{"id":2425,"date":"2015-04-09T10:00:00","date_gmt":"2015-04-09T18:00:00","guid":{"rendered":"http:\/\/thinkapps.com\/blog\/?p=2425"},"modified":"2015-11-11T16:31:12","modified_gmt":"2015-11-12T00:31:12","slug":"user-testing-guide","status":"publish","type":"post","link":"https:\/\/thinkapps.com\/blog\/design\/user-testing-guide\/","title":{"rendered":"User Testing: Why It’s Important and How to Do It Right"},"content":{"rendered":"

An explanation of what user testing is, why it’s important, and some tips for how to do it right.<\/em><\/p>\n

Understanding how people use your product is very important if you want to achieve\u00a0high user satisfaction.<\/p>\n

A large part of the answer can only be uncovered by actually talking to\u00a0users and watching them interact with your product. This is called user testing<\/a>, and it\u2019s one of the simplest and most effective ways of increasing your growth and profitability.<\/p>\n

Why is user testing important?<\/h2>\n

From a\u00a0business perspective, it all comes down to increasing your\u00a0user base and profitability.<\/p>\n

You want people to use\u00a0your<\/em> product instead of your competitor’s. To achieve that, you need intelligence. You want to collect\u00a0as much relevant knowledge as you can get in order to make the product that people really want.<\/p>\n

Generally speaking, there are two parts to the relationship your customers have with your product:<\/p>\n

    \n
  1. buying or downloading it<\/li>\n
  2. using it<\/li>\n<\/ol>\n

    For now, we\u2019ll leave the question of how to\u00a0make sure people buy your product and turn to the other side of the coin – how to\u00a0make sure people love your product<\/a>.<\/p>\n

    People buy things\u00a0because they want to achieve something. There\u2019s always a goal – a reason behind the product<\/a>. No one buys a hammer just to own a hammer; they buy it to hammer the nails.<\/p>\n

    If your product helps people achieve their goal, they will love it, they will continue to use it, and they will recommend it to others. So, you need to know how they use it. And that\u2019s where user testing comes in.<\/p>\n

    User testing helps you understand how people use your product and for what purpose. Do they use it for the same purpose you had in mind when you designed it? Do they get maximum value out of it, and if not, why not? Even if they achieve their ultimate goal, are they happy or frustrated?<\/p>\n

    Find the answers to these questions, and you will certainly create\u00a0better products for\u00a0your customers.<\/p>\n

    We’re\u00a0convinced! What now?<\/h2>\n

    Doing simple tests with users is fine. However, if you really want to get the most out of the process, you should have professionals working alongside you.<\/p>\n

    Not only will you\u00a0get answers from the customers, but they will also help you see your own product from a completely different perspective. Needless to say, you will learn a lot in the process.<\/p>\n

    Again, there are ways to do really quick and simple tests, but we will focus here on the tests performed by a professional user research agency.<\/p>\n

    How do we prepare?<\/h2>\n

    Before the tests begin, you’ll need to lay\u00a0some groundwork. Preparation will be\u00a0as important as the execution.<\/p>\n

    The preparation basically consists of answering two questions:<\/p>\n

    1) Who do we invite to participate?<\/p>\n

    2) Which user flows do we test?<\/p>\n

    Who do we invite to participate?<\/h2>\n

    There is a technique in user experience (UX) design<\/a> called personas.<\/i><\/p>\n

    You may have watched the show The Simpsons<\/i>. There is an episode in which Homer is supposed to design a car. While he was\u00a0going through the\u00a0design process, he took everybody\u2019s suggestions and included them all in\u00a0the final product. The result (seen below) was a hideous car that no one wanted to buy.<\/p>\n

    The moral of the story is that when you design for everyone, you design for no one<\/strong>.<\/p>\n

    There\u2019s no such thing as an average user. Your products will be used predominantly by certain groups of people, and you need to get to know them<\/em>.<\/p>\n

    \"Homer<\/a><\/p>\n

    Image credit<\/a><\/em><\/p>\n

    What are\u00a0personas?<\/h3>\n

    A persona is a representation of one of those groups of users. It is designed as an actual person. It has a name, a photograph, a back story, life goals, etc.<\/p>\n

    The persona also has specific expectations for your product and certain characteristics relevant to that product\u00a0(such as maybe age, tech proficiency, situations in which the product is used, etc.). You can read more about personas here<\/a>.<\/p>\n

    How do you create a\u00a0persona?<\/h3>\n

    You will model personas by interviewing real<\/i> people. Ask relevant questions, but also let them speak freely about their wishes, expectations, disappointments, and previous experiences.<\/p>\n

    You will get valuable knowledge about your product even by interviewing just a few people. Keep in mind that the rule of thumb in the industry is\u00a0that\u00a021 interviews will result in an average of four\u00a0personas.<\/p>\n

    It is important to design the personas based on the actual interviews. It’s possible to do it\u00a0internally (without interviews), but those will be\u00a0based on your own assumptions, rather than the actual field data. They can still be used, but proceed with caution<\/a>.<\/p>\n

    Once you have your personas designed, you can identify and find the people you want to participate\u00a0in your user tests, which are those individuals with the same characteristics as the personas.<\/p>\n

    Which user flows do we test?<\/h2>\n

    The second part of the preparation is figuring out what to test. This is based on another UX technique called the\u00a0red routes<\/i>.<\/p>\n

    The transportation office in London, UK, became aware that there are certain roads that carry a large portion of traffic in London, and any obstructions in these roads will cause huge delays. For that reason, stopping is prohibited at any time. Read more<\/a> about the\u00a0red routes in London.<\/p>\n

    By the same analogy, as explained by\u00a0Dr. David Travis<\/a> in\u00a0his Udemy course on usability<\/a>, each website or app\u00a0has a limited set of functionalities that are the most important to the users. They must be completely unobstructed and free of distractions.<\/p>\n

    Anything that stands in the way of the flows that follow these \u201cred routes\u201d takes away the value of your entire product. Learn more<\/a> about red routes and usability.<\/p>\n

    How do you know what your product\u2019s red routes are? Well, you already know your users (personas), and you know what their goals are. Now, it\u2019s just a question of understanding how they achieve those goals by using your product.<\/p>\n

    Flows that fulfill the users\u2019 most important goals are the red routes. These are the\u00a0scenarios you should test to see if they are really free-flowing or obstructed.<\/p>\n

    When you’re driving a car, the only red routes are “Getting to a destination” and “Not dying.” Everything else is a distraction.<\/b><\/p>\n

    \"Figure<\/a><\/p>\n

    Image credit<\/em><\/a><\/p>\n

    Where should the tests be held?<\/h2>\n

    You have two main options: the users’ premises or your own.<\/p>\n

    The advantage of visiting the user is the fact that you get to see how the product is used in the actual environment and real-life situations. The downside is that it might be hard to fully capture everything that the user does, and recording these sessions might not be possible.<\/p>\n

    \"Figure<\/a><\/p>\n

    Image credit<\/em><\/a><\/p>\n

    If the user is coming to you, you can use screen recording software to capture the user\u2019s activity. Some of those products offer the functionality of recording the user\u2019s face as well, adding an another layer of information to your tests.<\/p>\n

    The downside of this approach is that the environment may be too \u201clab-like\u201d and produce outputs that would be rare in real-life situations.<\/p>\n

    \"Figure<\/a><\/p>\n

    Image credit<\/a><\/p>\n

    Both approaches have upsides and downsides, so it is important to weigh your options and come up with an optimal plan based on your particular product and primary objectives for the user testing.<\/p>\n

    Conducting a test session<\/h2>\n

    This is the most important part of the whole process\u00a0of user testing: the\u00a0moment when you actually test your product with users.<\/p>\n

    To achieve the highest quality feedback, it is important to prepare the user for what they are about to do. First, they have to know that the session is being recorded and they have to agree to that.<\/p>\n

    (Note: In some countries, you are even legally obliged to disclose that information. Make sure that you cover your legalities<\/a>.)<\/p>\n

    Then explain, in general terms, what the user is supposed to do. What do you test, what product, etc. Make it perfectly clear to the user that you are testing the product and not him<\/em>.<\/p>\n

    They cannot make a mistake while working with it.<\/strong> If anything goes wrong, it is the product\u2019s fault, not theirs.<\/p>\n

    They should be encouraged to speak while working. It is very important to know what they are thinking, what they are\u00a0expecting, is there something surprising, something weird, something good, etc.<\/p>\n

    The users should behave as if they are doing this on their own, without expecting any external help. It is important to see how (and if) they\u00a0can handle any obstacles by themselves.<\/p>\n

    Conducting test sessions with users requires an experienced moderator. Every person is different, they react differently to certain situations, and their motivations can vary greatly. The moderator must get the most feedback\u00a0without\u00a0influencing the outcome of the test.<\/p>\n

    Internally, we compare this to the observer effect<\/i> in quantum physics – an effect in which the state of the particle is changed simply by measuring its state. Testing with users can be quite similar, and guiding the session towards an expected outcome can be very tricky.<\/p>\n

    Gathering data and reaching\u00a0conclusions<\/h2>\n

    Once the testing is done, you will wind up with\u00a0a lot of data. Some of it may be structured (according to a predefined system), and the rest is more like free-form comments. Both are very valuable and will help you with making\u00a0design decisions.<\/p>\n

    You will first want to resolve those difficulties that a majority of the users had. Or if not that, than those issues that appeared only occasionally but created a big difficulty for the users. Other issues should be prioritized based on their frequency of occurrence and impact they have\u00a0on users achieving their\u00a0goals.<\/p>\n

    It\u2019s not just problems that you will uncover, though. You will also learn what your users really love about the product, and which features they find very easy to use. Those aspects should definitely be preserved and maybe even used as selling points down the road.<\/p>\n

    In any case, you will learn a lot about the product, the way users see it, and how well the relationship between the two works.<\/p>\n

    Conclusion<\/h2>\n

    User testing is a usability technique that helps you gain very valuable insights from your users regarding why and\u00a0how they use your products.<\/p>\n

    The very preparation for the tests will help you to better understand who your users are and what they are trying to achieve with your product.<\/p>\n

    The actual test session will, of course, provide you with the most valuable feedback as you will be in possession of real-life facts, based on observing actual people, and not just on your own assumptions about them.<\/p>\n

    From our experience with doing the tests so far, about half of the problems found were already known to the product designers (although often they did not understand\u00a0exactly why<\/em> the problem was occurring, so solving it would have been difficult), and the other half of the problems were completely new to them. The latter could only have been found through this technique.<\/p>\n

    We encourage you to employ user testing to increase user satisfaction and, of course, create a\u00a0positive impact on\u00a0your business\u2019s bottom line. Happy testing!<\/p>\n

    <\/p>\n

    \n
    \n