Remote testing is a passion of mine, because it's exciting to get outside of the lab to see how people are interacting with products in the settings that they are using them in. The fact is by enlarge, most researchers these days are getting outside the lab at least part of the time. It really is because they need to get a better understanding of how it is that people are using products and not using products, is that they can get a sense for what it is they need to do to improve them in a more naturalistic setting. Very simply, in a remote test, you have a moderator and a participant in two different physical locations. Typically, participants are in their own home, office, or other location, and the moderator is watching remotely using either a webcam or some other type of technology. The benefits of this are that it's less expensive than doing research in a lab. Instead of having to spend time recruiting and getting people to come physically to your lab, you can essentially say, ''We'd like to have you run through this test on your time or a time that is mutually beneficial.'' That will enable us to get a good sense of what's happening in your own context. It enables you to test with geographically dispersed people. So, whether they are across the country or around the world, you can get a sense for what's happening in their home? What does that office look like? How are they cooking that recipe in their kitchen? It's also less time-consuming. That essentially means that especially in the unmoderated setting, you don't have to sit there and watch someone do an entire usability test while they're doing it. You can pick and choose later on the parts of the session that you're most interested in learning more about. Then finally, it highlights contexts. So, you'll never get an understanding that someone is actually being interrupted by their child who wants to get a snack when they're potentially cooking a recipe that you want to learn more about using an app that you're trying to learn more about, if you actually aren't doing that research while they are in their home. One of the biggest challenges with doing remote testing is that technical challenges can be very difficult to get around. So, despite the fact that we have significant improvements in speed and reliability around WiFi connectivity and high-speed connectivity, that's not the case all the way around the world, also not everyone has the ability to very quickly learn how to use some of the new technologies that you might be asking them to use so that you can capture their session. So, a little bit more about remote testing that is moderated. Here, you've got a moderator and participant in two different locations. The participants usually in their own home or maybe even sometimes in a store, or somewhere else physically, and the moderator is watching remotely. A moderated test enables you to do several things. First of all, moderators can help to guide people through a test. They can clarify questions or probe more deeply if people want to or give an indication that they want to speak more in depth about a particular issue that they're coming across. But the con here is, when is it right to probe? Sometimes, a simple ''aha'' is okay, but also sometimes you might bias the test you provide if you provide too much guidance or provide too much information about how to succeed during a task. Unmoderated testing essentially means that you have a participant in their own home following a script. This can be accomplished two different ways. You can do qualitative or quantitative research. So, in this sense, when you're doing qualitative remote unmoderated testing, participants used the think aloud method throughout the task and provide you with feedback that you then have to review and analyze by hand, which essentially means that you need to watch the session to get a good sense of what happened and why. For quantitative testing, that's information, the why is less important. In this case, you're much more interested in gathering that performance data and infrequently, are you collecting that feedback or that think aloud data. So, you would set in ideal metrics such as task success. So, what is that task success condition or time on task? A few thoughts about unmoderated user testing. So, the first is high volume enables you to get sometimes a very conclusive results, and it also enables you since you don't have to do all that testing in-person, to test with much larger groups of people, and if you're using a system, potentially all at once. So, you could be doing a moderated remote testing with sample sizes, upwards of several hundred in the course of several days, whereas that would not be a practical solution if you're trying to accomplish this in a lab setting. So, obviously, this great when time is of the essence, which is often late in the product design life-cycle. The drawback here is that you can't ask those follow-up questions if someone gets off track, there's no way that you can help them to get back on track. So, you don't have that real-time support to get people to succeed, or actually to complete a task the way that you might wish for them to do so, or to sometimes answer the question that you would want for them to answer. Remote testing is not limited to just usability testing. Often these days, UX researchers are doing tree testing, card sorting, light ethnography. First, click testing or even interviews, using remote moderated, or unmoderated methods. These are all really great ways to not only give yourself access to more users, but also a more diverse group of users than we've ever been able to access in the past. There are good reasons to stay in the lab, and there are many organizations that make the choice to stay in the lab for these reasons and others. The key reasons are really around sensitivity of new or highly secure apps or data. So, typically, in the financial services industry, or when you're creating a very new bleeding edge technology, you don't want other people to get a sense for what it looks like and how it works. So, you want to have control over access to it. The other is, if location doesn't really matter, then it's fine to stay in the lab where you can control the conditions of the experiment. You might also consider staying in the lab if it's very difficult to create the right conditions or to set up the test itself. So, it may mean that you need to remove cookies from a computer, or remove some data, or to reset people so they're starting from the same starting condition. That might be best given a computer over which you have all the control that you need to have. Finally, sometimes you're doing usability testing not just to answer a question, but to build the credibility of your team. If you do your testing in the lab, you can bring others from your design and development team to watch the tests so that they get a sense for, what are the things that users are saying and doing? Earlier in this course, we talked about what types of research questions would help you to stay on track and get the types of insights that you're most interested in collecting. So, I am going to talk a little bit about the types of questions that are best answered in the remote moderated and unmoderated contexts. First, around who, you can ask, what are people's preferences, and what are people's needs? This is best accomplished by using the interview method, which can be either moderated or unmoderated. Another approach is to understand what people are doing now, what they would prefer to do, and what their expectations are. Finally, and this is very typical of usability testing, what features are working or not working for people? Remote testing does not really lend itself to answering any when type of question. So, typically, in remote usability testing, you're wanting to understand more about where people are expecting to find content. So, having them take you through a guided tour of how it is that they would approach looking for the answer for a question, is critical here. Digging more deeply into where calls to action or control should be located, or where do people expect to find content? Why do people behave the way that they do is critical. So, this is something that is best went to remote moderated, as well as from remote unmoderated usability testing, because you get a sense for the difference between what people are saying, and what it is that they're doing. How people accomplish their goals is something that you can capture both in a lab setting, as well as remotely. So, how did they prefer to have information organized, or interact with that content? Getting a better sense of seeing a larger number of people making those choices about which pathway to follow through a remote usability test, enables you to get a better sense for, what is the natural way that people would expect to complete a task? So, you get a sense for how many people are exposed to key content, and what they do not choose to visit during an experience. Bottom line. Nowadays, people err on the side of attempting to test as much as possible with people in their own environments. That is because the more that you can get a sense for what the context constraints needs and behaviors are of people in the settings where they're actually using your products, the better you're going to be able to make those products function for them. Moderation is not always necessary because if you are providing people with sufficient instruction, they'll be able to complete the task that you need them to and you'll come away with the answers to the questions that you need.
