Framework · Research · 3 min read
Research questions to decisions
A five-step ladder from a vague business worry down to the decision, the research questions, the method and the evidence that would decide, so a study answers something the team will act on.
- 1The worry
The business question, written the way the team says it.
"Why do people stop?" - 2Decision
What changes depending on the answer, who decides and by when.
"Fix it or roll it back?" - 3Questions
Two to five open questions about behaviour and context.
"How do they book today?" - 4Method
Only now: how, and with whom. The question picks it.
"Watch six at their desks" - 5Evidence
The result that would decide, written before the study.
"If we see X, we'll Y"
Use it when
- A stakeholder asks "Do users like it?" or "Can we run a survey?" and you need to find the real question behind it.
- You're about to write a research plan and the goal still fits in one vague sentence.
- A study ended with interesting findings and no change, and you want the next one to land.
How to use it
- Write the business question. Note the worry the way the team says it, without fixing the wording. Output: one sentence, in the team's words.
- Name the decision. Ask what the team will do differently depending on the answer, who decides and by when. If nobody can name a decision, stop and check whether research is needed at all. Output: a decision, an owner and a date.
- List what you need to learn. Write two to five open research questions about behaviour and context, starting with who, what, when, where, why or how. Cut any question that existing data already answers or that can't move the decision. Output: two to five research questions.
- Choose the method and the people. Only now pick how and with whom, from the words of each question. Output: a method and a participant profile.
- Write the deciding evidence. For each likely result, finish the sentence "If we see…, we'll…". Agree it with the decision owner before the first session. Output: one or two "If we see…, we'll…" lines in the plan.
Worked example
A team runs a booking tool for a chain of physiotherapy clinics. The product manager asks: "Do receptionists like the new calendar?"
| Step | What the team wrote |
|---|---|
| Business question | "Receptionists keep calling support about the new calendar." |
| Decision | Before the next quarter's planning, the product manager decides whether to fix the calendar or roll it back. |
| Research questions | How do receptionists book a returning patient today? Where in that flow do they stop or ask for help? What do they do instead? |
| Method and people | Watch six receptionists book real appointments at their desks; read last month's support tickets first. |
| Deciding evidence | "If most of them get stuck on the same step, we fix that step. If they get stuck all over the flow, we roll back and redesign." |
"Do they like it?" became three questions about what people do, and a sentence everyone agreed before seeing a single session.
Watch out for
- Studying the business question as it is. It's too broad. You'll collect a lot of data and still not know what to do.
- Asking participants the research question. "How do you book returning patients?" gets a tidy, general answer. Ask for the story of the last time instead, in your interview guide.
- Questions about taste or the future. "Would you use…?" and "Do you like…?" ask people to predict or judge. Ask about what they did.
- Writing the deciding evidence afterwards. Once you've seen the results, it's too easy to read them as what you wanted. Write it first.
Template
Fill it in with the decision owner, then paste it at the top of the research plan.
Business question (in the team's words):
Decision: What will we do differently depending on the answer?
Owner: Who decides?
By when:
Research questions (2-5, open, about behaviour):
1.
2.
3.
Already answered by existing data? (analytics, tickets, past studies):
Method and people:
Deciding evidence:
If we see ..................................., we'll ...................................
If we see ..................................., we'll ...................................
- Just Enough ResearchErika HallBook · 2024 (Amazon)
- It's Our ResearchTomer SharonBook · 2012 (Amazon)
- Capturing research questionsGOV.UK (external site)
- Plan a round of user researchGOV.UK (external site)
- Discovery: DefinitionNN/g (external site)
"See on Amazon" links are affiliate links. As an Amazon Associate I earn from qualifying purchases, at no extra cost to you.