The principle of guerrilla research is to get useful insights in an efficient way without needing to create complex prototypes, recruit a certain demographic, or be a seasoned user research expert.
Guerrilla research is ideal for teams who want to test ideas early and often, allowing them to move on with a project as quickly as possible.
What you need
- An individual willing to run a session by themselves, ideally someone who can think on the spot with questions that might come up, or delve deeper into some UX problems if needed.
- A prototype you’ve tested with a colleague beforehand.
- A laptop and any devices you might need to show the prototype.
- Software to record the screen and audio (QuickTime does this).
- An environment where it’s acceptable to approach people. Student unions and slightly busier coffee shops are great for this as people tend to be relaxed and in a positive mood.
- Bonus: a gift to give as a thank you once research has been completed. A $5 coffee shop gift card is ideal.
Set aside about 2 hours and gather feedback from 4–5 people.
The shape of a session
A guerrilla research session should be quick enough to not bother the participant, but long enough to gather enough insight. Somewhere up to around 15 minutes is perfect, but I’ve gathered great feedback from people in as little as 5 minutes.
Here’s an outline of an ideal session:
- Approach an individual and ask if they have a little time to help you out with something. Introduce yourself and provide credentials (e.g. a business card) to make the person feel comfortable.
- Once they’ve accepted, show them your laptop and set the stage. Tell them you’re looking for some design feedback and that they aren’t being tested for their computer literacy skills.
- Ask for permission to record the screen and audio, reassuring them that it won’t record their face.
- Walk through the prototype, asking them to perform the tasks you’ve already piloted with a colleague.
- Any time you need more information, be sure to ask for clarification or dig deeper so the insights are useful.
- Once completed, thank them for their time and — if available — provide them with the incentive.
If at any point during the process it feels like the feedback you are receiving is unhelpful, thank them for their time and move on to another person when it is appropriate.
After the sessions
Once all tests are complete you should go back through the screen recordings and note down any key takeaways and your interpretations of the feedback, for example:
4 out of 5 people struggled to figure out how to reset their password, so we need to work on improving this flow.
BUT don’t forget to highlight the things that went well for to your team. Research is about validating or invalidating your work, so if you’ve got wins in there, flag them.
Test the right thing
Choosing how to validate or invalidate assumptions about designs is essential to ensuring any results you get are robust. I’ve found guerrilla research useful to test interactions that don’t require domain knowledge such as searching, registering, finding how to reset your password, or filtering results on a store. Otherwise, it is worth exploring other methodologies for your research needs.
When properly implemented, guerrilla research is an effective tactic for understanding how users interact with your product. The right tools and a discerning mindset let you gain valuable insight on what ideas are working well, and what may still need improvement. With these goals in mind, guerrilla research becomes an essential practice for any team regardless of budgets or timelines.
If you’re looking to learn more about guerrilla research, here is some further reading:
- The Pros and Cons of Guerrilla Research by the IDF
- What is guerrilla testing and how do I use it by Elizabeth Chesters
- Guerrilla research: quick, not dirty by Foolproof
Have you used guerrilla research and want to talk about it? The best thing to do is follow me on twitter and send a DM.