A beginner’s guide to guerrilla research

4 min read
Lade Tawak
  •  Oct 18, 2019
Link copied to clipboard

User research can be a hard sell for clients and stakeholders who don’t understand its value. They might have questions like:

  • Where does it fit into the budget?
  • Do we have enough time to make it happen?
  • Is research even worth it?

Unfortunately, some people can’t be convinced of the value of user research—so designers have had to get creative in developing time-effective, low-cost, and impactful methods.

Guerrilla research is one of these. 

Let’s discuss:

  • The definition of guerrilla research
  • When guerrilla research should, and shouldn’t, be used
  • How to implement guerrilla research methods

What is guerrilla research?

Guerrilla research is a quick, low-cost way of learning about and understanding experiences. It is usually done in public spaces and does not require a rigorous recruitment process, although it does require its own type of planning. 

Document your user research findings using InVision Freehand.

Especially when facing pushback from stakeholders regarding the cost of user research or the benefits of user-centered design, guerrilla research can prove the value of research with minimal investment. Its inherent flexibility—online or offline, one day or one week, sessions as short as ten minutes—makes it an easier sell than a full research program.

When should you use guerrilla research?

Guerrilla research is most effective for research that has a small scope, especially when used for intercept studies at the beginning of a project. An intercept study is a research method that involves gathering feedback from users onsite as they complete a process. For example, introducing a short survey after a user completes a transaction, or interviewing participants of a conference as they leave the conference. You can use intercepts to define and refine what areas to explore more deeply for more foundational research, such as contextual inquiries, interviews, participatory design research, and when necessary, surveys.

“Especially when facing pushback from stakeholders regarding the cost of user research or the benefits of user-centered design, guerrilla research can prove the value of research with minimal investment.”

Twitter Logo

Guerrilla research is also helpful for testing product concepts at all levels of fidelity, from paper prototypes to high-fidelity mockups. This will help you identify quick fixes and learn if potential users have a general understanding of what your concept is about. It is best to limit this to one feature, flow or 2 or 3 tasks as you want to keep it short and quick. Guerrilla testing can be implemented during design sprints or workshops or in agile or lean processes. 

When should you avoid guerrilla research?

The beauty and brilliance of guerrilla research lie in its flexibility. Therefore, projects with stringent needs from users aren’t a great fit. These projects include:

  • Sensitive subjects, like finance of healthcare
  • “Expert” personas who require specific domain knowledge
  • Context-dependent experiences

This is because people are unlikely to share sensitive topics or want to discuss domain-specific knowledge with a stranger who bumps into them at a mall. Imagine you went to a mall or a coffee shop to relax or do some work and someone comes to ask you about your sex life or spending habits? They don’t want to just check boxes in a survey, they want to have a full conversation with you. How would that make you feel? 

Watch people use your product or prototype in real-time. Photo by Charisse Kenion on Unsplash.

How do you conduct guerrilla research?

Like any research process, guerrilla research can only be successful when properly planned.

Define your goal

Like with any other form of research, you need to clearly define what you want to find out before stepping out into the world. If you already have a prototype, be it low- or high-fidelity, determine what tasks (limit this to two or three), feature, or flow you want to test or what questions you want to ask. Determine what metrics you’re tracking if any, and what the findings of the study will be used for.

Examples of goals for guerrilla research:

  • To understand if and why proposed investment tracker feature appeals to users
  • To assess if users understand the purpose of scan-to-pay feature
  • To understand how current law students study for examinations

Define your participants

While you’re usually testing concepts or ideas that do not require specialized knowledge, depending on what feature, flow, or idea you’re working on, you might need to define your participant group.

For example, when testing the usability of news sites, I wanted to talk with people who already read news online every day. I did not consider gender and age as participant criteria, but I made sure to talk to people of different ages and genders.

If you were working on a study app for law school students, available only as a mobile app, your participant criteria for usability testing would include people who are current law school students and use a smartphone.

Create a discussion guide and short screener

Once you’ve defined your goal and your target group, you should write out a short guide that helps you recruit people on the spot. This is known as a recruitment screener, which is a series of questions that allow you to qualify participants for your research, ensuring you’re talking to the right people. For example, if you want to understand how people order food online, you might want to include screener questions to find out if, how often, or where they order food.

Another reason screeners are useful is to help ensure you have the right mix of people. For example, if you want to find out their level of tech proficiency, you can include a question on that in your screener. Limit your screener to two or three questions. 

For my newspaper website study, my first screener question was:

How often do you read the news online (never, once a week, two to three times a week, four to six times a week, every day)

If they answered “never”, I thanked them for their time and moved to the next person. If they answered any of the other options, I would ask them this:

Where do you read the news online?

I would only proceed to the usability testing if they answered on their phone.

In addition to the screener, you’ll need a discussion guide, highlighting how you’ll introduce yourself to the participant as well as what areas you want to focus on in the short time you have together. You could start with something like

Hi, my name is Lade and I’m working on a project to understand people’s experiences using news websites. Would you mind trying out some things on a website and giving your feedback?

In your notes, you should include what specific areas you would like to cover, like experience using the website or checkout flow.

Choose your location

Determine where you’ll be conducting sessions, and get permission when necessary. You might not need permission if you’re working in a mall or park, for example, whereas you probably will if you’re inside a store.

Here are some factors to consider when choosing a location:

  • Consider places with easy access and a lot of human traffic like malls or parks
  • If you’re outdoors, find a shaded area to conduct sessions.
  • Ensure that you avoid glare and spots where the researcher will have difficulty watching what the participant is doing.
  • Pick a place with stable WiFi if you require an internet connection.
  • Consider where you might find people in your target audience. For example, if you’re targeting lawyers, then hang around a courthouse or a cafe near a courthouse. If you’re trying to understand college students, go to a school campus close to you.
Follow the large groups of people. Photo by Anna Dziubinska on Unsplash.

Before you start

Combine your research goal, participant criteria, discussion guide, and screener, as well as study location in one document that you can share with your team. Always do a test run to ensure that you’re not going over time and you’re asking the right questions.

Try to have everything you need backed up offline so that you avoid connectivity issues. If your prototype requires an internet connection, ensure that you have a backup in case WiFi fails. Remember to bring along a backup power source if you will be using a device.

Don’t rely on voice or video recordings—instead, bring a notetaker if you can. This is because it is best to avoid the additional time required to get permissions and set up recordings—and because public spaces are usually noisy, which would make recordings unclear. You should only have one notetaker, as having too many people crowded around the participant might make them uncomfortable. If you’re using a mobile phone or laptop, you could try using a screen recorder. 

Decide on whether or not you’ll be giving incentives, like $10 gift cards or branded items such as stickers or shirts.

Approaching participants

When conducting guerrilla research, there is a tendency to select only “friendly” faces. Avoid this potential bias by creating a system to randomize who you approach. For example, stop every 3rd person. People will say no, so be willing to take no for an answer and back off.

Don’t approach people who look busy or in a hurry. Only ask people who are clearly not in a hurry, and seem available for a discussion.

During the session

When someone agrees to participate in your study, thank them for their time and remind them that you’ll only be spending the agreed-upon amount of time. Remember to inform them that they’re free to stop at any time.

Before you start, remind your partner (if you have one) to stay quiet during the session. Their job is only to observe and take notes. If they have anything to add, it should be at the end of the session and should be limited to one or two questions

Stick to the discussion guide as much as possible, and avoid the temptation to go on tangents, no matter how interesting. Make the participant feel comfortable during the session, especially if you have someone else with you.

Remember, the same guidelines apply to guerrilla research as any other form of user research: no leading questions, observe what they do, listen to what they say.

Be ready to adapt to whatever happens as you go along. If the participant decides to end the session before you go through the discussion guide, be willing to wrap it up.

After the session

At the end of each session, say thank you to the participants and give them incentives if any. If you’re conducting the study alone, immediately after each session, write down your observations, quotes, ideas, or anything else you think is important.

If you have a partner, talk through the notes and do a quick debrief. Write down everything you remember. You can do this digitally or with debrief sheets, and then refer to this document for your final analysis later.

If you have the resources to conduct larger, deeper research, then do so. However, guerrilla research can help you identify quick fixes and work around tight budget and time constraints to get direct user feedback while showing stakeholders the value of conducting user research and involving the users in the design process.

For more tips and examples on getting started with guerrilla research, check out the following:

Collaborate in real time on a digital whiteboard