Mastering UX Design Basics: Usability Testing | Anny Liu | Skillshare

Mastering UX Design Basics: Usability Testing

Anny Liu

Play Speed
  • 0.5x
  • 1x (Normal)
  • 1.25x
  • 1.5x
  • 2x
8 Lessons (27m) View My Notes
    • 1. Usability Testing Intro

    • 2. Setting Up a Test

    • 3. Who Do I Test?

    • 4. Prepping the User

    • 5. Conducting the Test

    • 6. Calculating KPIs

    • 7. Data Mapping

    • 8. Usability Testing Outro


About This Class

This class goes over the basics you’ll need to know when Usability Testing. Usability Testing is a keystone research component of UX design-it informs designers of the reality of the experience users are facing when using their sites. Usability testing is most commonly used to test prototypes, but for this course we won’t be requiring you to create one-instead you can use any website you like for a usability test, unless you already have a prototype ready from a previous course. Here we will be learning how to Set Up a Usability Test, Create Task Scenarios, Deciding Who to Test, Prep the Users, Conduct the Test, KPIs, and Data Mapping.

In this class project, you will be conducting a real-life usability test to apply what you’ve learned.


1. Usability Testing Intro: Hi, My name is Anne, and today I'll be teaching this class from one of the most important skills in you Exercise . Usability testing. So, just a little bit about me. I've been living in Brooklyn, New York, for three years now, and I've been a UX designer for two of those years. As a UX designer, I've come to find that usability testing is one of the most important tools you can use to ground your product in reality. So usability testing basically is when you walk users from a guy to test on one of your products. Beautiful website, mobile app, etcetera. Ah, lot of times when companies put out products, they're answering what they think the user wants. But your job is a UX designer is to do the research to find out the things that user company and even you might not have realized where the actual solutions to let the user is looking for and then designed. Based on that, that's where usability testing comes in. So some of the key skills you can expect to learn from this class include how to set up a usability test, conduct it, mapped the data and execute your findings for that design, I would say this class condemn, flee, be for beginners People who are interested in you X but don't have much or any experience in the field and want to learn of the key aspects of the job. Usability testing. You will be required to download any tools you should in order to have, such as a computer or a WiFi connection. A great example of how usability testing is practiced in the real world is this article about Jared Spool called the $300 million Button. Here he talks about how he did usability testing for an e commerce company and found that they were asking users to register too early in the purchasing process. So after redesigning this flow and removing that registration, but it the company saw an increase in sales the next year up to $300 million. As you can see, the skills you'll learn in this class are really important to being a UX designer because it's about bridging the gap between vision and reality. Usability testing is what helps you form an accurate, realistic depiction of what the user is going through. It's about trying to help users with the problems that they're genuinely facing, not just the ones you think they are. So I really hope you'll get to learn this and more throughout this course. I look forward to seeing your projects in the project gallery once you're done. Also feel free to follow my profile. Believe review of the of the course, Thank you. 2. Setting Up a Test: So now it's time to start setting up figure test. One of the most important aspects of conducting a usability test is how you prepare for it before you even conduct the test, you should be prepared to look out for certain things such as KP eyes or behavioral. I'll get more, indicate the eyes in a minute. You should also prepare your test recording methods. What are you going to be recording this test on? And finally, what are your goals? What are you looking to get out of this test? So I mentioned KP eyes earlier. Thes air key performance indicators. Some examples include test success rate time on task, use of search versus navigation, user error rate and system usability scale. I'll go more into calculating these KP eyes later on, but for now I'm going to show you an example of the use of search versus navigation before you test. It's important to decide which KP eyes you'd like to test. This can be discussed for client or company beforehand, and usually you'll test KP eyes based on the goals or the hypothesis you formed about the test you're about to conduct. But I'll be showing you an example of search versus navigation soon. Another thing to look out for our behavioral. These are the things users do without telling you things that they themselves might not realize what they're doing. These are the actions that they're taking during the test, but not necessarily saying out loud. Here is the website of the company. You might have heard of Wal Mart. So you can see here they have a big search far and kind of a more hidden navigation. So here is a great way to test search first in navigation. Obviously, a user is probably more inclined to go to this big surge far rather than this navigation on the side. So you can tell Walmart is a company trying to promote its search more than its navigation . A site like Apple, however you might think, is doing the opposite where they have everything kind of laid out on the navigation and the search is a little more tertiary. If you're a website like Walmart, you might want to use a K P. I like surfers and navigation. Search versus navigation is a K B. I. Where you just compare that amount of times a user will. It's important, however, before you start testing to know if you're going to be looking for this KP I. But also, it's something that you should be watching out for when the user is going through the test . You might not notice that the users hovering over this menu, even though they're saying they would go to the search for that's an example of a behavioral. Sometimes the users action doesn't actually lineup of what they're doing. This is a website that's pretty famous for its bad ux and you I It's very distracting. It's hard to find things, but just as example of website that could really do it better. You acts were going to use lings cars dot com. Basically, if you were here to look for anything, let's say you're tryingto Lisa BMW. You not only don't have the option of searching for it, the navigation is pretty terrible. You want to look for a BMW specifically, you would either have toe look through this navigation on the left or through all these pictures and on the right. And look how long this list goes, Really? So basically the intention of these usability test is to kind of see the different KP eyes and behave rules you can get as well as going through the task scenarios that I'll go over later. So just prepare and keep in mind what your goals are, what kind of KP eyes you're looking for and what kind of behavioral zehr looking forward before you test, because it's much easier to understand the direction you want to go with your tests when you have these things clearly mapped out, also important to the side and prepare what method of recording and note taking you'll be using for your tests. For example, I like to record the chest with a Webcam, iPad or iPhone camera for both video and audio. If it's not an in person usability test, then I'll usually have a sky call of the user where I'll have them screen share as they go through the site, and I'll use quick time to record the test. But this is just what I use, so there are actually a lot of different tools out there that you can use to record audio or video, and it's up to you and your personal preference to use whichever one you're most comfortable with. I just happen to use quick time Skype, iPad and iPhone a lot of the time. But no matter what tools I'm using, I always like to take my notes afterwards during a test. It's tempting to take the notes starring because you'll save time instead of having to go back and write all the notes later. But personally, I think that there's some stuff you missed during the test. When you're busy taking notes, however, this is definitely just my personal preference, and you can decide the way to record and take notes based on what makes you comfortable or what your company your client prefers. 3. Who Do I Test?: So now that we're getting closer to conducting the usability test, there's still a question that we need to ask who exactly in my testing. Most websites and APS have ah set demographic that they're trying to appeal to. For example, you probably wouldn't testes site like WebMD with a bunch of fifth graders. So this is something to discuss of your client or company before you go into testing. And it's important to try to test within those parameters being age, gender, occupation, location, etcetera, psychographic sehr. Also important. And it's a way to ensure your tested users aren't just physically appropriate, but emotionally and mentally appropriate. Like I said before, usability testing is trying to best see the ways your users actually interact with your website and psycho. Graphics and demographics can have a big effect on that. So how do you ensure this before you test? There are several right ways to reach out to users. Some companies might have an email list that you can access, or you can send out something called a screener survey to their current users. I use Google forms, but you can use any sort of survey template. Just make sure to use one that allows you to make the answers private to everyone but you. You don't want a security issue later on. So on the screen of survey, I'll usually asked my demographic and psychographic questions to parse out the less targeted or relevant users. There are other sides you can use for screener surveys other than Google forms and other websites you can use to reach out to users. If your if your company your client doesn't provide, um when I schedule with users, I usually use something called Callin Lee. It allows them to book time with me. That's best for them. However, you can also schedule the test with them using Google Calendar etcetera. So just do whatever you think is most effective or that your company requires. Just make sure if you're testing someone in a different time zone to take that into account . The typical amount of time for usability tests in my experience could be from half on hour to an hour. But that also depends on the length and number of your task scenarios, and the resource is available to you from your company or client. So the number of users used to test per round is a really well argued topic in the U X industry lately, and it can depend on who you are and what your testing. Personally, I usually aim for 4 to 5 users per round, and for your project, I would recommend the same. But in reality, the number of users you should test can depend on the goals and the type of task scenarios you created. And that's completely fine, because again, it is a science, but not a perfect science. The number of rounds of tests can also vary with your goals, and your client and company and it's good toe, have multiple rounds of testing to compare KP eyes. But for our project, I'll be having you just do one round of testing. 4. Prepping the User: all right, so we're almost at conducting the test. One important aspect of testing is prepping the user to get ready to do a task scenario. Most users haven't gone through a task scenario or usability test before, so it's important to remind them when you're doing this test that you're not testing them, you're testing the Web site or the prototype. The task scenarios air simply a way for you to understand user better, and the chest will just give you the data that you need for your designs. It's a good idea to have a conversation with the user about any questions they might have beforehand as well, especially in this section, because you want the user to be as comfortable and relax is possible during this test. That way you get his natural result. It's possible explaining prototype versus existing site. I am guilty of forgetting this sometimes, but it actually is a very important part of propping the user. A lot of the times users come to your site knowing what website you're testing or having used it before. So when they when you show them a prototype that isn't finished, they might ask wire certain things left out. Or why are certain things not clickable. This will take them out of the test little bit, and you might lose time having the question asked, or even having to answer the question. So it's really a good idea. If you are testing a prototype toe, let them know that it's not the finished product. It's just the thing they'll be testing on. It is incredibly important that you have the user think out loud as they go through your website. There are a lot of times where you might be tempted to assume a user is doing something just because it looks like that's what they're doing. But when they tell you their motivations and what they're about to do out loud, you can get a lot of very important and very good dad out of that. So make sure that's something you prep the user on before you go into testing. I've always recorded every usability test I've ever done, and for you that may not be the case, depending on who you're working with. But if you are recording and sharing the usability test with the rest of your team, you have to ask the user beforehand for permission. Usability testing is basically the re still in the research portion of the UX design, so it is still kind of the science of everything. And like any science, you need to make sure that your subject is maintaining their objectivity as much as possible. Obviously, there are a lot of situations where the user has already used the website before and it has a certain emotional bias towards it. But even so, you let them know that whatever they say or do, you won't be offended by it. And again, that goes with explaining to the user how you are testing the prototype and not them. It's also important that you tell the user to speak as honestly as possible because you really want what the genuine reaction is to your website. 5. Conducting the Test: All right, now you're finally here. You're ready to actually start conducting the test now, there are a lot of things to keep in mind when you're conducting this test. Remember to try to let the users go through the test as naturally as possible. Don't ask them any leading questions. In turn, don't answer any leading questions so that you can maintain the objectivity. During the test, Users might be tempted to ask you questions like, What should I do here? What am I doing wrong? What am I doing right? It's usually best toe. Answer these kinds of questions of something like, What do you think you're doing? What do you think is going wrong? What do you think is going right? How do you feel about that? It might sound like you're copping out of answering the question, but really, it allows the user to think for themselves, and it gives you a better idea of what users might be thinking when they go through this website alone and at home. Or keep in mind your goals. A swell KP eyes, your hypothesis, all the things that you need to do all the information that you need to be getting out of this test are important for you to note down when you see them happening in front of you, make sure that your equipment is recording us. Well, I've had this happen to me where sometimes I think something is recording, and I realized I haven't actually pressed the record button. It's the worst thing to lose an entire usability test just because you didn't do your technical check correctly. And also make sure your testing environment is appropriate. You don't want a lot of background noise. It can distract your user and make the audio much harder to understand if you have to play back the recording later. Also, watch out for Behavioral Zell's. Like I said before, behavioral Zehr, the things users do without realising or without telling you about. So these are things that they don't exactly verbalize to you, and it's your job during the test to watch out for those not the users job to point that out. Let's go through the formatting of a usability test again. Now that you're conducting it, it's important that we go into this in more detail. I usually like to have my questions laid out in front of me when I'm conducting the test, but it's also a great idea to try to memorize them if you can. But just make sure you're hitting all the points that you need because, ah, usability test has to often be done in person, and you really only have one opportunity to ask your questions unless you do decide to send a follow up email. However, even if you do set a follow up email, it's never as good as having the questions asked in person while the test is still fresh in their mind. So like I said before, you want to start off with prepping the user, testing the prototype, telling them that they're not being tested, that the prototype is telling him to think out loud, asking if you can record etcetera. Then you go into your intro questions again. Here are some examples of intro questions that you could use demographic question, psychographic questions, behavioral and use context. Then, of course, you're gonna want to go through your task scenarios, so I usually do one or two tassan area per person per test. But depending on your company or your clients, you could go through much more or much less follow up. Questions are also really important for usability test. It allows you to cover stuff that you've missed during the actual test. Like I said before, you should have certain ones ready for each test, whether it's to ask about something that answers or KP I or something your company or client wants to know about. It's also a time toe. Ask about anything. You might have noticed the user do strangely or differently during the test, since it important to let the user go through the Tass scenarios as naturally as possible. This is the part where you get to ask the questions you held in during that time, and you'll usually find that you might have questions like, Why didn't the user press the big red button in front of them on the home page? Why they click the hamburger menu instead of using the search bar. All those burning questions are more appropriate for the end and allow you an opportunity to go over these points without disrupting the actual test. And when you're conducting the tests, just know overall, your goal is to get this user to be comfortable. Offer them coffee or tea. If you're doing it in person, try to have light conversation beforehand and just really kind of get to know your user without making them feel like they have to answer your test a certain way. A usability test is one of your greatest tools to see what is really happening when the user is behind the wheel, so next will be going over KP eyes. And then finally, I will be telling you how to math the data and then you'll be ready to start your project. 6. Calculating KPIs: So now we're at one of my favorite parts of usability testing. All right, so now we're at one of my favorite points and usability testing, which is calculating the KP eyes. KP eyes to me are kind of the reward at the end of the usability test, where you can kind of where you concede, where you can compare how your usability tests have improved, the numbers and usability, or where you can just kind of quantify all that, a data that you've been collecting. Comparing users to each other and comparing different rounds of users will often give you different results. But either way you're getting a lot of very good information. So these are some examples. And so these are some examples of some KP eyes you can expect to be testing past success rate, so task success rate is the average of all users on old tasks. So if you have cast success rates for three different tasks and the rates are 70 80% and 100% than the overall past, success rate is actually the average of that over three. That's because there were three test total and the individual success rates were these numbers at 70 80 and 100 so the total tax success rate is 83%. The next KP I you could test for is time on task. This one's a little bit more straightforward. It's just the average amount of time it takes users to complete a given task from the moment they start until they're done again. You have to calculate this based on individuals and then do it in the context of all the users together and get that average use of search for his navigation. We've gone over, and that's more straightforward as well, just measuring the number of users as a percentage that used search versus navigation. And you can decide which which one you look for. So if you're hoping to test and see how many people you search, you might end up of a number that says 60% of users you search to complete a task user error rate is the errors over opportunities for errors. I found that this one does confuse a lot of people, and for me as well I was. I had a hard time learning this one at first, but basically what you do is you go through a website, and basically what you do is you go through your tests and calculate how many opportunities there are for error. For example, if I'm trying to get my users to do a task scenario that requires the search far one of the potential errors is clicking on the hamburger menu instead of the search bar. Some other errors, however, even if however you are, However, this becomes difficult in that you have to find every potential error that your user can go through throughout your task scenario, and that becomes the number you compare the actual errors they do commit to. However, over time, I think with practice this KP, I gets easier. Miss KP. I gets easier to understand, and as you as you get more practice forming task scenarios, it will become much easier for you to kind of see where those opportunities for errors are . Just remember the top when it says the errors over opportunity for errors. The errors on the top are the errors that the action that the actually commit and the bottom number is the potential errors they could have committed. And the last one is system usability scale To be honest, I don't really use it in the same content system. Usability scale is kind of a longer wounded one. Ah, lot of these questions, I would say, can also be follow up questions as opposed to KP eyes. But if you scale a monarch, you scale. What I usually do is I ask users to answer C. Answer these questions on a scale of 1 to 5, as opposed to strongly agree or strongly disagree. To me, that just makes it easier for me to compare what each user says and kind of instead of having only a smaller. It kind of gives me more of a range of what users can potentially answer. And, of course, if I find that, of course, I don't use every single one of these. I don't use every single one of these questions or I don't ask every single one of these. Sometimes if I find a task doesn't require me Teoh. Sometimes I find if a project doesn't require me to ask for a certain KP I, it might not be necessary, and I'll leave that K B I out. Either way, though, when it comes to KP eyes it is really important to be organized. I usually like to do my KP I calculations in Google sheets. I find that that actually really helps in terms of. I find it's really great because you can calculate the numbers right in Google sheets, and you can also just share it with your You're gonna easily share it with your client right after and you can ease it. I find that using Google sheets allows you to share things with your client quickly as well . 7. Data Mapping: So now we're at the last point of content in our class. Data mapping data mapping is very important in how you present the information you've gained from usability testing to accompany or client. But there are several ways that you can choose to map the data after you do a usability test. Personally, I think I get a lot of great insight when I use something called a plus Delta map. So here's an example of something I like to use when I'm data mapping. This is called Real time board dot com. It's just a way for you to create a Plus Delta map and share it with your client easily. Here you can divvy things up into sections very easily. They even have a sticking up function, which, if I were working on a bigger team, usually you would actually do sticky notes in person. But I find that the method of working online and collaborating here is great as well. So as you can see, there are things on the left section under this label called Plus Plotless are the things that your user liked or found made the process easier for them during the usability test basically anything positive that happened during the test and implies a feature a section needs to be kept in Delta, on the other hand, are things that need to change things that may be made the users experience worse or slower things that were frustrating or things that you just noticed. Behavioral e The user wasn't able to get through. Of course, these are all just things that you have observed. These aren't necessarily going to be the changes that get made at the very end of the process. That that always is a discussion with your team and your client. Just because a bunch of users might want something doesn't necessarily mean that's the top priority off what the users want. There's something called on M V P, which is the minimum viable product, and a lot of companies are looking to do that when they're putting out something after a usability test. It doesn't mean that you're not listening to the users. It just means that you're prioritizing what the users needed the most. And that's why this chart will kind of help you visualize what exactly is a priority for you to be looking at more closely do what you should add and what you should change 8. Usability Testing Outro: So, as you can see, a lot of websites have benefited from good UX aside, and others could benefit from more of it. Through usability testing, you'll be helping both you, your user and your company get a more accurate representation of what the user needs when using your product. Never assume that you know everything about the user when researching her designing. And never underestimate the power of usability testing when creating a great you exercise. So I would like to thank you for taking this class, and I really look forward to seeing what you do for your projects just to explain the project a little bit more. I'd like you to take your favorite website and conduct a usability test on it. If you already have a prototype ready from another class, feel free to test on that instead. But just make sure you record the test on whatever device you'd like and take notes during or after making note of which KP eyes you'd like to take down beforehand. Once you're done, I'd like you to map the data you received, and if you still have time, you could test another competitor website or another similar website and compare the results of the 1st 1 you tested. See if you can find any similarities or differences and ask yourself, Does anything stand out to you? Is there one thing that one website does better than the other? How did the user seem to feel during the test and how their experience is different? Have what made the experience is different or similar. So feel free to answer these arm or in your documentation, and I really look forward to seeing what you guys come up with. Thanks again for taking this class.