The art of guerrilla usability testing

(David Peter Simon Ellazou & Crispinzhu Lynnwang Correction Check the original reproduced please specify the source)

Guerrilla usability testing is a very effective technique. "It's an art, like a random pickup in a café or public place, and a quick shot of a few minutes of video while he's browsing the web," says designer Martin Belland. "We now skip the part of the random pick-up technique and go to its subtleties, including how to get and share feedback with the team."

I recently participated in a Quick Launch project that required my team to complete a responsive web site in a very short time. We have been given very little time to develop, let alone do user research. However, thanks to the guerrilla usability test, we gained user feedback on brand positioning. Ultimately, our design meets both user expectations and business goals.

During the project, we test a different type of prototype every week to make the business idea a reality. For example, in the middle of development, we draw the prototype on the index card, and quickly evaluate it. The evaluation exposes the problem of navigation (which will guide us to rethink the core of the user's view of the whole station), and even stop the shaping of some brand media material. In addition, the tests let project-focused people subvert the "users" they instinctively understand.

What is the nature of guerrilla usability testing? The nature of testing requires ease of operation. It can help us to verify (or falsify) the core assumptions at low cost and at the fastest speed.

decomposition

Is it hard to understand how the guerrilla usability test is magical and how does it work? Before you begin, consider a few basic questions:

, what are we going to test? Where do we test? Who are we going to test? How do we test? What are we going to test?

The best thing about this kind of testing is that it's not open-ended, from a hand-drawn prototype behind a paper towel to a full-featured prototype. I very much agree with Stephen Crugger's proposal, before we think we should do the test before the start, the important thing is to go out as soon as possible.

Determine what the product should be made of by measuring the appearance of the product we can measure. Even a piecemeal definition of a UI sketch is a good way to evaluate future products. In fact, recent research has shown that low-fidelity prototypes can better win the cooperation of senior and junior users.

where are we going to test?

Where to start the test will affect how we operate and record it. For example, if we want to test a new retail chain of mobile apps, we might as well go to the store, wandering in the aisle, if we are doing "regular" office software, we might as well test the staff in different office areas, and so on. The principle is: decide how to carry out by the scene of need.

, who are we going to test?

When designing for big supermarkets, it's easier to ask strangers who seem friendly to have a few minutes of free time. Public places and shopping centers are good choices for crowded places and rest areas. In any case, it is useful to test the user's situation, as mentioned above: a scenario that covers a variety of environments and behaviors.

The coffee shop is a good choice because you can often find target test users in different cultural backgrounds and age stages.

How do we test?

The test approach is fairly straightforward: let testers think when they are performing their tasks. Using sound thinking to test a product's overall understanding is better than simply focusing on whether the task is complete or not. The principle is to observe how the user manipulates the product and silently evaluates its usability. As Shala Harrison said, "Watching users is like picking teeth--everyone knows they should do it every day, but they can't." Just do it, it's no big deal. ”

Always start with an open, boot-free problem:

How do you understand it? What are you going to do now? What would you do?

Through these questions, the participants describe their understanding of the product as a 7788. In this way, we can develop some ideas about how to improve the design in the next phase.

Use technology

Guerrilla usability testing is deeply influenced by context. Even so, some useful insights have been drawn in my long-standing international practice:

Beware of implicit prejudices. When you find a test user in a café, pay attention to those who repeatedly refer to our product as biased. Simply keeping an awareness of implicit prejudices can help designers maintain their subjective neutrality and count individual differences. Remember to seek to be equal in gender. explain why. The designer should introduce himself truthfully, test the reason and expect to get the feedback of that kind. Often, it is better to distribute forms so that participants can fully understand the implications of their participation-for example, the study will be shared only at the national level in international conferences. These forms, while somewhat boring to carry, can help build trust. Ethical. Of course, honesty doesn't mean we have to be transparent. Sometimes it is better to avoid some information, such as the product you are testing is designed by us. Instead, we can spread a white lie about our research goals. We just have to make sure that each round is the last to tell the truth: Trust is a prerequisite for successful cooperation. Keep it casual. The test could be a reward for being tried to buy a cup of coffee or a meal. In the process of waiting in line or ordering, we can inquire about the life condition of the subjects and introduce the way of testing. Keep a sense of participation. Eliminate communication barriers by getting users involved: try to draw the third or fourth interface of the process in a tissue or notebook. There is no need for the subjects to draw a complete interface, only they can express the concept of the first approximation of the mind. You never know what you will learn from the inspirational imagination. Do not direct participants. When you feel puzzled, ask the user what they think. Like "I don't quite understand." How do you think about it? "Such questions can open the user's words. Users often feel that they are being tested in a test scenario, so it's best to start with apologies to avoid user resistance. Observe attentively. The idea of a slightly ephemeral, left for follow-up analysis is very important. Ethnographic observations can help you review what you think in your test. Don't dwell on neat notes, and in most cases you can do quite a lot of work. It only needs to be used to inspire memories, not to be presented at academic conferences. Capture feedback. The core work of any test is to capture the harvest. While this is a complete personal choice, some of the tools that are available are recommended: Silverback app or UX recorder the face response of the test user when recording the active view. Other researchers have developed mobile rigs tools. Remember to use tools to meet future sharing needs. Control time. Remember, this is not a paid test in a usability lab. Pay attention to the time spent by each user and remind them to leave at any time. You certainly don't want grumpy users to distort your feedback. Share Feedback

Only half of the battle was completed after the test. To present a compelling, significant guerrilla usability test report, designers need to strategically decide how to share our findings with colleagues.

When analyzing and preparing feedback, consider your audience first. The best feedback is to understand the important conversations between the project's focus person and the initiator. For example, development needs to assess the bugs that new requirements bring, and the person in charge wants to prioritize new requirements.

Furthermore, when submitting feedback, align your audience's expectations. Try editing a imovie video or ppt slide. Your partner is probably as busy as you are, so editing a "trailer" to refine a major conclusion or focus on a strong user's words can keep your audience focused.

Please try a guerrilla usability test

So far, guerrilla-type usability tests have been in different forms. The art has yet to be perfected. It's a cheeky impromptu speech. Remember to build your own approach: learn in practice.