Welcome to my talk Usability Testing, Short and simple. Um My name is Ines, we um I'm from Austria as you can probably hear not from Germany but from Austria. And I'm very happy that you have joined this session in the next 20 minutes.I want to give you a brief introduction into usability testing so that you can implement it in your um next project. That's OK. Before we start into the, the um testing itself, I want to give you a few um information about myself. Um After doing my A levels here in Vienna, I've studied business administration in first in Vienna and then at the University of Wales in Swansea and also during my art even then I was very interested in it. So I was always looking for positions where I could combine my knowledge in business and it, and I've done that since uh finishing my, my university degree. And about five years ago, I started as a senior consultant at a EO uh which is the German um it service provider. Currently, I mainly work as requirements engineer and also as a UX designer. And what I like really like about my role as a requirements engineer is that I can translate business needs into requirements that will be understood by um developers. And uh I enjoy being that translated between business and it as both areas.

Um Yeah, I, I like very much during um more in the last few years, I realized that a lot of it projects don't care enough about the user who has to use it systems in the end. Um So I decided that I wanted to learn more about how to create systems that put the ID D user in the middle. And that's why about 18 months ago, I started my part time studies of user experience management at the techni com here in Vienna. And currently I write my master thesis and hope to be finished in October. Uh Most people believe that UX design is mainly about um creating beautiful user interfaces or building prototypes, but that's just not true. It's really about understanding the user and you can't understand the user without testing and without doing usability testing. So why are people tell people that they should do usability testing? I always have to wait, the slight things change very slowly. And when I, I tell people that I want to do usability testing or that they should do, I get a lot of statements like um mm Yeah, we, we hope this will come up soon. Uh We need a laptop test testing takes forever. Where will we find all those users. Um And then also we need specialists to do usability testing. And um my favorite is uh we don't need to test, we know our users, which is just brilliant.

Um And I guess that some of you also believe that some of these statements are true and that's why you're here to find out if they are or if they are not. And um I want to show you that they are not OK before we go into testing, just a quick reminder what usability is and why we aren't testing the user experience. Um usability as you can see from the picture is just a part of user experience. Um It is really about um completing a specific task or the ability of a specific user to complete a specific task in a specific context. Use experience is a wider, it is about how a person perceives and um a system, what beliefs he has about the system or the or which um feelings, emotions also before during and after the use, the user experience is a much more emotional thing. Whereas usability is something very um strict. And that's why we can test usability using a usability test because we can test if a specific task can be solved by a specific user in a specific context. Uh When we want to evaluate user experience, then we need other methods such as interviews and questionnaires. Um And yes, you can combine those methods in one go.

So do a usability test first and then perhaps an interview afterwards to find more about the emotions of the user or the beliefs or what he thinks about the system or how he feels about the system. Now on to um testing itself, how do we do a usability test first? You have to decide what your goal is. So what do you want to find out with the test? It's really like um doing research. Like perhaps if, if you did it um graduate degree at university, you remember how research works? Um You have to find what your research, the research questions are or the hypothesis that what you want to test. For example, we could ask, does the user understand the navigation of our website? Or can you complete a specific task based on that goal? Um We will define them the use of tasks that the user has to complete. Yeah, that was 1 to 5. Um which task um which data should the user enter? When is the task complete? So you have to give the user really specific instructions, what you want him to do. And then you can observe the behavior of the user to find out if the usability is good or better. Very important before you actually do um testing, do a test run. Um We know communication is really difficult.

So what I write down as a US um user task and I find very logical and completely clear um must not be for you. So it might be that your task some misunderstood, it's not clear what you mean. And also you have to check if the prototype or the software really work. If there's no dead end error messages, some tasks that cannot be completed. Perhaps you have to do several iterations before you then can finally go on to testing itself and um hope this comes up soon. And um during testing, I recommend to record um the usability test um because that will make evaluation easier. Um But at the same time, what you observe during the testing is what does the user do you, where does he click? Uh What data does he input? Where does he look? Um Yeah. Does he find information or what does he do then also where does he have difficulties? That's the interesting part. What's not clear to him? Um Which questions does he ask? You will see that the testers always ask questions or most of the time because they are not sure what to do. And then they ask you, we talk about this also a little bit later and also very interesting.

If uh several paths lead to the same result, then it's interesting to observe uh which path a user takes. Um It also might be that um you want to use it to search something on your website, but he decides to use a Google search instead instead of the search of of your website. That would be very interesting to find out why. And um that's why it's necessary to observe this kind of um behavior. After you finished your testing, you can then evaluate what you found, um which problems did arise and, and how often, um how serious are they and how difficult will it be to solve them? Sometimes serious problems are very easy to solve. For example, if um your labels are not clear and the user not sure which data to put in or the navigation is not logical for the user, then it might be just a few clicks of development to change that. And the usability problem is solved could also be that um you have too many fields on one screen or um the the the flow of a of perhaps the login screen is not clear or the registration screen and you have to redo the whole screen flow then um use solving usability problems could also take um quite some time and might be difficult for development.

So it's really tricky when it doesn't, the slides don't change. Um have to check what my next slide is while I wait. Yeah. The next um the good thing is we Yeah. What what we need, sorry, sorry. Yeah. What do you need? Um Not for usability testing. We you don't need a lab. Um What you need is the system that you want to be tested that can be a paper prototype Um So you can already test a sketch of the system that you create or perhaps um you build something prototype in a um design thinking workshop. You can already test that. Then it could be a click dummy that you create with software like sketch fig A or Adobe XD or even powerpoint. Yeah, you can create um prototypes with powerpoint or it could be the the real IT system, either the test system or the production system depending on the development stage. Then um you need user tasks um as we discussed before, that's the instructions for the user um that explain to the user what he or she should do during the test. Then we need the test leader.

That's the person who will guide the, the testers through the whole process from welcoming them to thanking them at the end and perhaps bringing them out to the door again so that they always have one person they can uh talk to and who will take care of them. Possibly you need to have minute keepers if um to take notes. Um as you know, from personal experience, it's very difficult to watch something like um a talk like this and take notes at the point of time. Because normally when we take notes, we look at the paper that we take notes on and then you will miss something that's happening on the screen. So a minute keeper um can then just concentrate on, on taking uh minutes. And uh I also recommend to have technical support close by, especially if you really work on a real it system. Um In case the it system doesn't work, there's a back and new error message, it's good to have someone close by to be able to fix that quickly. Then what else do we need? Yeah, we need tools to um record uh both the screen and the user in times like where we do virtual testing. Um Like I would today um It's quite simple. You have the um the newer tool with like teams or zoom and then you can record the user and the screen at the same point of time, it's more difficult.

If you do testing on site, then you would need to record the screen that also the software for that but also have a webcam, a separate webcam for the user that you can um record the user. Why is the user interesting because it's interesting to see the facial expressions of the user during the test. Uh Does he frown? Is he surprised? Whatever you can see is perhaps angry or insecure. That's you can see that from the face, especially if users don't talk or are very quiet during the test, then it's nice to, to see the face afterwards so that you can um find out what they thought during the test. And then last thing that you might need um is tools to broadcast the test in a different room if, um, several people want to watch the test more than just really the core persons. Um, they should sit in a different room and watch on a separate screen. Testers are really, really nervous and it doesn't help if 10 people are sitting opposite of them and watch what they are doing. Um, that will make them feel terribly uncomfortable, like in an exam situation. So it's better to have other people watching in a separate room. That might be the development team, your clients, the product owner management, who yeah, whoever is interested in in the result then now comes the very good news as to the number of um users as you can see from this graph or my my heading, you only need five users to test that is if you want to have qualitative data.

So just a list of usability problems in your system. If you go for qualitative data, um this KPIS, then you would need more testers to have um valid data. Obviously, if you want to have know what's the average uh success rate, the average time, average time of completion. And obviously with five users, um your results will not be statistically relevant wi five users as you can see from this very famous curve of Mr Nelson and Landau. Um up to five with each additional user, you get a lot of more insight into the usability problems. Whereas after five, the curve starts to flatten out. So more users um won't yield more insights. So that's also the question or the answer to it takes forever. No, it's quite easy to test five people. You can even do that. Perhaps if you're fast and experienced within a day, then I've brought you an example for a user task um that I made up for this um for this talk. Imagine the system to be tested as the transport of London journey planner. Um And you want to test what your goal is to find out if um people, foreigners can plan their journeys with um the journey planner. So you've created one task, we would have several tasks but um one task could be you will land at Heathrow airport in the evening and you reckon that you will be able to catch public transport by 7:15 p.m.

You want to go directly to your hotel, the RS using public transport first task or question to answer. Where do you need to change? Secondly, when will you be at the hotel? And thirdly how long does it take you to walk from the final stop to the hotel? And then you can, the user will try to, to answer these questions and while he he or she does that you can observe and see how they um do it. And if they do it the way you thought they would or differently, that's always very fascinating what they come up with. Um things that you, you didn't think about. So what you really see is that I've given uh concrete data because that's what you, you do as a user, you know what you want to do on the system, you have um data in your head that you want to use to complete the task. And it's also very clear that when the um when the um task is complete, what what has to be answered. So I have to be quick time is running out. So during the test, remember that all testers are nervous um explain to them that you test the system, not them, it's really like an exam situation. Um And perhaps repeat that either you will react to questions.

Um Either you answer questions when, when um testers have them or you don't answer them, you tell them things like just behave as you would if I won't be there. Um But make sure that your reaction is consistent otherwise you will influence the results. And then thirdly, that's very fascinating to observe your behavior does influence the user. They will try to read your body language to find out if they are doing things um right or wrong. And um so you have to be absolutely neutral when you sit there and um try not to show anything that goes on in your mind. Yeah. So about those statements we had um we don't need a lab to test. You. Saw testing doesn't take forever, you don't need a lot of testers, but you don't need specialists. You can do that yourself. And about the, we know our users, we don't need to test things. Try yourself. You will be fascinated and surprised what you find out. And um after you've done, done that, let me know on linkedin, what you find out and what you surprised me most what surprised you most that would interest me very much. OK. That's it. From my usability testing. Talk good. Well, then it's, it's really if something happens, then I would, I would stay there just in case to um handle the problem. Uh So I wouldn't leave the test alone just in case something happens. So I would, I would stay in the room. So thanks for watching.

Um have a great time at the conference. Um So many interesting talks there and perhaps we see each other soon or talk to each other on linkedin. Bye.