IBM and thus says we have an if also want used Google. We have those APIs as well. There's not that many out of the box APIs, but we'll every new update that adding more and more. So thank you for watching and I will see you in the next video. 31. Lesson 4.8 - Building Strong Flows: Creating strong and reliable processes is crucial if we don't want our processes to break. One of the main reasons why processes break is because the underlying application or web pages were trying to run our process on has changed or slightly be modified. And our tools and our robot is not able to find whatever it's supposed to interact with. Therefore, building strong and reliable selectors is crucial for creating reliable problems. In this class, we are going to learn a couple of tips, suggestions, and very great insight on how we can create very strong and reliable process says that will never break. We're going to start with selectors. We're gonna take a look at how to build dynamic and strong selectors. After that, we are going to take a couple of look at some insights that I have learned that will help you make reliable robots. All right, let's start with the selectors. So I have selected two webpages cross. One is this. And what we're going to try to do is we're gonna try to click on the Follow button, this button right here. So what we want our process to do is click on this button repeatedly until it follows a certain amount of people. We don't wanna do it manually. We want a robot to do it for us. Now we also don't want to spy every single button. Now remember there's different ways that we can do every single process. There is countless ways of we can do one single automation. However, in order to create strong and robust processes, the least steps that we have in our application that is the best. Therefore, we have tried to be as quick as possible and as I tried to utilize as little actions as possible, the least actions, the last maintenance, the last development, and the stronger the robot. So let's run our process and go to the page that we want to spy, different elements. This is the page we want to spy, the 10 first people. So let's see how we can create a dynamic selected. The easiest step to do is use the web recorder and to spy different elements. So we can spite 10 buttons right here. So let's buy a couple of buttons. Now here is a useful Dep. You should always try to find the button on which you would want to click. So this is a button I would like to click on the button or on the link. In case I cannot find the button or on the link, I'm going to try to select the smallest, the smallest element possible. So instead of selecting the one that is above it, I'm going to try to select the smallest y because this is going to give me the most reliable selector. So let's select this. We've selected this span and obviously it can vary from page to page, certain pages you can do this kind of things and different pages you cannot. So on one page, the innermost element will walk. And on another page, the outermost element will walk. It is up to you to try and figure out what is the best. For my experience on the majority of cases, the inner element is the best. So we have recorded them. We can see that we have recorded the same element, spans, spans, span and span. They are all different because it has created a different UI element for each one of them. Now let's see if I work work, click on something different. So we see that now it created a div tools. This is a different UI element. We have to be very careful while we click on our page. So let's, let's finish this. We have spied enough elements and let's try to run our process and see if it's going to walk right out of the box. So it is opening Chrome, the first step is working fine and it's following the first people. However, it is, it has broken already. It is not able to identify this element right here. So why is that? Why is, why is it not being able to identify this element here? Let's try to debug and understand why that is. So if we go to the span three right here, and let's go to the, let's take one of these. And let's go right here and try to compare these two elements. So I'm adding a new element. And now I have the previous span and this one together so that I can see what are, if any, what are the differences? So we can already see a first difference, which is right here. So we can see that this changes. So this is four and the previous one span two, which is the one that has failed is span it is for so that could be the only difference. Let's see if there's anything else to this. And all right, and let's see this 12 divs and more or less the same? The same? Yes, exactly the same. The same. The same formation. So let's try to spider web recorder one more time. Let's see what we get. So what we are trying to do is work trying to spy this button right here and understand why hasn't been why hasn't it been identified? So I'm going to click on it and we see that it has been identified as span five. So let's finish this and let's take a look inside. So let's let's go to span five. And let's try to understand why. Why did it not walk? So to do that, we're gonna go back here. We're going to add one more of these. And we can already see that the syntax of, the syntax of this selector is different. So we see that this number changes. However, the syntax afterwards is different. This one has a third div right here. So that probably means that let's go back to the web page that there is something extra between probably there is an extra container or next day or next as something that is making this changes occasions and have it, have another div. So right now, let's figure out first how we can make this into a dynamic live. You click on all of the buttons. So to do that, we see that this is primarily the number that changes afterwards. We can have different changes. What we will deal with that in a second before that, everything is the same. So this is the only number that changes. So if I'm able to put a variable in here, I can create a dynamic selector. If I can put a variable that goes from 123456789 and so on. I can probably have it click here, here, here, here, here, here, and so on. If all of these have an incrementing number right here. So let's just make sure Let's select one more of these. So let's see this one. And this one is 5 Friday. So if I, just for curiosity's sake, I copied, I go back to the 2D one that I'm working on. I create a new one. I paste this. And now we can see that. We can see the difference. So we can see that it is incremented 34345. And this four right here is the follow that was not being able to click. Now when I have added all of these, I can see that now I can create a dynamic selector. To create the name selector, we have to add a variable. So let's add set variable. And we're going to add this variable to 0, 0, let's add it to one. So this variable right here, we're also going to rename it to make it a little bit more understandable. We're going to call it increments. And then we're gonna go to our selector right here. And we're going to insert our variable in place of that number that we saw that it always changes. So just to see one more time its place instead of this number right here. So I will go right here. I will take away this number four and I would insert a variable increment. So that is pretty much it. Now I don't need this anymore, and I don't need this one anymore. I don't need this one. So right now, I know that I have two selectors. I could have possibly two different selectors right here. So I will leave this other selector and I will, instead of the four, I will also insert my variable. So the increments variable right here. So what should happen right now? So right now, it should take our one as viable and inserted here. However, it is not dynamic yet. It is not going to change. This number is not going to change. So let's make this number change. We can make this number change by adding just a simple loop. So let's add the most simple loop. This loop is going to start at 0, it is going to add at five, and it's going to increment by one. We will move our clicking actions inside and we can remove all of the other buttons and clicks. We do not need anything else in our flow. So right here, we're not using also the span we're using, I believe it's spanning tree, the one we have inserted our dynamic variable. Yes, we're using a spanning tree. We also can delete all of these other. And to do that, I can also remove all unused UI elements just to keep things clean. And I can see that I only have, I'm only using one element right now. So right now I'll process is not complete yet, because right now it will just go and use this variable as the one. We have to increase this variable at the end of every cycle so that it changes from one to two to three to four to five. So variable we're going to increment is going to be this. We're going to increment by one. And that is going to happen at the end of our loop. So let's run this process and let's see if this is going to follow five of our friends. And now we can see how the robot is going to follow the different users. Now by having integrated the different, the different elements for the follow button, we have created a selector that is strong by adding a variable inside of the incrementing number that we found inside of our selector. We made it so that now it can spy as many elements as we're able to load on the webpage. So it does 20, 30, 40, 50 different elements and we make it loop. And then 50, it is going to click on all of the different buttons that it has. And here we can see that it is just continuing, it is going to continue clicking through the different buttons. So, ladies and gentlemen, that is how we create strong selectors very quickly. And creating strong selectors is going to make our processes reliable and not fail. Thank you so much for watching. Right now we are going to discuss a couple of tips and tricks that are going to make your processes even better. 32. Lesson 5.1 - Lesson Overview : Lesson 5, it has been quite the journey, ladies and gentlemen, in Lesson 5, we're going to be putting everything together. What are we going to do? We are going to connect Power Automate Flow and Power Automate desktop. We're going to pass information from flow to the desktop. So let's say we have a process that has collects information from the Cloud, from an API microsoft Cloud. We want to pass that information to a desktop process. So we will learn how we can work with cloud and flow at the same time. We'll also learn how we can schedule, That's what flows. So that's the flows currently you can just push a button and sugar them. But if you connect desktop to the Cloud, you will be able to schedule your desktop flows as well and have them work together, which is really going to give you the ability to do end-to-end process automation and complete and do really great processes. But at this point, you have learned all of the knowledge that you are required to automate pretty much any process. But now a question arises, what should you automate? So in end-to-end process automation, we are going to go through a series of methodologies and frameworks that are used and that you should utilize when selecting the right process for automation. So here you will learn what process is going to be the best candidate for you to automate. That is going to make you save money. And it is going to be the lowest hanging fruit and it will give you the best benefits the sooner. We'll also take a look at the final project that I have prepared for you. And I will give you complete at the end of this course. And we will also take a look at what you should do after this. Now with all of this knowledge, what is next? Let's jump right into this. 33. Lesson 5.2 - Process Selection : So you have learned how to use Power, Automate Flow and desktop. And you can build great automations and you can automate any process inside of your company or for yourself. But how do you actually find those processes for automation? That is a really important question. Without knowing the answer to that question, your automation skills are pretty much useless. Automating wrong tasks, wrong processes, wrong activities is just going to be a waste of effort and also a very negative showcase of the technology. Therefore, in this chapter, you will learn how to best choose the processes for automation. But it's not only about choosing the process for automation, it is also about other things as well. Keep your ears open and listen to what I have to say next. The main triggers points for identifying a process for automation are the following. So for example, if we have processes where we have lots and lots of data entry inside of the application. So for example, Let's imagine a hospital. Would patients entering the hospital all time, somebody needs to input all of this information. Therefore, this R, this could be a good use case for automation, maybe processes where we have high error rates that are that, that could be checked automatically by a robot, reports and so on. Or maybe work that doesn't have a lot of low value rework. Maybe work that is just repetitive or manual. Numerous processes such as moving application of data application from one application to another, copying PDF information inside of an application, just very numerous manual activities that are repetitive. Or for example, we have integrated a legacy system. So we have a process where we have two different legacy systems communicating or moving data between them. That is probably a good example for automation because there's probably a human endeavor that is taking the data from one place or another. And if it's not a human, maybe it's a script, but that script has to be launched, manage and maintain, and all of that could be done with a robot. Now, there could be also a process that is high in turnover due to repetitive and low value-added activities, which means that people usually quit tasks and processes where they have to do very repetitive activities. Either people just Satellite accepted and they repeat the same task forever, or they get tired and they look for something better. The majority of people gets tired of this kind of processes and those should be a red flag for you to go there and automate that process. Now, with that knowledge in mind, let's take a look at a couple of examples, real life examples of good use cases inside of a company that can be automated. This is an illustration of arche potential different functions inside of companies. So in information system, we have processes like batch management or pipeline Management, Server Management. All of these are activities that are repeated daily, if not whenever something happens, that could be easily done with automation. I have personally worked on numerous IP automations. One of the latest project was for managing data pipelines for a specific for a specific client. And they had different data pipelines that they had to check and make sure that data pipelines completed all of the data migration and they have login every day and check that that was highly repetitive, highly manual. So we automate that automation and with the robot. Then, for example, in production we have inventory management plan, management, logistic management. And to give you an example of production, we automated a process for a large, very large company that has a webinar about this process as well. We managed to save 7 million and it was in the logistics management. We managed to automate a warehouse process that was making sure of the inventory that was inside of the inside of the warehouse and was making sure of the expiry dates. Now before we had the robot to check this and compare all different information, there was a human that was going in and they were nearly not quick enough to be able to keep up with the demand and with all of the, all of the data that was coming inside. So they The plan was to have an additional person, but we automated the process. Human resources with financing are some great, great use cases for automation in financial department, you are going to always find repetitive and manual activities. Why? Because they deal with many ripples creating, paying invoices, receiving data, bank reconciliations, and many tasks and activities that are very repetitive. They are done on a regular basis and they always have to be done. Therefore, the finance departments should also be one of your main focus if you're looking for processes to automate inside of your company, finance department is always one of the initial use cases as well. Many companies start their RPA journey from the finance and accounting area. You can go through the rest of this, of this, I wanted to highlight some of my experiences and also the main area for starting with with automation. So how do we choose the right process for automation? Let's say we've identified 2030, 40 different processes for automation inside of our finance department. Now what, how do we prioritize among these processes and how do we choose the right one so that we are 100% sure that we are going to be getting the right benefits. And the complexity of that process is going to allow me to actually automate it with the budget that I have. Now, there is different approaches to this. You can use a digitalized tools such as process lands or should boomy or automation hub that can help you digitalize all of your information. However, the methodology is the same. There are a couple of things that you have to keep in mind for this to work. First of all, you do an initial collection of all of the ideas and all of the processes that you want to automate and that you have that you have identified for automation. And this ends up in your automation funnel. Now, once you have identified this, then you take some key criteria. And based on this key criteria, you select what would be the best processes to automate. And this key criteria is like frequency, process frequency. How often is this process repeated? Is it repeated every minute, every 10 minutes, every hour, every one month? Obviously, distributor will vary one month. The process is not a great use for automation. If it's repeated every day, that might help more than if it's repeated every month. Data into volumes, how much data is this is entering and how many, how many forms, for example, are being inserted or or utilized. How many people are working on this process or FTEs full-time equivalent. Also, if the rule if the process is a rule-based process and other criteria as well. So once you have identified a few candidates for automation, you shouldn't just jump into automation and start automating. You should restructure your process. Don't automate the steps of the process and not redo your process with automation just the way that it used to be before. Now's your chance to reshape and reimagine your process with the new technologies that you have available. Now, if you have AI automation, RPA, Cloud flow desktop, and you have all of these technologies available. What will your process B, therefore eliminate all of the extra actions that are currently inside of your process. No need to do them. Is there really a reason for doing this things? Ask yourself, simplify your process as much as possible. And I have highlighted this a couple of times inside of my training. Build simple robots. The simpler, the better processes should be as simple as possible. Standardize your process flow. This means that you should not have as many exceptions as possible. If you have as few exceptions as possible, means that your process, if you want to automate it and do it in a very reliable and stable way, you should have the least exceptions or variations of your process possible. And then once all of this is done, this doesn't take a lot of times, it just takes a little bit of thinking. And once this is done, you are finally ready to automate your process with RPA and intelligent automation. Now, a couple of points on what is going to make your adoption of RPA success and what is not based on the experience that I have. So if you're an organization or an individual, you should use an agile business led approach to implementing automation. Automation projects did not happen over a long period of time. They are rapid in development, especially because you are using low-code tools. Therefore, you can use agile methodology to build your your automation roadmap and build your processes. And that has to be your frame of work when working with intelligent automation. You also have to act fast. You have moved fast and demonstrate results quickly. Rpa is a new technology that hasn't been yet adopted very widely inside of organizations, not many stakeholders are aware of the benefits. Therefore, you need to show KP main KPI metrics and show them as soon as possible. You also should focus on scale and frequency, and not just end-to-end process automation. This is a very important point. Don't try to automate the process from the end until the finished. It might be easier, just automate certain parts inside of a process. So for example, if a payroll process, if it's involved for many different actions, maybe just automate the initial actions that extract all the information from PDFs and make it available for the person to continue with the work that they have to do. Therefore, keep in mind that it is not always the best automated process from the beginning until the end, even though like that, you unlock great value, but it is faster to build different little pieces of automation here and there where they are required. Once you have completed those little initial pieces of automation, you can build on top of those and add other components. Once you have realized initial benefits. How you're going to fail in automation is if you document your existing processes way too much. So with traditional IT, there is a lot of documentation that is involved in all kinds of requirements, counter requirements, project plans, exceptions, and so on. That is all great documentation. I'm not saying that those documents are not required. However, if your objective is to scale a 100 automations, automations, 300 dollars donations in a year. How can you manage all of that documentation for all of those processes. In addition to that, creating, hey, projects doesn't require this kind of level of detail using local tools and technologies that are added on top of your existing systems. There is no reason to overtake and over-engineer this processes. They have to be reliable and stable and managed. But that is, that goes well beyond what our currently the requirements for many software vendors regarding the documentation that should be created for RPA processes. Also, don't overtake where you start. Think about the processes that are quick, wins and processes that are low hanging fruits, those would be your first priority. Instead of automating processes that are maybe big pains in the ass for the company. And the processes that could potentially be in the eyes of everybody, focus on the low-hanging fruits and on the processes that are going to give you a benefit immediately so that you can prove the value of RPA inside of your company. And then you have justification to go after the larger processes. And also many organizations, they start with the pilot, but they don't plan the roadmap. They don't think about the long-term. They focus on one single automation to automation, three automations, a small pilot, but they don't consider the big vision and big roadmap that they require. They don't plan the budget. They don't allocate resources, training, and people to this kind of commitment. Therefore, with a small initial investment that doesn't get sustained, it is absolutely not enough to start a successful automation journey. Now, with all of these things in mind, usually able to identify processes inside of your company, be able to automate them with different technologies, such as Power Automate Flow or Power Automate Desktop, and be able to communicate and present the results of what you have automated solid, you can get further support for further automation. 34. Lesson 5.3 - Desktop To Cloud: You have learned how to build flows in the Cloud that get information from APIs. You have learned how to build flows on the desktop that integrate applications on your computer and different web applications. Now, let's put the two things together and just imagine how powerful your clothes will become when you can put together AI flows and all of the different API flows that you can create in Cloud flow. And you can combine it with the Power of Power Automate Desktop. Now let's see how that is done. In one of the initial classes. We have set up on our computer, we have set up the connection, the connector between when the gateway between flow and desktop. Now let's see how we can create a process that is going to be triggered from the cloud. So let's create a process on our desktops. So this process is going to be called desktop process. It is just a simple process that we'll get some data that we are going to connect, collect with our Cloud flow. So let's say we're gonna get some data like the weather information for the weather for specific location. And we want to give it inside of our floor, ikea. And here we can do different information with it. So let's just display this information with a message. So let's put a message and let's display a message. And let's try it. The weather report. And here we will put our message. So to do that, we first have to add some input, output variables. In this case, we are going to be using input variables. And input variables are just basically variables that we create and we collect information from them. Now also keep in mind that once we're going to be creating our Cloud flow, we're going to see this name right here, external name inside of our Cloud flow. When we connect to this, we're going to leave this as new input. This is going to be a text. The default value is going to be empty and we're just going to create this variable, new input. After that, I will put this variable inside of my message and I will add it as new input right here. So like that, we will get the variable and we will display it in a pop-up message with this flow desktop process. That's the name of the flow. Let's save this flow and now let's go to our Cloud process. Let's create a new process. And you should be very familiar with how to do this. We're going to create the Cloud is manually triggered and we're going to call it trigger and pass data. Always name your flows with something meaningful. Alright, so the first thing that we have to do is we have to get the weather. So let's get the weather with Microsoft weather. And we're going to get the forecast for today. So let's get the location. We're going to write MSL. And we're also going to use metric units because I'm not very familiar with impaired. The next thing we have to do is we have to trigger our desktop flow so we can search for desktop. And once we find desktop flow, we can add a running flow with Power Automate Desktop. So this is the action that we need to use. Run a flow build with Power Automate Desktop. Now, once we connect this action, it is going to load all of our processes right here. Now, I have to select the correct connection that I'm using. And once I select the correct connection, I will be able to see all of the different processes that I have on my desktop. We have seen this previously, so this should be nothing new. Let's select the desktop process that we want to use. And let's select also the run mode. In this case it's going to be attended runs when you assign in, inside of Power Automate Desktop. And here we have the new input. Once again, if we recap in our, in our desktop flow, we have an input variable that is called, that is called new input. Now right here we're going to add inside of the new input, we are going to add the conditions of them are coming from here in a dynamic manner. And again, 0 code required to get all this API information from Microsoft, whether we are going to save our process. And we are going to also change the connection and make sure that I'm using the right connection. Once our process is saved, we can test it out. So what this process is going to do is going to be triggered manually. It is going to get the weather forecast. It will activate the desktop process. And if we pass some information like the weather condition, that is going to tell us how the weather is in MSL. So I can run the process from right here, but let's run this process from my flows right here, trigger desktop process. Let's click on Run. And let's run this process. We can see what we're using right here and that all far connectors are walking at the same time. I also want to remind you that you can also run it from your mobile device by creating also a button. We'll just a simple click of a button. You can trigger both of the flows. So let's continue and run this flow. Let's run it. So it is done. The process is currently running. It will take a couple of minutes to trigger Power Automate Power Automate Desktop. So we can add the same time we can go inside of our process and we can see how it is working. We can see that it is currently working right now and it has executed. So it is telling us the message has returned and is telling us mostly cloudy. Now, once I stop it, we can wait or we can refresh this and wait a couple of more seconds. And we'll see that our process has completed running. So if we refresh this and we can see that our process has completed successfully, we can click inside and we can see all of the different information that has been passed. So new input, that was the data that we were passing to our desktop flow. It has been populated as mostly cloudy. So we got the weather forecast, we got all of this information we chose just to pass the mostly cloudy. That is, how we can pass data from our desktop flows to our Cloud flows and vice versa. So right now we've just seen how we can pass the data with an input. But if we wanted to collect something from our desktop and pass it to the Cloud, it will be the same concept. We would create an output variable right here. We would fill out the information. And in our Cloud, we can, once we connect to our, once we connect to our desktop flow, we can collect the information from it. So that is how we use the two technologies together. Remember that this is also, this can also allow you to schedule your desktop flows. So from your Power Automate, if you go inside of if you go inside the floor flows and you decide to create a scheduled flow, you can create scheduled flows for Power Automate Desktop. Therefore, you can automatically trigger your Power Automate Desktop flows. So that is done just by creating a scheduled flow, adding the name of the flow, selecting their occurrence and that we want, in this case, we're not going to use five minutes creating it, adding a new step, connecting to the desktop, adding a Power, Automate Desktop flow, and just choosing the process that we want to use with this, with this, with this flow. So that is how simple it is. And I really invite you to try this out for yourself and see how the Power of Power Automate Flow and desktop when combined together. Also when combined with Power Automate AI. 35. Lesson 5.4 - Final Project: Welcome to your challenge. So what are you going to do? You're going to create a news robot. Imagine having a robot that is going to everyday search the news, find the top news, convert the headline of this news into voice, create a video, a short YouTube video about it, and posted automatically to YouTube. You should have the knowledge and the resources to completely create this without any problem. In this video, I will break it down to you and show you what should be the main components for you to create inside of this flow. And then I will show you a small recording of this process. I would like to give credit also to fab your lawn or Donny, who is one of my students that has created hyper news, which is a robot that does exactly this and that I will be presenting to you at the end of my description of this robot. So with that in mind, let's go over this robot. So your robot will start by collecting some news from top RPA sites. So you can collect news from different websites, either with Cloud, with RSS feeds or API connections, and store that information inside of an Excel or inside of a Google Sheet document. After that, you should also check and make sure that when you import new news inside of your document, that the news is always reason and that you have that you're not over writing the same news. After that, you need to get the headline of the news and converted to text to speech. Different ways to do this as well, you can use a API and convert text-to-speech with flow. Or you can use desktop and go to an actual webpage that does text-to-speech, copy and paste your text inside and just download information like that. After that, you need to download a couple of images so that you can create this 15-second video. And you also can enhance the description of the video by adding sentiment analysis or saying if this news is fake or not. Also via API or other ways, you need to download images that you need to create the video. After that, in the final part of your robot, you can create a using whatever tool that you want. But I will show in the video that we've used Microsoft Movie Maker or the one that is attached with every licenses of Windows. So you can use that and with desktop recorder and imagery color and by sending keys, you can automate and create different create different movies with all of the stuff that we have collected. So you will add the pictures that we've collected regarding the title of the image. So the robot made the search on Google for the images associated with that, US. And it now it's adding them inside of the video. After that, you should post the video to YouTube and schedule it for specific date and time. So that is the video. Let's, let me show you what this robot looks like. It is called hyper news. And this is just a demonstration video for any questions or information regarding this robot. If you have any problems, please make sure that you leave some comments, your questions in here, and I will do my best to answer your questions. Okay. Two. Here is a okay. 36. Lesson 5.5 - What's Next: Well done on completing this course, you have learned how to create clouds in the flow clouds on your desktop. We put them together, we create an AI flows, we automated pretty much anything you should be able to automate pretty much anything. So what's next? You might be asking yourself, well, what's next is up to you to decide if you want to continue your learning journey. There's definitely lots and lots of resources that you can go through and continue your learning journey. You can also jump right into building automations and flows. You should have all of the skills required to build complex automations that require even a BI Desktop APIs and many different components together. However, if you want to also learn more and continue your learning journey or Microsoft Power Automate, there is loss of more things that you can learn. Now, two easy ways to access the learning components of Microsoft Power Automate right here from flow. If you log into your flow and you click on your panel all the way down, you have a button that says, Learn. If you click here, you will be redirected to the Microsoft Ignite page. And you will have lots of great content grouped into different categories on Power Automate Flow and on Power Automate Desktop. This is great content, great videos. It has been grouped together. It has been grouped together according to logical to logical grouping. After that, if you also go to your Power Automate Desktop and from Power Automate Desktop, you go to Help and you go to learn right here. And if you click here, you will be redirected to the training of the filtered side, where you will find different trainings that are already grouped for Microsoft Power Platform. Now, these are all courses that once you complete, they will give you towards your experience level as a Microsoft Developer or Microsoft member. And it will award you badges, as well as the more courses you complete. That is different certifications that you can do on Microsoft Power Automate Desktop Power Automate Flow, and Power Automate apps. Now, if you're interested in certifications, certifications will definitely allow you to get a job as a developer with Microsoft Power Automate Desktop or flow, this kind of jobs right now are on the rise and there is very high demand for this kind of scales in the marketplace. Therefore, learning it is very important. There is other resources that I will share in the links below, and they are useful links like this in this resource that is, Microsoft Power Automate learning resource is just a large collection of different links that are useful, that will have you covered for pretty much anything that you need in Power Automate. They are grouped with different levels. So for beginner, intermediate, advanced, you have Power, Automate Desktop administrator, and also community User channels that are primarily on YouTube. All of these are great creators that are available on YouTube that make amazing content of Microsoft Power Automate. And I highly invite you to go check them out if you're interested in learning more about flow and desktop. I hope that you have enjoyed this course, that you will get involved with me and the community. Ask me your questions and I'm really looking forward to your questions. I'm looking forward to know if you're how you're implementing the automation powers that you have learned and how your automation journey is going. There's different ways to reach out to me. First one is to leave me a comment bounce bound in the comment box. I'm really looking forward to it. And once again, thank you for watching my course. I really hope that you got the most out of it.