Data Visualization: From Data to Insight

1. Introduction: Hi there. My name is Josh and I'm the Manager of data and analytics for multi-billion-dollar organization. I also have a particular passion for telling effective stories with data visualization and data communication. In my role, it is crucial to tell effective data stories so that they can influence and drive data-driven change in the organization. I'm also an instructor on all things data and coding. And I'm excited to welcome you as part of this community of enthusiastic data professionals these days, data is one of the most powerful assets and organization has every top performing organization in any market that you want to pick is there because they have better data capabilities than their competitors and dates professionals are some of the most in-demand skill sets that recruiters are looking for today, organizations are currently investing billions of dollars to enhance their data capabilities. But let me share a little data-driven fact of view. Data costs. Let me share a Data Factory view. Most data initiatives fail to yield any results. Now, this isn't because data initiatives fail to gather the right data, or they failed to implement the right technology, or they don't have the correct people. So the contrary, most data initiatives set out to collect certain data and analyze it. And they achieve on that goal. The gap is, after presenting this data, it doesn't actually influence positive change in the organization. This is because data isn't presented in a way that's compelling and influential, they will actually drive actionable change to the organization. In other words, the problem is, people aren't utilizing data-driven storytelling, which is exactly what we're gonna be covering in this course. Storytelling is one of the most powerful ways to influence, teach, and inspire. When you compare that with the analytical truth of data, then you have data storytelling, which is one of the most powerful tools in any organizational toolkit if you want to drive business value as you join me in this course, we're going to learn to take borrowing normal data and turn it into powerful, story-driven messages that will inspire your audience to take change that you want to influence through your analytics. We're going to explore the five key principles of visual perception and visual design and how these can apply to the world of data visualization to tell impactful stories that will change an organization. Your project throughout this course will be to take some data and turn it into a powerful message using the techniques that we learned in this class. So welcome to the class. I look forward to seeing you in the next lesson. 2. Data In our Modern World: Hi there and welcome to the first proper lesson of this Skillshare class. In this lesson, we're going to explore why data is important to us in the modern world and why data visualization is a crucial skill for everybody these days, we consume data every second of every day. In fact, over 3 quadrillion bytes of data are produced every single day. And the pace in which we're creating data is accelerating rapidly as more people, organizations, and even computers are generating more and more data. We're currently truly living in a data-driven world. The reason we collect all of this data is to make informed decisions with this process is known as data-driven decision-making, which is a pretty appropriate name actually. Now, if you think about it, all decisions are actually data-driven decision-making. Unless of course you're just consulting the magic eight ball to demonstrate, let's think of a mundane decision that needs to be taken, such as which hotel you're going to book for a trip. The decision you make about which hotel is based on data that you collect. The data you are collecting in this case is things such as the amenities and the facilities and the location of the hotel, the cost of the hotel, how much available budget you have to spend on that hotel. And comparing this with what facilities you want and what amenities are most important to you. You'll also need to consider how long do you want to be in the hotel and the type of room that you want to stay in. There's lots of different data elements that go into making a decision, such as staying in a hotel as we go through life, we use data to inform our decision-making. Oftentimes without really thinking about what we're actually doing, about how we're constantly taking data as an input, using it and analyzing it, and coming out with decisions which result in actions that we take, such as booking a hotel. What I want you to do is take a moment to reflect on a couple of recent decisions you've made. Try and break them down into the input data that you use to make that decision and how that data influenced the decision that you wanted to make. I encourage you to post this in the discussion section down below of this course so you can share them with some others. But just reflect for a moment on some of the day-to-day decisions you make and how data is used to drive and inform those decisions. Now, this type of decision-making is not unique to our day-to-day lives. It's used in businesses all the time, only at a much larger scale. As information and technology has advanced, it is allowed organizations to collect more and more data and use it in much more sophisticated ways to make better and more powerful decisions. Now with this rapid increase in technology and the collection and storage of all of this data, we've had to develop more sophisticated ways to analyze that data and come out with the informed and correct decision. Traditionally speaking, only a few data professionals needed to utilize data skills to analyze the data that they collected. However, as markets are becoming more and more competitive, and the ground that they're competing on is data capabilities. These organizations that utilize that data the best are the ones that are succeeding the most. And in order to compete with these organizations and achieve the status of being truly data-driven as an organisation. It isn't sufficient anymore, but just a few select individuals have someday to skills. Now, everyone in the organization needs to be able to converse or speak the language of data. It isn't something that just the Buffon's and IT are doing. It's something that everybody in the organization needs to get on board with because data is the lifeblood of business. So what is data visualization on? What do we actually use it for? Well, if data is the lifeblood of any organization, then data visualization is like the answering. It's what allows data to be carried around the business and be used in useful ways to drive value to the business, which is done through influencing people to make the correct decision. Data Visualization therefore, is an encompassing term that just describes the way that we communicate data and information to an audience visually through the use of the different objects, such as the lines and bars on different graphs. Data visualization, simply put, is the visual display of data. Any instance where you take some raw data and present it visually, is a case of data visualization. And the complexity of what you present can range from something very basic to something much more advanced. But it's important to note that data visualization isn't effective if you're simply displaying numbers at its core, data visualization is simply the numbers presented visually. That doesn't really mean anything to anyone. It's only when those numbers are looked at in the correct business context with the intention of driving some kind of decision, does it actually become useful? This is where I draw the distinction between data visualization and data storytelling. Data visualization is. Turning any graph into a visualization. Whereas data storytelling is presenting data in a way that's there to influence change. And this can only be done in the proper context of decision-making, e.g. let's take a look at a data story. So this is a data story that I put together and we'll be exploring it as we progress through this course. Just for your reference, you'll be able to download this somewhere in the project section down below. I recommend you download it so you can refer back to it in future if it's not always going to be available on your screen as we go through this course. So what this data story tells us is that during lockdown in COVID times, that was an increase in physical activity for different people in America. So this is a data story because it uses specific design choices to present a story to an audience. It isn't just a simple visualization of the underlying data. It's used techniques and visual design choices that I've made deliberately to tell the audience a story. What I want you to take away is the data is used as the evidence to the message. The message takes front-and-center stage on the actual data elements, the lines in this line graph. That's a reinforce the evidence for the message. And that's to me what separates a data visualization to a data story. A data visualization would just display these lines and say here's the number of people doing outdoor activity over time. Whereas the story tells you a story about it. It gives you a key message that COVID-19 inspired more physical activity and specific intentional design choices have gone into this to tell that story effectively, what this visualization shows is that we can take data and through intentional design choices, we can turn it into a message that can inspire and influence an audience. These days, everybody in the organization needs to be able to speak this language if the organization wants to be considered data-driven in its approach. And since these days we're living in a data-driven world, then everybody is expected to participate in this new language of data. Data visualization these days is incredibly prevailing. They can take many forms such as dashboards, infographics, presentations over pods. Just think about how often you encounter a data visualization in your day-to-day life. Perhaps like me, you have something like this, which is like a step tracker app that visualizes for me my heart rate over time and exercise done and how many steps I've taken daily. And at the end of the week it produces a small report that I can look over. This is an example of visualization in day-to-day life, data visualization is absolutely everywhere. It's on advertising and billboards. You can see products advertising themselves using data to compare performance between themselves on different products. You can see data visualization all the time on the news to inform you of stories that are happening. Now these are good examples of data telling stories once again. And importantly, how many meetings have you intended recently where someone has presented a graph or some kind of data visualization. These days, it's everywhere all businesses are attempting to do data-driven decision-making by using data to inform their decisions. And as we said earlier, most of these initiatives fail because the visualization itself doesn't actually tell a compelling story that leads to change in decisions being made. All of these examples show the surveillance of data communication and the need for everybody to be able to speak and communicate effectively with data. Companies are so relying on this scale today that being able to confidently work a spreadsheet and create some graphs using something like Excel is now no longer going to cut it in the modern day world. Nowadays, you need to be able to confidently work and analyse data and turn it into compelling and influential data stories that will actually drive change. They just Storytelling isn't really about analytics or math or statistics. It's much more about visual design and taking your borrowing standard graphs that get produced in tools such as Excel and turning them into rich, compelling narratives that will resonate with an audience through intentional graphical design choices. The key I want you to take away is just because the data is accurate and factually presented. No longer enough to actually drive change in the organization, which is, after all, why we take decisions with data, because we want to be able to act upon those decisions. And if you're presenting lackluster graphs that don't resonate or inspire with anyone, then you're not really utilizing data to its full potential. And with some design choices, you can tell much more compelling stories. I'm sure your next question is, well, how do I do that? What does this actually mean? 3. Why Data Visualization is Important: So vitality Friedman, founder of Smashing Magazine and overall design lover, put it best in 2008 when he said, the main goal of data visualization is to communicate information clearly and effectively through graphical means. It doesn't mean that data visualization needs to look borrowing to be functional or extremely sophisticated to look beautiful, to convey ideas effectively, both static form and functionality needs to go hand-in-hand, providing insights into a rather sparse and complex dataset by communicating its key aspects in a more intuitive way, designers often fail to achieve a balance between form and function, either creating gorgeous data visualizations which fail to serve the main purpose to communicate information. Or they create standard data visualizations that fail to engage and compel. Friedman touched on something very important in that quote. And that is that data visualizations are effective. Data visualizations need to strike that balance between form and function. There's no point having a beautiful, well-designed visualization if it doesn't actually drive change in an organization. And just because you've taken data and presented it in formatively through data visualization, doesn't mean that it's actually going to inspire an audience unless you make deliberate design choices to achieve that goal, graphs can be accurate and truthful, but they're just numbers unless they're presented in the right way with the right context than the not actually going to drive change. And if your data analytics don't actually drive decision-making, then you're effectively just presenting numbers for the sake of presenting numbers, this is why most data initiatives fail because they have the correct data. They're just not presented in a way that's actually going to inspire people to make correct decisions with similarly, you can invest far too much time in the other side of things. You can take some data and make it beautiful, infographic out of it. That looks really visually compelling to an audience. But it doesn't actually say much with it. It's not a data story, it's just nicely polished data. The key to an effective data story is striking that right balance. The design needs to incorporate storytelling elements to inspire an audience, but the data also needs to be informative and factual. And as we progress through the different lessons, this is exactly what we're going to explore. We're going to look at graphs that are truthful but ineffective. And we're going to look at what design choices can be made to turn them into compelling data stories. Before we do that though, I wanted to share another quote from Edward Tufte and he wrote a book in 1983, colds. And I'm going to have to read this because there's no way I can remember it. It's called the Visual Display of Quantitative Information. Now, this defines what an effective data visualization actually should be. Now he says in the following passage, excellence in statistical graphics consists of complex ideas communicated with clarity, precision, and efficiency. Graphical displays to show the data and introduce the viewer to think about the substance rather than about the methodology, graphic design, all the technology of the production, It should avoid distorting what the data has to say. It should present many numbers in a small space and make large datasets coherent and encourage the eye to compare different pieces of information and data. They should also reveal the data at several levels of detail from a broad overview to find structure. And it should serve a reasonably clear purpose. Description, exploration, tabulation, or decoration, and be closely integrated with the statistical and verbal discrepancies of a dataset. So that was a lot of words, but it's important to get to the key message to Edward Tufte was saying that what he's saying is that an effective data visualization is one that presents the numbers truthfully, but it's also one that compels the audience by presenting information in a way that's actually going to tell a story. A data visualization isn't something that is just produced for the sake of having the data. And effective data visualization is one that drives decision-making without the proper consideration and design choices, then the data visualizations you create are going to fall shots of effective these days with tools such as Power, BI and Excel and Tableau and Qlik Sense and many, many other tools that we use in organizations. It's really easy to plug in some data and with a click of a button, you can produce graphs. But these graphs don't take into account any of the context of why you're doing the graph. It doesn't take into account any design choices. And therefore, being able to one-click create these graphs is actually not creating effective graphs. It's just producing more graphs. These graphs are certainly not going to resonate with an audience or be inspirational when it comes to driving decision-making. So join me now in the next lesson where we're going to grab some of our data and start visualizing it and discovering the tools and techniques we can use to turn our standard one-click graph into something inspirational and influential. But before you do that, I want you to think about a couple of recent data visualizations you've seen in your organization online, or you can hop onto Google and have a look for some data visualization. I don't want you to think about whether they would just presentations of data, as in they were just visual displays of data. Or whether they were actual effective data stories that contained design elements used to create a narrative. And when you've done that, I'd love for you to share that in the discussions down below. So we can see some examples of other data visualizations and we can pick apart what makes them effective and highlight if they told a story or if they were just presenting visual data. Once you've done that, join me in the next lesson, we'll start exploring visual design. 4. Gathering Your Data: Hello. In this video, I'm going to share with you where you can get some interesting datasets so you can start creating some wonderful data visualizations. Now, for this course, you can use any data set you want. It doesn't have to be one of these recommendations, but I'm going to recommend a couple of interesting, valuable websites you should know about anyway. The first one is Kaggle, which is right here. So Kaggle is an interesting website full of really good datasets. And the best part is there's a whole community of people who talk about data and what they did with it. Interesting if you're into that kind of stuff to find the datasets, you just go up here and the column two datasets. And then you can find lots of fascinating datasets. So organized by category or sometimes the best ones or just hit on the trending. E.g. here we have all the college majors and their graduates, all the New York City Air BnB reviews, let's say we want to explore this one. You just click on it. It tells you a bit about this date set where it came from and who scraped it. And then you can go down and see more about the data. So this is describing the columns that are found in that dataset. And you can see we have the date the review is posted and the contents of that review. And you can do lots of interesting things. Perhaps you could create a map and do a heat map of the best and worst locations in New York to rent an Airbnb. Once you're ready, you just click Download and it will download the dataset for you. So that's Kaggle. The next one is Makeover Monday. So just go to make over monday.code.uk. And you'll find this very interesting website that publishes a lot of fascinating datasets. And the best part is every week they have a short sort of competition where they post an interesting data set and then encourage the participants to visualize that data in interesting ways and then they review and discuss it. So to find those, you just go over to datasets and you'll see all the different years up here that it's been active. So it's been quite a number of years now. And for each week in that year, you will find a link to the data. So you just click this link here will take you to the date set. You can find more about the information that dataset. So this one is a top ten military budgets or the American business formations are e.g. FDA inspection information. Then you also have two links which I encourage you to check out. So one is called watch movies where the creative Makeover Monday, go see the dates set and explains his thought process as he's creating a visualization and understanding the data. So really fascinating to watch as well as the visit reviews. So this is where participants post the visualization of the data. So lots of different people posting their visualizations of the same dataset. And then the creators of Makeover Monday just go through and review it and talk about the good and bad aspects. So the one I'm using, you don't have to copy me, but if you want to, you can is in 2021 and it's the very first week called the great bicycle boom of 2020. And to download it, you just click on the data. This will bring you to a link to download the dataset, you might need to sign up and create an account on this website. But don't worry, I've had an account here for quite awhile and I've never received any spam emails or anything like that from it. So reliable if that's your concern. Anyway, you just go here and you'll just follow the instructions to download the dataset. And it will give you a nice CVS file and you can start working on it. So those are a couple of great websites you can find data for this course. So what you should do now is find it interesting dataset. Have a little play with it and look in the dataset, make sure you understand it, and even go ahead and create your first visualization if you want. Once you've done that, join me in the next video. 5. An Effective Data Story: Hi there. So we've seen why data is so important to organizations today. And we've also explored why creating a strong impactful visualization is the key to telling an effective data story, which is what drives change in the business. You will also be aware that over the last few years the increased importance in data and therefore creation of data visualizations. Unfortunately, most of the visualizations being creative are pretty ineffective actually conveying a story. In fact, there's probably more in effective graphs being created than there are effective ones. And this circles back to something we touched on earlier, that most data initiatives fail not because they don't have the right data. It's because it's not presented in a compelling way. If you want to be able to create effective data-driven stories, then you're going to have to be able to identify what makes these graphs so ineffective and what are the key elements of an effective graph? And that's what we're going to learn in this lesson. In this lesson, we're going to explore the key elements that distinct between a visually appealing looking graph and one that's actually effectively telling you a data-driven story. We're going to review some graphs in the context of Tufte's principles and identify the aspects that make the most effective and the aspects that should be changed to make ineffective graphs more compelling and visual. So let's start by looking at a graph that's pretty ineffective or learning to identify the key components that should be changed to make this graph more effective. And let me warn you right here. Once you start being able to identify the ineffectiveness of certain graphs, then you're going to start seeing them everywhere and it's going to start driving you crazy. I know. I go around all the time and ethnography, IC, I can't help but look at it in the context of these principles and start identifying what makes them so ineffective. So let's jump into it by reviewing our first graph. So let's take a look at our first graph now. You can pause the video if you need to, but let's look at this graph here. And I want you to just reflect on it for a moment and ask yourself, what is the message intended by the author of this graph? In other words, what's the story of this data as you're looking at it? Consider the principles that Tufte gave us earlier. Ask yourself questions such as, is this graph coherent? Does it encourage you to compare different pieces of data? Does it show you the data at different levels of detail? Importantly, does this graph serve a reasonably clear purpose? Are you supposed to look at a trend? Is it something you're comparing? Are you looking at a positive or a negative message? Which we can really sum up the most key principles of effective data visualization as is it accurate and not misleading? And does it tell you a story? Or is it just data being presented visually with those principles in mind? Look at this graph. You have to decipher a message or is there a clear upfront message being told to you? So let's break this graph down. Clearly that isn't really a message here. At least there isn't one that stands out or jumps out at you and tells you exactly what this message is. It's something we have to dig around for. Now. That could be because the graph is quite cluttered and messy and there's a lot of different visual elements on it. There's lots of labels, there's lots of bars, There's lots of graphic elements such as grid lines, and these serve to distract you from the core message. So try and look past those distracting elements and hunt down. And I want you to look at this graph and decide for yourself what the story is here. So we're obviously looking at a sales graph over a period of a few different months. I want you to tell me what the message for the audience from this graph actually is. So what did you come out with a message? For me, there's actually several different potential messages in this data. And this is critical because no one message jumps out. There isn't an upfront story being told. We just presented data visually and it's up to the audience to interpret what that message actually is. In this case, the story could have been that Q4 was a relatively steady sales period, or perhaps the author intended to communicate that Q1 was a very varying sales period. However, both of these are ultimately just guesses. They're prone to bias from the audience and the context that audience member has. This is just something that I've looked at and maybe you had entirely different stories that you picked out of the graph. And that's totally valid because the author of this graph didn't present a visual story to the audience. They just simply visualize the data they had as opposed to presenting a story and the data is used as evidence to that message. In this example of a data visualization, the audience is left to decipher their own message, which is definitely not what you want to happen when you're trying to communicate visually with data. Earlier we discussed vitality Friedman and his belief that an effective data communication is one that inherently communicates a message first and foremost before graphical design is incorporated into it. So in this example, the data presented is in fact accurate. This is there a sales data for this organization? However, because it wasn't presented in a way that tells a story, then it can't be considered an effective communication even though the graph is relatively tidy and well-presented. It isn't a piece of communication because it doesn't actually tell us any message. So next, I want you to review this graph and you can pause the screen if you need to review it for a moment. In this graph, I've taken the exact same data and I've cleared away all the distracting elements. So it's a much more clean all minimalist graph. However, it is the exact same data. So this graph is more organized and more clutter-free. But does that make it more effective at telling us a story? Well, the answer to that is no, not really, just because of graph can be more well represented on a cleaner, more minimalist, less cluttered graph. Doesn't mean it becomes an effective piece of communication. Which goes back to one of the principles vitality told us earlier. Just because the graph is well-presented, it can still be devoid of meaning. You don't want to focus too much on creating visually appealing elements that ultimately don't tell us a message. And that's what this graph represents. It's cleaner and more well-presented. However, that doesn't inherently make it a more effective graph. However, we can introduce some visual design and visual perception elements to bring the story out of this graph and make it more visual and compelling fall and audience. Now, take a look at this graph where I've made some intentional design choices that show us a more compelling story. Once again, using the exact same data, this visualization is much more impactful, persuasive, and poignant for an audience member. The key message I want you to take away from this little exercise is to reflect upon what Vitale Friedman told us earlier. Designers often fail to achieve a balance between form and function, creating gorgeous data visualizations which failed to serve their main purpose to communicate. So we can create a contrast between the two graphs we've seen. So both of these graphs use the same data. One of them tells a story, whereas the other one is just presented. Well, it's visually appealing, but it fails that main purpose to communicate information. Your data visualizations need to clearly and coherently communicate and effective message for your audience. These days, data is so important in organizations and with the number of people fluent in the language of data, than simply creating a graph that is visually clean and appealing is no longer good enough. You need to create a graph that intentionally uses design elements to effectively communicates a clear and impactful message. So now let's look at this in the context of the graph that I'm creating. So this is just the default graph that was created by simply inputting the data into a visualization tool, unclicking graph create, the One Button graph creation tool. Hopefully now you can see that this isn't very effective. However, let's introduce some design elements to it. So what I'm gonna do is I'm going to clean this graph up and make it more visually appealing. Let's not just use the default colors. Let's add a nice theme to this and present it in a visually interesting way. So here you can see the same data but presented more cleanly. Hopefully, now you can identify that this doesn't tell us a story. It might be visually appealing to look at. However, as Vitale Friedman says, it fails to achieve its core purpose, which is to communicate a story to the audience. So what I want you to do is create the default graph from the data that you've selected. And then once you've done that, you can post it in the discussion and project areas below. And we can sort of look at the starting point you're coming from. Once you've done that, join me in the next lesson and then we'll start exploring the visual design and visual perception techniques that we can introduce to our data visualizations to make them a compelling and impactful data story. 6. Visual Perception: So far we've explored the language of data and why it's critical in today's data-driven world. We've also had to look at effective and ineffective data visualizations and identified the key aspects that separates a well-formatted, presentable, good-looking data visualization with something that actually communicates on impactful message to an audience. In this lesson, we're going to explore visual perception. We're going to look at the five key aspects of visual design and visual perception and how they can be integrated into data visualizations to turn them from graphs to data stories. To understand how to incorporate these visual design aspects, we're going to explore some visual perception. We're going to borrow some visual perception theory and how they are applied in the real-world through visual design. And then we're going to see how we can apply that to data visualization. Creating effective data visualizations is a lot like cooking. Okay? Okay, Hear me out on this one, right. So anybody can sit down and have a meal and everyone will form an opinion on the meal. They just stay wherever they liked it or wherever they didn't like it. And they could even identify aspects that they did like and aspects that they didn't like. And you don't need a Food Science degree. In order to do that, you may even be able to describe in detail specifics about whether or not you liked the meal, such as the texture was enjoyable, or it had a richness or a savory aspect that you particularly enjoyed. However, if you're eating a steak, you might say that it tastes good. Whereas what you're actually technically doing is describing the male yard reaction of the amino acids and how that related to flavor. But you don't need to know that in order to say whether or not you enjoyed that stake. So the same can be said about data visualization. When you present a graph to an audience, then they can instinctually know whether or not it resonated with them. And they may even be able to describe certain aspects that were effective or not. They could say they liked the color theme or it was clean and presented. It was informative. They can't necessarily hone in on exactly what they're doing much in the same way. They don't know the technical, scientific aspects going into cooking a steak. The audience doesn't know the technical aspects going into the visual design, but they will instinctually know whether or not that graph was effective or it was just a data visualization. In much the same way that a good chef will have a functional knowledge of Food Science. A designer or someone who works with data should have a functional knowledge of visual design and how that can be applied to their graphs to make them resonates with an audience. So let's explore the broad concepts of visual perception and design and how they can be applied to data visualization. So firstly, what is visual perception? Visual perception is how we interpret the visible light in the spectrum reflecting off objects in our environment. In other words, it's how we see objects and interpret meaning from them. It's how we look at visuals. And it gets emotion and understanding and communicate information visually. And it doesn't just apply to the appreciation of art. It applies to everything. Visual perception is driving down the road and seeing the cars as objects in front of you and navigating through traffic. It's reading a book. It's everything to do with our visual system. It's looking at objects and understanding what that object is. That is visual perception. But it goes beyond just looking at something like adeno, this book. I can look at this book and I can say, this is a book I can see with my eyes. I know what this is. It goes beyond that visual perception is about visually communicating a message through colors and shapes with just an image. Not only can we communicate what's in the image, we can communicate a message and a story. I'm not just talking about animation or cartoons. I'm talking about a single image can communicate a story. And to demonstrate, I want you to take a look at this advertisement for Harley-Davidson motorcycles. This is a brilliant example of visual storytelling. We're looking at this image and without any context around it, it tells a complete story. You can see in the rear view mirror of this car a busy, hectic work environment. You can tell by the fact that it's in the rearview mirror, someone is driving away through iconography. We can understand that it's a rearview mirror of a Harley-Davidson motorcycle. And you can see these beautiful vistas someone's driving through. This tells a story to us. It says. Escape the busy, humdrum life of the real-world by getting on a Harley-Davidson and going on an adventure drive into the unknown with one of these bikes. And it's an effective marketing image. But what I want you to take away, not that you should be buying Harley-Davidson. That's not the message here. The message for us in our context is that there was a story told her entirely visually. There was no text, there was no pre context to what we're seeing. We looked at this image and we understood not just the objects contained within the image. We created a story around it. And that is what visual perception is. The way this image achieve that is by using some design techniques to draw our attention in the right places, to subdue our attention in the correct places to create a narrative entirely visually, they made a lot of deliberate and intentional design choices that independently a very simple, but collectively form and impactful message. And we can steal some of these design choices and apply them to our graphs to get the same result. In fact, these design choices are so impactful. If you understand how visual perception works, I, how our brains take the inputs sensory information we get visually and turn that into meaning and understanding. You can bypass a lot of these things. You can essentially hack someone's brain to communicate a message. Now, you probably sat there thinking, No, you can't really, can you? Well, let me show you something that will demonstrate this effect. So I want to share this short video with you and I'm not gonna give you any context going into it. I just want you to watch this short clip for me to test just how much attention the attention stealing design of the new Skoda phobia actually steals. We loved one parked on this ordinary road in West London. We wanted to see if it's sharp crystalline shapes, bold lines and lower wider profile would attract the desired level of attention. Will the 17 inch black alloy wheels stop passers-by in their tracks? Will the angular headlights attract the attention of other road users? When a crowd gathered to check out it's fresh, sporty look. Well, not quite. But did the attention stealing design distract you from noticing that the entire street has been changing right before your very eyes? Believers. Have another look. Did you spot the van changing to a taxi? How about the scooter changing to a pair of bicycles? Or the lady holding a pig, let alone the fact that the entire street is now completely different, I didn't think so. So there we have it. Proof that the new Skoda Fabio is truly attention stealing. Okay, so pretty powerful. We saw in that short video that despite active attention, we were looking directly at this video. We weren't distracted or at least I hope you weren't you weren't distracted. We were actively participating, watching this video. Yet we missed so much information. The key to take away is most of what we actually visually see, or in terms of what our brain sees, all of this visual information comes in. Almost all of it is discarded as waste. Our brain has a process to identify visual things and take away the meaning from it. And it's very good at ignoring pretty much everything else. And what makes this short clips or effective? Is it the terms hat our brains by giving us clear information to look at whilst distracting us from the rest of the information by the road. What you should know is this can be applied to data visualizations. So picture yourself, right now, you're in a meeting and you've got a graph on the screen behind you and you're presenting some information, your audience is actually sat in the room looking at the graph. Most of what they actually see is being ignored. What's actually happening is your audience is looking at the graph. Visual perception happens so they're taking away the key meaning or information from that graph. And then despite continuing looking at it, are being active participants in this meeting. They're ignoring all, pretty much all of the information actually being given to them. And what you want to do is make sure that the small piece of information they actually take away is the message you intend your audience to take away. What happens is we, as humans are actually very, very poor at multitasking. And we're not very good at concentration, particularly when that concentration is divided into multiple things. So when you're showing someone a graph there actually ignoring most of that information. But the good news is, what you can do is make intentional design choices to almost hack the brain and bypasses and deliver the message you want to communicate. There are entire fields of research around visual perception and understanding meaning from images. Child development, psychology, medicine, science, graphic design and advertising and cinema all use visual perception techniques to make sure what they're communicating visually is the message they intend their audience to take away. In the next few lessons, we're going to explore the five key themes of visual perception and how we can translate them into intentional design choices to create effective, data-driven stories. 7. Order: Hi there. In this lesson, we're going to explore the first visual perception technique, which is order. Now, order in visual perception is the principle that when we look at visual objects, we don't look at them as a whole. We look at them as individual components which we understand to complete a whole picture. So in order to understand order, you can think of reading. When we read a book, we're looking at individual letters which are formed into words, into sentences, into paragraphs, into pages, and into ideas. What we're not doing is opening a book and then looking at both pages simultaneously, looking at all the words on that page, and then understanding the whole story presented on that page. We're looking at individual components that come together to form ideas. So the Visual Perception Theory of order is very similar in that we're not looking at the whole, we're looking at individual components that come together to form an understanding of what it is. We're looking at the speed of which our brain does. This is measured in microseconds. It happens very, very quickly when we look at something, our eyes scan across the entire image and then pick out the individual components and create meaning from what we're looking at. Now, the visual perception of order can come in two forms, structured and unstructured. Structured order is when there's a prescribed way in which we're supposed to look at something. Again, think of reading, if you are reading an English book, you start in the top left and scan across letter by letter from left to right. Then you drop down a line and you go back to the beginning and you read across horizontally. Does it prescribed structured order to the way we're supposed to look at that. However, there's a second farm which is unstructured order. This is where there isn't a particular rule of what we're supposed to look at first. And when presented with a unstructured order of objects, what our brain does is it highlights what is most eye-catching at the time. So whatever stands out to our brain, initially, that's the order it's going to look at. It's going to look at things, brightest color, or the biggest objects, or most in the foreground, it's going to capture things that stand out to it to understand what it is. If it needs more understanding from one object, it moves to the next object. And then if it's still needs to figure out what it's looking at, it moves to a third object and a falloff object and so on. Again, this all happens in microseconds, so we're not really consciously aware of what it is we're doing. But that's how it works. We look at something and whatever stands out to us the most is what we start using two as a foundation to build understanding of what we've visually looking at. Now, graphs and data visualizations, they come under the unstructured category because there's no particular order of what you're supposed to look up. If you picture a data visualization, it's broken into many components. You have the bars, lines, gridlines, labels, axis labels, titles, call-outs. You have a lot of different components that come together to form a data visualization. And what happens is our brain looks at this data visualization of all the individual components together to form idea of what the message of that data visualization is. Now, as we saw in the previous lesson, our brain is really, really good at ignoring almost all visual information. So what actually happens in a microsecond is your audience will look at a graph and they'd very, very quickly come to an understanding of what they think the message of that graph is. And then despite actively participating and continuing to look at that graph, that understanding isn't really changing or evolving. So to demonstrate, I'm going to show you a quick graph on a screen. So take a look at this graph. What I want you to do is try and be consciously aware of what I was doing. What you're actually looking at in this graph. I'm trying to remember the order you looked at it. This is the order that I looked at. The graph in yours will probably be a little different. This is because this graph is unstructured. So there's two key takeaways here. Firstly, everybody's order is going to be a little bit different because it depends on what stands out to that person first. And what's happens is our brain will look at four or five objects and then decide what the message of this graph is. The second key takeaway is that you can influence that order through the use of visual design. When you're creating data visualizations, you need to be aware of the order in which your audience is going to view the graph. You then need to make intentional design choices. Lead the audience's eyes around the graph so that they come to an easy understanding of the intended message of that data visualization. And this is important because after your presentation, you've presented the graphs of the audience and they leave the meeting room the next day or the next week when they're recalling this meeting, then not actually picturing in their mind the entire data visualization and then analyzing it once again to understand it, the only thing that goes into a member about it is the key message that go into a member that one object was much larger than the other, or there was a large comparison, or there was a trend over time. Whatever the intended message of the graph is, is the only piece of information they're going to remember later on. If you aren't controlling the order of the graph, the message they take away may not be the one you intended them to take away. As we saw previously with the sales example, there were actually many different messages someone could have taken away from this graph. That was because design choices weren't made to influence the order in which people view the components of the graph to come to an understanding when designing visualizations, you can make intentional design choices to make it easier for people to understand. Also to convey the correct message when it comes to understanding how to guide someone's eyes around the graph. There's a few things that stand out. Firstly, did you notice how when you looked at your graph, you didn't read the title first. In fact, it was probably one of the last things you looked at and the same in the order of what I locked out and when I presented it, I didn't look at the title first. And that is because almost no one loves at the title first, It's not the most visually striking elements in a data visualization. It's usually the bright colors that form the actual data points, the bars and the lines. Those are more visually compelling because it's unstructured order. That's what we look at first. It's usually then afterwards we turn to the title amongst people get titles wrong in graphs, the title they use to describe the graph. However, that's only effective if you look at the title first, which most people don't. So you want to use the title to give a context clue about what the person is looking at. You should think about your data visualization in two terms. Firstly, you have the data, which is the evidence to your message, and then you have the visual elements such as the title and call-outs and labels which tell your story. The data itself is actually the evidence of that story. So when it comes to designing graphs to visually, which we will explore in some later lessons. You should think of them in those two terms, the message, which is all the other elements, and then the data itself is just the evidence to that message. Let's take a look at my example that I've created for this course. If we look at this data story, I wanted to highlight all the visual elements inside this graph. I have these call-out labels, I have a title, and I've made deliberate color choices to highlight aspects of the graph that I want the person to see. Now, whilst it's still subtle differences between the order different audience members will view this. Most people's viewing of this graph, the order will be the same pretty much across all the different people that view it. And that's because I've influenced the way people view this graph so that when they're presented with it, the story comes out in a way that I intend the message to be. You can see how I've used brighter and more muted colors and different places. I've used a dashed lines and solid lines to create separation. And I've added labels in the appropriate places, and I've paid attention to the brightness and size and position of those labels to influence the story. When you look at a graph like this, you may notice initially this large texts that's contrasted with the background so it stands out the most. So what you're looking at is a peak in the data and then you have this label that tells you a little bit about it. Then as your eyes move and follow the trend lines, there's labels in the appropriate places. And then you may notice the title and contextual story written there, which gives you more information about the story. As you recall, later on tomorrow or the next day or in a week's time after you finish this course. If you were to remember this data visualization, you're not going to remember how many people went on the trail or you're not going to remember the exact height of the peak or the, the dip in the peak, or how long the period of data was. What you actually go into. Remember, is the key message that COVID-19 caused a spurt in healthy habits by the increased number of people going on these trials. And that was the message I crafted intentionally so that you can do that. So let's look at another example. Here I have one basic standard graph and this is the graph that would be produced by any data visualization tool. It's the default graph. And the issue of the default graphs are they don't pay particular attention to the order because that tools, they don't have the necessary contexts to craft the message. So I'm going to show you a couple of different versions of this graph. And all I've done is made some basic intentional design choices to it. And you can see how each graph looks very different and presents a bit of a different story to the audience member. This is the same data presented in the same way. All I've done is made a few subtle changes, such as changing the colors or the brightness or the contrast to really enhance the story within each one. And the same graph now tells two different things. So depending on the audience member, if we presented them one graph, like graph a, they'd come away with a different message the next day than the same graph being presented to the second set of audience. So graph B being presented to them, they'd come away with an entirely different message. And we've done this just through subtle design choices. That is the power of visual perception of order. If we go back to our Harley-Davidson example, they've made some deliberate design choices here as well. They've used brighter contrasting colors to highlight the background of the office work, which is what you see first, you can bet they did this intentionally. And you can see how more muted background of the sunset and the person driving away into it is more muted to bring it into the background. So the older they want you to look at this in is to notice the hectic lifestyle, then to notice the background and the fact that it's in a rearview mirror and think, oh, someone's driving away and then you notice the Harley-Davidson logo and think, Harley-Davidson, what you get out of this is the intentional message that Harley-Davidson allows you to escape this lifestyle and it offers you this kind of escape from reality, if you will. These are intentional design choices following the principle of Alda to make you see things specific order to craft a particular narrative that they want to say. And that's why water is so important in data visualization. So that was just the first principle. And there's a lot going on, a lot of the other principles we're going to explore. Ways you can influence older and ways you can accentuate certain parts of your data visualization are ways you can downplay certain aspects of your data visualization. So join me in the next few lessons and we'll explore more visual perception concepts. 8. Hierarchy: In this lesson, we're going to explore the visual perception concept of hierarchy. Hierarchy refers to the way that everything our brain sees is in terms of a foreground and a background. So everything we see, our brain categorizes into these two buckets. And anything in the foreground is what our brain focuses on and pays attention to. Whereas everything in the background is largely ignored by our brain. To demonstrate this concept, let me show you this image now. This is something you may be familiar with. Is it to people facing each other? Or is it a candlestick? As you focus on the image and start looking around it, it switches between the two. Now this is an optical illusion. And the way the optical illusion works is by hacking our brain and exploiting this concept of foreground and background. So what it does is it uses different shapes with contrasting colors and influences the space the images on to make it unclear to our brains what exactly the foreground is and what the background is. As the brain switches between the two. That's when it thinks it's looking at two people are facing each other, or it's looking at a candlestick. And that's sort of the way this image works. It makes it unclear where the foreground and background elements are. If you are particularly perceptive, you can almost feel it inside your head. Switching between what is the foreground and what is the background. For me, at least there's a definite sense of the image are shifting and that is the brain deciding to put something in the foreground and therefore focus on it, or put those elements in the background and instead focus on the other elements in terms of data visualization, this concept is particularly important because as your audience views your graph, that going to do so in terms of foreground and background elements and whatever's in the background is therefore going to have less attention paid to it. And sometimes it's the background elements that actually enhance the story and the readability of the graph that you want to communicate. And if you rely on default graphs with a few basic changes being made to them, such as the type of graphs produced out of Tableau or Power BI, or Excel, then the tool itself lacks the necessary context to understand what elements are important to the story. So instead, it just defaults to selecting some elements that are important and placing them in the foreground. Whereas These may not align to the intended message that you want to communicate. However, with some intentional design choices, you can influence the foreground and background elements of your data visualization. This will influence an audience's understanding of your graph and communicate the intended narrative. So let me demonstrate with an example how to apply this concept to data visualizations. So here we can see three different data visualizations that all the same data, basically the same graph. The only thing I've done in each one is change some of the foreground and background elements by using design choices such as brighter, bolder, larger colors to put things in the foreground. Muted, duller colors to push things into the background. And just with a few simple changes, we can see three very different graphs. When you look at these graphs, think about how you're visually perceiving them. You're not looking at them and going well, these are three identical graphs with minimal, barely noticeable changes. It's like the three different messages being told to you. The changes I've made were very intentional, deliberate design choices, but they're not particularly complex ones. But they do illustrate that you can change the audience's understanding of a graph without them even being aware of it. As you look at these graphs, I'm intentionally manipulating your perception of those graphs. And if we think back to the previous lesson, the order in which you view the objects, which in turn influences the understanding and feeling of that graph. As a little tip, you should remember about this is when I'm designing data visualizations, I always start in terms of brighter and muted elements. I think about which elements I want to be in the foreground and which ones to go in the background, then I just color them brighter or muted colors accordingly later on, once the composition of the graph is right, and I'm pleased with it. That's when I start introducing design and visual flat to the graphs, the most effective data visualizations, unjust ones that highlight some important elements of a graph and mute some of the background elements. They intentionally use design choices to influence the order in which you view the graph and therefore the understanding or the intended message of the graph. And the concept of hierarchy is basically that it's focusing attention on the narrative elements of a graph and doling the parts that don't really contribute to the narrative fully. And it does this to intentionally create a story out of the data visualization when you create graphs out of the basic tools without making these intentional design choices. Then you end up with graphs that aren't really effective in utilizing the visual perception concept of hierarchy. To demonstrate, let's take a look at an example, and this was produced as the default graph in Microsoft Excel. You can see now that we understand a little bit about hierarchy. How it's placing certain elements in the foreground and which ones it's placing in the background. We can see from this graph that it's got the order of those objects all sort of muddled up. E.g. this graph has access elements in the foreground by using a brighter color, they stand out to the audience. And we don't really need access labels or lines because these elements themselves aren't contributing to the story or narrative we want to create with this graph. This is just the default produced by Excel. And because of its lack of understanding of context, it's southern just chooses which elements should be in the foreground and background. And as we saw earlier, these are important to crafting the narrative. So a graph like this doesn't really have a narrative. However, with a few mindful design choices, we can manipulate the narrative of this graph by highlighting the aspects that contribute to the story and muting the aspects that don't contribute to that story. An ineffective way to use this concept is by thinking that everything on the graph contributes to the story. If the axis labels that this is data and the graph itself is our data story. Therefore, they should be foreground elements, whereas this isn't correct. What you end up with is too much competing for someone's attention. You end up with a graph like this where everything is placed in the foreground and it's all competing for the audience's attention. As we would call familiar, when everything wants our attention, then nothing really wants our attention. We end up in that blindness state we saw from the video example. And we stopped taking things in the key to using this concept correctly. It's to highlight the important aspects to actually contribute to the narrative. So of course, your data elements are important, but which ones are particularly important to contribute to the narrative? E.g. it's always things like the largest difference, a trend line, a peak, or a trough, or two competitors against each other. Highlight these aspects of your graphs are not just all the data elements of the graph. Otherwise, you're not really using this concept effectively. The goal when it comes to designing for hierarchy is to highlight the narrative elements so that the graph becomes easier to understand and the story is more prominent. To demonstrate how we can do that, let's take a look at another example. So take a look at this graph here. This is the revenue generated for a small cafe. So you may be able to tell this is just the default graph produced by some data visualization software. And you can probably by now highlight some of the elements that don't align properly to the concept of hierarchy. So what we can do is we can tidy this up a bit by highlighting the elements that actually contribute to the narrative and meeting the ones that don't. By making these changes, we can influence the order in which you view this graphing. Your audio is probably very similar to this. You notice the line for the revenue, then the sharp drop, and then the competitor line. So I want you to think about how these design choices of influence the understanding of the narrative. If this graph, you would have thought something along the lines of a cafe was doing steadily growth in revenue. Then a new competitor opened, and this caused the cafe to take a large dip in sales. And that's because these key elements of the graph are highlighted with brighter, more bold colors. Whilst author elements of the graph where muted and gray and pushed into the background. And this is a powerful way to influence someone's understanding of the narrative of the graph. So much so in fact, that you probably miss the actual story of this data when you delve a little deeper, when you use the concept of hierarchy, then you could influence the message of the graph. When using the concept of hierarchy, it's really important to make mindful, deliberate design choices to influence for the right narrative. Because as in this example, these design choices made the audience take away a completely different narrative to the one that was intended. And this was achieved by mindlessly highlighting the data elements because they're the data elements. That's the story. Let's make them the foreground. And let's push everything else to the background. And in doing so, we actually missed the story of this graph. So the message of this graph is actually that the cafe has experienced continuous growth in revenue over the entire period of the graph, despite a competitor opening and a temporary dip in sales. The key message to take away for this audience is that the cafe is doing really well. It's experienced nonstop growth over the entire period. Because of our design choices, we buried that message and highlighted a completely different narrative of the graph. And that's really the key of the concept of hierarchy. It's about mindful, deliberate choices on the right elements to create a proper narrative. Not just highlighting aspects of the graph that semen parts and if they don't actually contribute to the story. Because as we saw, making these design choices really does influence the understanding and story of a graph. So how can we make these design choices to this graph to highlight the correct narrative? So let's take a look at the revised version of this graph. This one has used the same sort of design choices, is the exact same data. All we've done is apply those design choices differently to highlight a completely different narrative of this graph. So through the use of deliberate color changes, we've completely manipulated the narrative if this graph a few different times, which I hope demonstrates the power of the visual perception concept of hierarchy. So when it comes to designing your project, I want you to be very mindful about which elements actually contribute to the narrative of the graph. And those are the only ones that should appear in your foreground. So before we move on to the next lesson, I just want you to take a moment to reflect on your data visualization for the project and think about which elements are actually key to the correct narrative you want to say. And by highlighting the other parts, could that change your story in any way? 9. Clarity: Our brains are the apex of complexity. Trillions of neurons working together to make sense of the world form consciousness. And this allowed us as a species to reach the top of the ladder. However, when it comes to visual understanding, our brains, a surprisingly immature and uncomplicated. They can actually handle a rather low capacity of visual information. Our brains are constantly seeking to turn whatever they see into the simplest form to understand as possible because it simply can't handle too much. We touched on this in an earlier video. If you remember, the blindness to everything changing, That's because we can only concentrate on one thing at a time and we are almost blind to everything else. When our brains see something complicated, it seeks to simplify it as much as possible. This is the same phenomenon as when you look into the clouds, you start seeing shapes out of those clouds. This is because our brain wants to simplify everything It's sees as much as possible to reduce the amount of processing power it has to do in order to understand what it's looking at. Well, that's a gross oversimplification of the entire process. However, when it comes to data visualization, that's about as much as you need to know about the topic our brain seeks to simplify when it comes to data visualization. This turns up in a surprising number of places when it comes to data visualization, simpler truly is better. To demonstrate. Take a look at this graph we have here. This graph is formed with over 20 different lines over four different categories, resulting in 80 individual data points. However, when you look at it, you're not absorbing the information of 80 different data points. You're basically simplifying that information into a couple of key messages. Even if we wanted to try and understand the AT individual data points, we simply cannot handle that amount of capacity. So we narrow it down to whatever stands out to us, a couple of key themes or messages. This is simplifying those AT data points into their basic traits. What we understand from this graph is not 80 data points. It's a general downward trend and one of the lines is a bit of an outlier. And this is the visual perception rule of clarity, inaction. Even if we wanted to try and understand AT data points and the trends and the movement across all these categories. We're simply not going to be able to. So we just simplify and pick out some messages. And that's the same with old graphs. And it isn't just complicated graphs where we do this. It's even the most basic visual graphs. We seek to narrow down that visual information to a couple of different key messages from each graph. And it's not just the lower limit, there's also no upper limit to the amount of data or information we can present visually to someone. And we're instantly able to narrow it down and condense it into a couple of key messages. So let's kick it up a notch by looking at this next graph. This graph has over 15,000 data points. And again, we're not really picking out any individual data points on this. We're narrowing it down to a general trend or key message from this graph. What this graph shows is the number of entries in a listicle and the number of likes that listicle receives online. And if you're unfamiliar, a listicle is one of those articles like top ten, friends quotes or something like that. It's an article, but it's essentially a list, hence the name listicle. This graph shows 15,000 articles and plots the number of likes and the number of entries in each listicle. And despite it being incredibly complicated in terms of visual information being presented to us, I 15,000 plus data points. We can actually simplify it really, really easily and almost instantaneously, what we see is the general trend. The more entries in the listicle, the more light to get some Facebook. And that's quite impressive. If you take a moment to reflect on that. We instantly took 15,000 data points and came out with a couple of outcomes are key messages from that data. We did it relatively quickly. We didn't have to look at this graph for particularly long. And when you break it down, what's actually happening is we're looking for context clues. We're looking at all that data presented. We're looking at a couple of the title and information about what's plotted on this graph. And then we solve, generally see a trend out of it. And that's really what happens across all data visualization. We're not picking out every individual component and understanding the data. We're just looking at it and understanding a key message from it. This graph demonstrates that quite nicely. There's a lot of information, but really there's only one piece of information that we're taking away and that is a trend. However, that isn't to say that you're free to just present as much information as possible to the audience. Know, he should not be bombarding an audience with data and information and hoping to fall back on the idea that, well, the audience will pick a message out of that. You should definitely be presenting information to guide them to that message. But you shouldn't think of your graphs. And we've talked about this before, but you shouldn't think about your graphs as the data itself. You're plotting the data and the message will come. You need to be thinking about them as, here's the message you want to present. The data is there as the evidence to that message. To demonstrate that we can't just present lots of information at the audience and hope they'll understand it. Take a look at this new graph. This is the same data we saw just before this other graph. And now it's incredibly difficult to understand. This is pole design choices have gotten into that. There's now so many categories per bar, the colors don't make sense. The quintessential visual display of this data is not conducive to being able to be understood. And this is an example where you can't just throw information at an audience that has to be some consideration of the design. In the previous example, we saw 15,000 data points and could easily pick out some information. This one, we're only seeing about 80 data points and it makes absolutely no sense to us anymore. So whilst the principle of clarity seeks to understand the complicated, there are limits you have to design in a way that makes it easy for the audience to understand by picking the right graph for the right data. Helping the audience along through intentional design choices to interpret the message you intend in your graphs. So when it comes to visual clarity, There's a couple of key messages you need to understand about this principle. Firstly, it's that we can take a lot of information and condense it down into its most simple form and take away some messages. And the way we do that is not by looking at the whole graph, understanding every single data point and element on that graph, and then interpreting in thinking about it and coming out with a message. What actually happens when our audience looks at a graph is they scan over the graph, visually, pick out some context clues, and then conclude with some kind of outcome or key message. The second principle to understand about clarity is that we can help them along with that message and understanding through visual design. And it's the message that people understand when they finish your presentation on meeting. Again, when you show someone a data visualization, they will take away the message and that's what they remember the next day, the next week, the next month, they don't remember the individual data points. What they remember is the key message, e.g. the listicle example. Again, when you finish this course, you're not going to remember 15,000 individual data points go into member is a general trend of listicle entries to Facebook likes. You will not remember the data points. And that's really what clarity is all about. It's about intentional design choices that allow the audience to condense the data to form the narrative that you intend to tell. Clarity is also inherently linked to the principle of order. Because through both of these principles working together, your goal is to lead the audience through the graph to the right elements to draw the right conclusions. And you're there to support that journey as opposed to trying to hinder that journey and make it more complicated for people to understand. So if we once again turn to the cycling example, let's see if we can pick apart the elements that support clarity and order. Now, you can see in this graph, I've used highlighted information almost in the center of the graph to draw the audience's attention to it. This is a deliberate design choice. I intentionally made this bold contrasting font with this sort of br element and make it stand out. And I intentionally designed the title to be a less standout font than the rest of it. So the first thing the audience sees is this elements at the graph. And what they see is two elements that they're comparing and one being much larger than the other. Then as we said that go into sort of look around for some context clues. And what they'll notice are these call-out boxes which give them more of the story. They'll notice the title, which tells them what this data visualization is all about. And essentially the key message I'm trying to communicate. What you should notice is the actual graphical data display elements i, these lines are not really inherent or apparent to this message. The narrative is there. These elements take second stage and they're sort of at the back of all this information. So the story takes center stage. The actual graphical elements are used as the evidence to that message. So I'm seeing these elements are your message. That's why they're bold. They stand out that the most forefront elements of this graph. And then the actual graph part itself is just the evidence to say that is why I am communicating this message. That is order and clarity, working together through design choices to complete a narrative. As an audience member. Once you've finished this course and you thought back on this graph, what you'll remember is not these actual data elements, it's these narrative elements. You'll remember the actual key message that COVID lockdown inspired some outdoor activity. You don't have to remember precisely how many people use the bike trail in the month of May, e.g. you'll remember the message or not the means of that message. So what you should do now is you don't have to physically design anything on the graph in your project. But you should think about what is the key message of your data? What is it you want to actually communicate? So have a little play around and analyze a bit of your data and come out with the key message. So you have the message up front. And then we'll go into the next chapters and talk about how you can specifically design for these choices. And how you can highlight certain elements and how you can downplay other elements. 10. Relationships: In this lesson, we're going to explore their visual perception concept of relationships. So our brains are hard-coded to try and seek understanding through everything we see. Whenever we look at something, we are trying to use relationships to the context of what we're looking at and through our own personal experience to try and understand what it is we're looking at and we're hard-coded to do this. There's no way around it. It's just how we function, everything we look at, we're trying to seek meaning in what it's actually trying to convey. In other words, we're not looking at just an inanimate object. We're forming a relationship with its context and our understanding. And we're forming almost a narrative or understanding from what we look at. Now to demonstrate, let's look once again at our listicle example. Reflect on this visualization for a moment. What is the message being conveyed here? What is our understanding through looking at this data? What is our key takeaway, our key outcome, our key message, the narrative, this data visualization. So I'm guessing that when you look to this, you took away that most listicles contained 10-20 items. And then once you understood this, you started looking for the y in the data. You started looking for the message. And this is just instinctual. This is what we do as humans. This is the quintessential visual understanding. You looked at this visualization and you started forming relationships with the context, the way it's presented, your understanding of elements within this graph, and you started putting 2.2 together to form a story. Again, this is instinctual, it's natural, it's just what we do. You didn't necessarily consciously do this. You just subconsciously made and message or narrative or understanding from this data. Most people will look at a graph like this and they'll understand this trend line. They'll sort of see this trend in the data that suggests that the more listicle entries you have, the more light you get on Facebook until you reach a certain point, then that trend starts reversing. Once you cross this threshold of number of listicle entries, then you start getting less likes on Facebook. So you started farming the story after this. Why is that the case? We look for the y in the data all the time? It's just natural. We looked at it and thought, well, probably because if a listicle has too few entries, then it's not really worth reading or sharing with people. Whereas if it has more entries, then we start seeing it as something worthwhile and sharing. And then if it has too many entries, then we probably don't finish the article and then bother sharing it because we never reached a satisfactory conclusion having read that listicle. So that is probably the story or narrative that you took away from this data. And it's a perfectly reasonable story. However, the data doesn't say that there's no message saying that's what the interpretation of this data should be. In fact, there's plenty of flaws with that interpretation. However, it's perfectly reasonable to come out with that narrative based entirely from the data presented to us. And that is because we always seek to find the story and that is the concept of relationships. In visual perception. We look at this data or anything really, and we form a narrative or story out of it. The way the other principles played within this data, the principle of hierarchy, of order of clarity. All of this came together to form a narrative. And just because the data doesn't actually tell us this narrative inherently, it's perfectly reasonable that your audience will create their narratives out of whatever you presented them. And sometimes even a conflicting narrative will conflict with their interpretation of their data. And they'll always go with their own interpretation. Never mind the message you present to them even subconsciously. And we'll explore that in a moment. So despite creating a story out of this graph, there's plenty of things that it doesn't actually tell us which invalidate the narrative. We came out with. Things such as how long have the article has been online. Mediums where they shared with how many active readers of this website where they're at certain times, where their internal policies that stopped. It's not unreasonable to think that Facebook got more popular over time. Maybe the listicles, they've got less shares, which is posted before Facebook reached peak popularity. There's plenty of unanswered questions. Isn't this data on all of these questions will invalidate the story we said. So the conclusion we drew from this data isn't actually truthful, or at least we can validate how truthful it is. However, that won't stop anyone from forming that narrative. When you present data to an audience and you don't provide the narrative for them, then they're going to make their own narrative. And that is the principle of relationships. And again, even if you presented a narrative that conflicts with the data you presented. They will not accept the message you presented them. In most cases, they will always default to what they see with their own eyes. So let's take a look at this in action. Here's another visualization. Just take a moment, pause the video if you need to understand what's going on here. So this shows the sales from a coffee shop over time. And what you'll pick out is, you'll see the sales line. Then you'll see this competitor line in there coinciding with a dip in sales and then a subsequent rise back in sales. Now, you're going to start forming a narrative out of everything you've seen here. What was your story you took from this? I encourage you to post it in the discussions below and share with others the key message you took away from this. So it would have been perfectly reasonable to interpret a message from this graph along the lines of sales, we're steadily growing in this coffee shop. A new competitor opened and they started taking away sales from this coffee shop. And after some time, customers returned to this coffee shop and sales picked up again. And again, this would be perfectly reasonable narrative to form from the data because of the way it was presented. That's not necessarily the correct interpretation of this graph. What if I said that the competitor isn't really much of an influence on the sales. In fact, the coffee shop went through a refurbishment period and they just had less available seating for people. So they had a period of time where they had less customers, and then the refurbishment ended and they picked up their customers again. Now, I bet that didn't enter your thoughts when you looked at this graph. And well, why would it, none of that information was presented. Therefore, how is the audience supposed to interpret that message? Again, they're going to only interpret the message based on the context they see in the graph. Now, despite me saying that you're still going to think the competitive must have had something to do with these sales. And it's because the graph being presented conflicts with a message. I'm telling you, it's hard to accept that. You'll go back to think, Well, it's still the competitive must have had something to do with the sales. It kind of just been the refurbishment period. There must have been some impact. And that was because we put elements on the graph that said a competitor opened. And you just instinctually have to form a narrative out of what you see in the context in which you're viewing the graph. So you might be familiar with the phrase, correlation does not mean causation. And it just says just because things correlate or seem to work together doesn't inherently mean that they do. This phrase exists to remind us that just because it's presented on the graph doesn't mean it necessarily played into the narrative of that graph. However, despite the existence of this phrase, It's instinctual that we form a narrative visually out of what we see and it's hard to overcome that hurdle. So you should never really present any conflicting elements in your graph that allow your audience to form their own narrative. Again, remember, I've said this a few times and this, if you take anything away from this course, it should be this message, your data, the elements of the data in your visualization or your evidence. The message should always take front-and-center stage. Again quickly looking at our bicycle example, all the design was made to stand out these narrative elements and the data itself is barely a part of it. I leave no room for the audience to conclude from their own narrative because I present the narrative to the audience. So presenting this graph through an audience and saying, Oh look, here's a competitive followed by a dip in sales and then almost pulling the wool from everyone's eyes are saying, hi, trip to you. It's nothing to do with the competitor. Obviously, we shouldn't be doing anything like that. But let's take a more realistic example of this principle in action. So here's the same graph again, and this time it's been formatted to be more conducive of the actual message. So this is a more realistic example. The manager of the store wants to know about the renovation period and how it's impacting overall customers to the coffee shop. And they're also interested in that coffee sales. Now, we haven't included a narrative in this message, but despite that, you're probably still connecting the dots on this. You're saying, well, the renovation period didn't seem to affect coffee sales. Therefore, the renovation period only affected seated customers and the coffee sales. It must be pretty much primarily takeaway coffee for this cafe. It doesn't actually say that on this graph. However, we started connecting the dots and just because these lines were presented together, we had to form a relationship between them and come up with some kind of narrative that fits these pieces together. Looking at my cycling example, we can see how this plays in. You want to make sure that the audience takes away the right message and you don't leave any room for them to make up their own. So you do this by highlighting the story elements and downplaying the data elements, but they should also be presented in the right way. At its core, my visualization is one of comparison. So I want to compare this year to last year. So instead of having it as one line showing data over a period of time, I took the two years and place them on top of each other to make that comparison easier. So there's no other room for people to come up with their own message about this visualization. So that's the principle of relationships. You don't have to act on anything in your data visualization for your project just yet. But remember, think about what your core message actually is and how you're going to bring that out to the audience. And that's what the principle of relationship is all about. At its car. You don't want to leave room for anyone to interpret their own message. Once again, you should always use the data as the evidence and your message is the primary parts of any data visualization. So think about your own data visualization of what the core message actually is. Then join me in the next video. 11. Convention: The final theme of visual perception that would go into explore is the Visual Perception Theory of convention. So to understand convention, it can be simply put as just the way we do things. There's not much more to convention than that's just the way it's done. However, it should be noted how it impacts data visualizations and visual perception. So Convention is really just about the way we all unwritten only agree to display certain things. Now, of course, that's a really broad definition. So let's narrow it down to how it impacts data visualizations in certain situations, in certain ways, we're just hardwired to look at things and expect them to be presented in a certain way to demonstrate, let me show you a picture of a map. So this right here is a picture of the map of the world. Now, let me ask you, is it wrong? Is there something wrong with this map? Most people would say, yes, there's definitely something wrong with this map. It's upside down. But when you think about it, there's no right way up. The new map should be. It's just that we've always been presented a map in this correct way. Therefore, anything that goes against that conflicts with our expectation. And this is the principle of convention. Every map we've ever seen has had Australia on the right, America on the left. It's been presented with north facing up. So when we're presented with something that conflicts with this accepted standard, then it's really hard for us to get over this hurdle. You see the visual perception concept of convention is almost like a barrier. If anything, conflicts with our accepted standard, then it's really hard for us to get over this hurdle, almost impossible. In fact, when we see a map that's upside down, we don't just think, oh, this is a map. It's just not the right way up. We just, it's almost like we reject it. We just can't use it. We have to turn it the other way round before we can start accepting that this is a map. When it is upside down, there's something wrong with it. Certain things I just hard-coded to us in data visualization, there's other elements such as green, meaning good or positive, or higher sales or better, trend lines. And red means negative, some things down. There's some negative aspects about this data, something wrong that needs to be highlighted to the audience. We just expect green to mean good and red to mean bad. If we were to flip this around, then it's really hard for us to get over. It presents a hurdle to our understanding and we just can't get over this barrier to absolutely understand the message. You should be aware of this because sometimes when you're trying to create graphs, maybe you're doing this as part of an organization and that color theme contains green and red. And I've been in this situation myself, where the theme of the organization was red. It was quite a strong element in this organization to use company colors. So people always present in these graphs that looked really negative and say, Oh, we're doing terrible in sales because it's all red, whereas they meant it to be a positive thing. Again, convention acts as a barrier. When you can flip to have Convention, you lose the understanding from the audience. So even if you tell them, Oh no, no, in this situation, red means good. They'll go. Okay. But they won't really interpret it and subconsciously, they'll always be thinking it's a bit negative. Even though you tell them the right story. It's the same with relationships. What you present is what delivers the message, what you say to the audience. If it conflicts with the message that you've taken away, then it doesn't really penetrate as far. You don't really get that message sinking in. And the same can be said with convention. If you conflict with convention, you're going to have a hard time delivering the narrative. So you should always go with convention in data visualization over them red or green. What are some other conventional elements you shouldn't conflict with? Whilst there may not be an accepted standard to what elements we look at in a graph or what elements should be in a graph. It is possible to put them rank. Let me show you the cycling example we've been working on. So I've made some edits to this example and now it's wrong, things that just out of place. And this distracts us from the message. So when we look at this graph, we can't get over the unconventional elements of it to start interpreting the narrative. The text is wrong. The title is just in the wrong place. The moms don't go from left to right. They go from right to left. And this just breaks our convention and we can no longer understand it as a trend, even though looking left to right or right to left shouldn't really make a difference. It really does. And convention is so strong that when you conflict with it, the audience can interpret the message from it. No matter what you say to them, they can't get over this lack of convention. So take a look at this other graph as an example. Here we're looking at a wholesaler that's selling products over four different stores. And they've used a line graph to show this information. Now whilst the data is correct, a line is typically going to be used to denote a relationship with time. However, in this case, no such relationship exists. We're comparing the sales across different stores. So a line graph conflicts with our convention. It should be a bar graph for comparison. However, the line graph muddies up the interpretation of this graph and makes it difficult to interpret the message easily. So things such as comparison should be done as bars over time as a line graph and percentage to hole should be used as a pie graph. These data conventions that should always be followed because when you break them, it leads to misunderstanding and harder to interpret narratives. You end up having people distracted by these elements as opposed to accepting the information they said, which if you remember, was one of the principles that tough laid out earlier. He said, and I'm paraphrasing here that the design of the graph shouldn't make people think about how that graph is produced. They should just be thinking about the message that it's trying to create. When a typical audience member views a graph like this one. If you remember, back to our sort of steak cooking example where I said that a person eating food can describe if it's good or bad and use certain aspects, but they're not necessarily aware of the scientific process of what went into that cooking. The same can be said with a data visualization like this, they wouldn't describe this graph as insightful or compelling or influential. They might just say it was okay. Just something didn't quite resonate with them. In this case, it's the lack of convention being followed. Even if your audience member can articulate exactly, they don't find compelling about the graph. Now, there's other ways you can break convention using graphs. Consider this graph from the same wholesaler. Now we've drilled down from the stores to the different products sales. And the colors indicate to the viewer that there's some relationship between the similarly colored and products for sale. However, no such relationship rarely exists. If you look closely, you shouldn't notice that things are muddled up. Why bananas and medium-sized t-shirts colored the same. This again implies relationship. However, the fact that there isn't a relationship breaks that convention. If we were to present the same data only this time we will group the objects logically, then it resonates with the audience a bit better because it follows convention as opposed to conflicts with it. So let's turn to our cycling example and identify the elements that conform to convention. Firstly, you'll notice that the time goes from left to right. This should always, always be followed. You'll also note that these two lines have visual differences, and that's because they indicate that they're measuring different things. Everything about the text elements on the graph all aligned to convention there in the typical places, a title up here is larger, it's bold, stands out more. It's placed at the top of the page because things at the top tend to be titles and important messaging, they're not tucked away down at the bottom somewhere else. So everything about this graph confirms a convention and this makes it much easier for the person to interpret the message of the graph and not be distracted by conflicting conventional elements. So that is the Visual Perception Theory of convention. Remember, it's always conform with convention and not conflict with convention. So join me in the next lesson where we'll learn to bring all of this together and go from raw data to present double influential messages that use all of the visual perception theory and the tools and techniques that you should be following in order to do that, look forward to seeing you there. 12. Bringing it all Together: So far we've explored how people go from visual images to messages and how all of these different visual perception concepts apply to the world of data visualization. So let's quickly recap some of the most important messages you should take away. And then we'll go into the methods and techniques you should actually use and how you go from raw data to impactful visual communication. Firstly, we explored the concept of order, which states that in order to visually interpret meaning from something, we don't look at it as a whole. We build it up over different parts that buildup to form a narrative. And food design techniques which we'll explore in a moment. You can influence that older. Secondly, was the principle of hierarchy, which states that everything we see is in terms of foreground and background. And anything in the foreground is therefore the focus of our attention. Thirdly, we explored clarity which states that everything we look out, we seek to simplify in order to understand. This is how you can throw a lot of data at someone 15,000 data points, e.g. and they can interpret that into a couple of key messages. However, that doesn't mean you can just throw data out to an audience and expect them to understand it. You still need to make design choices that will allow them to understand the key message. Next, we looked at relationships, which states that everything we look at, we seek to understand in context and form a narrative out of the elements. And this is particularly true when it comes to data visualization. Everything you present on a graph is going to be interpreted by your audience into some kind of narrative. And unless you present them the narrative, they're going to present their own. Or if you present a visualization and the narrative that conflicts with the narrative they come up with from that visualization, then they're not really going to take away your message as strongly. And finally, we looked at Convention, which acts as a barrier to understanding if things conflict with the accepted standard of the way things should be, then people aren't really going to interpret the message because they can't get over this barrier of the Convention being conflicted with. Always conform to convention. How exactly do you go from borrowing data to impactful data story? Well, firstly, you want to do the analysis of your data and decide the key message you want to present is, once you've reached that stage, there's two stages that you need to follow in order to turn that into an impactful, influential message. Step one begins with clearing away all the destructive elements on the visualization. The default graphs created by all these tools contain far too many elements. Every line, every label, every bar, FE grid line, every title, legend, all these elements seek to distract the audience. If you think back to all of these are all elements that people are going to look at to come up with that narrative. And unless every element conforms to the narrative you're trying to tell, all they do is act as distraction to the message. So step one, clear them all the way. So let's take a look at cycling example to learn a bit more about this. This is the default graph created and all these elements need to be gotten rid of. What I like to do is start with the absolute bare essentials. So just the data elements themselves. And then I'm going to introduce different elements one by one until they can be understood. So I just have two lines here. We need to know what these lines are. So let's bring back the month labels into this. Now that we have that, let's bring back in some elements that introduce what this actually is measuring over time. So I'm going to bring in the number of people using these outdoor activities into this graph. And we can basically stop here. This is all that's needed to interpret the message. People often ask me, should I have labels on my data? Should I have an axis? Should I use labels there? So the answer is very much up to you. You should never have both, but you should always have one. Which one goes with the message? The strongest is the one you should go for. Here, it's not so much the individual data points that are important. It's the trend over time. Therefore, I've gone with axis labels where if I really wanted to highlight a specific value, I'd probably go with data labels. The next thing you want to think about is the composition of your graph and it's caught. All data visualizations are about comparison. And you want to sort of identify what it is that you're comparing and is it a positive story or a negative story? Are you comparing two points in time? Are you comparing sales of two different stores or products? You have one sort of boil it down to what is the actual comparison. And that is what you want to highlight in your story. In this visualization, I'm comparing two different years. So I want to use the composition of the graph to really highlight that. It's better to always keep these lines on top of each other. So make sure the comparisons really easy. I should separate them visually somehow. One stands out against the other so that people can see there's a visual difference between them, which allows them to be compared because they see them as two different things. So I'm gonna make these few changes here. This is essentially our blank canvas. We've cleared away all the destructive elements. We've got the basic composition of our graph ready? Here's where we begin to highlight the elements that go with that story and downplay the elements that don't tell the story we want. So there's lots of different ways you can visually make something stand out. Here's all the different ways you can do that. You can circle things, you can make stuff bold, brighter, more standout colors. You can put a box around them. You can move them slightly so they appear different. There's lots of different ways. This, by the way, is a downloadable resource you can get in the projects and resources section so you can download it and take it away. So what we're gonna do is we're going to think about which elements we want to highlight. And then we're going to pick from this list, how are we going to hire them? Because not all of them will be applicable all of the time. So it's carb. This visualization is about comparing this peak month to the other peak moms across these two years. So that is the element I want to stand out the most in my visualization. So we can look at this page and think about which elements will actually make that stand out. The way I'm gonna make it stand out is I'm going to put a box around it and I'm going to use a much brighter color on that box. And this is sort of combining two elements together, which you're totally free to do. This sort of highlights this aspect of the graph and makes it stand out. So you may want to do more or less of this highlighting depending on your story. For me, it's all about this element and I'll section right here. So it's the only thing I need to highlight in this data. Now we want to introduce some narrative elements to it. And remember, people don't always look to the title first. We want to add some narrative call-outs to the graph that explain the story a bit better. So that's what I've done here. I've added these three narrative call-outs to the graph that explain the story. Now think about the older your audience is going to see them in which ones you want them to see first, because that's the one that should be the most highlighted. So you can see I've used some of the techniques of highlighting on the text itself. I've used bolder text to highlight and stand out why I've made this font much bigger than this other fun. So I can make this font elements stand out the most. I've placed it inside the container to make it stand out more. These will deliberate design choices that highlight the elements I wanted to highlight and downplay the ones that I don't. I've got a title here and this is very important to note. Your title should always be a few words on what your key message actually is. And you can also use, as you see here, I've got this smaller narrative element underneath it to sort of tell that story. Titles on the first thing people see, but they do look at them for some context on what they're seeing. So I've made sure to include it. So now I've got the core of my data visualization. We started by removing all of the distracting elements. Then we used techniques to highlight the aspects we want in the graph. We introduced our story elements. So now we're at my favorite stage which I call adding flap. This is where you introduce your own creativity and design elements to make it look however you want, as long as you don't disrupt the work we've done before, I'm pretty much anything goes. You can even incorporate some design elements to fervor influence the story. You'll see how I've used the background colors inside my narrative elements here, which make them stand out because they contrast against the background, so they stand out even more than that. Here's a side-by-side comparison of where we started and where we arrived with our visual design. I think it's pretty obvious which one stands out and tells an effective date To story. Hopefully you can highlight all the elements that go into this that make it impactful and turn it from a data visualization to a data story. So your project now is to implement some of these ideas into your own data story and then post them into the projects and resources section. I really look forward to seeing what you've done with your data stories and I'll be providing some feedback. Also in the project section, I encourage you to go there and look. I've posted a couple of my own examples of before and afters and the changes I made that made them an effective data story. So work on your own project and post it into the section. I really look forward to seeing you. In the next lesson. We'll be going through a couple of more examples of before and afters and highlighting the elements that make them effective. 13. Data Story Examples: So let's explore a couple of other data stories and hide the elements that make them an effective narrative, starting with this one right here. So just take a moment, pause if you need to sort of understand this little graph that's going on here. Let's see if we can identify which elements make it effective. Firstly, I want you to note how little that is to this graph has gone a lot there. And that's because we cleared away all those destructive elements and just reduced it to the bare minimum number of elements that tell the story. You see we have a trend line along the bottom that just tells you the year. We didn't need to drill into mumps. We didn't need to drill in any further than this few year period to provide the necessary context. I also want you to know how there isn't actually any data labels telling you the volume of searches that were done on each of these topics because there were relevant to the actual message, this visualization. All we have in this visualization or two elements, the two search terms that are being compared and they used bright standout colors to really highlight to the audience the comparison that should be made between them. Then the center stage, we have this bold callout that sort of right in the middle of the page that really draws the audience's eye to it and it gives them the necessary context to the message. We think about the audit, someone's going to look at this, they'll see this sort of thing first, Netflix releases worldwide than making a comparison between these two different line elements. So let's think about the order. Someone's going to view this in. When they first see this graph. This call-out grabs their attention because it's got this bright neon arrow. The text is highlighted. Why conflicting against the dark background and it's sort of on its own. It's placed away from other elements which you all concepts we looked at in that downloadable resource. All of these come together. There's little grab attention. It's going to be one of the first things people see. And it says Netflix releases worldwide. So the good thing about arrows, this is just a little bonus tip. His people instinctually follow where the arrow is pointing. So we've got this call-out in the center of the page. It points into the arrow, they see the downward trend. It says Netflix releases worldwide. Now, gonna be hunting for context clues about what does that mean? Now, the title comes in play, is this about free stuff for convenience, the rise of online streaming curves pirating. And then they'll read these labels which are intentionally down in the bottom left corner because I don't want them as noticeable as the callout. Then they understand that the green line is taunting search terms and the blue line is Netflix to stems. And then they can use all these to form that relationship, that narrative out of the elements that looking at the message they take away is Netflix released on search terms for tolerance plummeted. So that was a direct correlation between the two. So you can see how everything went into this narrative. And once again, the data is the evidence to the message. It's the narrative elements that take front-and-center stage in this visualization. What you should notice as you create some impactful data stories is there's not a lot of actual data elements needed to tell the story. You just wanted to tell the message and boil the elements down to the bare minimum to tell that message. And that's what we're looking at head. It's just two small lines. They don't even have labels. But the way they've been presented in the context of the narrative tells a pretty impactful story. And when you go away from this course, you'll remember the key message and not the data elements. So this is another example and this was just a bit of a fun one I put together. It's how much time was spent playing video games before and after the birth of my son. And let's think about what you want to compare. In this example. You want to compare two different points in time, the precision time and the posts on time. In order to draw that comparison, I've used two different colors. One of them stands out a lot more than the rest, although they're both important data elements. And then I push too far into the background, but it really does make the comparison a lot easier. And because of that, your eyes are drawn to this comparison and this label that stands out by being placed in its own space, which was an intentional design choice. It's in its own space, so it grabs your attention. The contrast between the bright green and the blue. Draw your attention to this space and then you see the label that son was bombed. Then you go looking for context clues and you see the title next, before and after baby. And then visualizing how much time I had video games before and after the baby. And you can see this large drop, this period where there wasn't much of a drop, wise paternity leave. So we had a lot of paternity leave to use up. This was just a fun visualization. I produced on Instagram and shed it over there. And you can follow me if you want to see more of these graphs. But it does highlight a lot of the concepts we've talked about and intentional design choices to tell the message again in future after he finished his costs. And you reflect upon a graph like this, thinking about the actual time spent on video games in, let's say, April 2018. That's not what you're thinking. You're thinking the core message I told you the narrative I presented to you, which was after some bing bomb lost a lot of free time. So you play video games. And that was the intended narrative from this graph. So now what you should do is work on your own graph and then post it in the projects and resources section where you'll also find a couple more examples. I've made, work on your own project and I encourage you to post it there as I'll be giving feedback. And I encourage you all to explore each of these graphs on share some ideas and feedback. There.