"Anytime we ask anybody, “What would you do?” — whether it’s about pricing, behavior, or anything else — that response is garbage. Throw that answer away. It is completely unreliable.”
— Teresa Torres, author of Continuous Discovery Habits.
I love this quote. Last week, I showed one of the ways Monzo fuel their discovery process, by recruiting willing customers for research.
Once you’ve invited them for a chat (research interview) though, what on earth should be on your question sheet?
Some rules of thumb:
- Avoid questions that require people to speculate on their own behaviour. This means avoiding hypotheticals like “What would you do in situation X” or “What would you pay for product Y, given these benefits?”.
- Avoid context-less, direct questions like “What criteria did you use, when deciding to purchase product A?”
- Don’t ask them for solutions like “What do you want?” or “How could we best solve this for you?”
The answers will be “garbage”. However much they may validate your ideas, or you’d love them to be true. Humans are dire at reliably predicting their own future behaviour or accurately recalling past behaviour.
Instead, get them to tell you relevant stories about times they tried to solve the problem your product solves. Using real events helps improve their recall and accuracy of self-reporting. Dig into the details. If they speculate by saying “Oh I usually do…” bring them back to “What did you do in this specific situation though?” Dig in some more.
Record the sessions. Note any distinct tasks in their process and any information about them. Expose as many colleagues to your notes and recordings as you can.
Then start spotting opportunities to improve their life by making those tasks easier.