Guerrilla testing is a simple method for those cases where there is no time to explain. Guerrilla testing is conducted where the target audience gathers: in the streets or shopping malls. The researcher interviews ordinary passersby. Guerilla testing is a great option if:
- The product has a wide target audience. This allows you to collect feedback from customers in a store or anywhere else.
- Testing takes up to five minutes. After that, people are in a hurry for business, and they don’t have time to answer any questions.
- There’s no budget for a test at all. In this case, guerrilla testing is a possible way out.
Guerilla testing is a bad option if:
- It would help if you had a specific audience. The likelihood of meeting a 35-year-old man on the street actively using a particular bank application tends to zero.
- You need to do an in-depth interview. Unfortunately, people are rarely willing to spend an hour discussing a product. They are always in a hurry to get somewhere, except when a person has to wait a long time for something. For example, they are standing in line at the visa center or sitting at the airport.
There are no free respondents, and finding them is always spending of working time. So the cheapest way to recruit ten respondents is guerrilla testing.
5 Guerilla Testing Steps
Guerilla testing consists of 5 steps at which you should be maximally prepared. Let’s discuss them in detail.
Determine what to test
It would help if you understood that you are unlikely to test the whole website or application. So, formulate a single goal of the guerilla testing and build tasks for it. This will allow you to make the usability guerilla testing qualitative and collect more data for further analysis. What should be tested first? Pay attention to the selling elements and entry points at your website or application. These include the home page, catalog sections, product page, shopping cart (for online stores), checkout page, and service or product description page.
Shortlist your list of tasks to test
Now shortlist the tasks you’d like to include in your guerrilla testing. Then construct a step-by-step path for the user to achieve the goal of the testing (e.g., to place an order, find a certain product, etc.). This will be the scenario (task) according to which users will act at your site or application during the guerrilla test.
Pay attention to the number of tasks in the scenario, and the optimal number is no more than 5. Do not set complex tasks. One task = one action. Formulate tasks as briefly and simply as possible. Guerrilla test participants should not have any questions about what is required of them.
But! The participants themselves should ask additional questions on each task. This will allow you to understand the chain of their thinking better while performing the task and trace the problem points. Such questions can be included in the script or formalized in a separate conversation with testers after the guerrilla test is completed.
Conduct guerrilla usability tests
An introductory briefing is recommended before the guerilla test begins. During the test, it is important to record: the remarks of the participants, mimicry of the test participant (filmed on video), the screen of the computer with which the participant works, and user actions during the task, including moving the mouse, switching between browser tabs, using the keyboard, etc. All of these parameters should be time-synchronized.
After completing the guerilla test, collect the data and structure it. Record the number of errors made by users, the number of completed tasks, total testing time, and the number of problems identified. Systematize the data on what specific difficulties users had and at what stage of the task, their comments, and answers to additional questions. When analyzing results in this way, try to stick to specific wording, i.e., record a specific user action.
Each guerilla test, prepared in advance, can identify more than one problem on your website or application. Therefore, it makes sense to divide the entire list of identified problems into critical, serious, and unimportant. This will not allow you to get bogged down in small things and quickly fix the really important deficiencies. In the end, formulate the guerilla testing conclusions and proceed to the edits.
Fix the issues
If you were testing out prototypes, fixing problems shouldn’t take too much time. However, if you were testing out a nearly completed project, you’d likely have to spend some time and money to fix the issues.
Guerilla testing should not replace more formal user research. But if you don’t have the budget for formal testing, it can be a promising option. Try guerrilla testing and get valuable insights!