Skip to content
Design Academy My progress

Stage 2 · Lesson 2 of 13

Research questions and assumptions

Writing questions that research can actually answer.

  • Beginner
  • 30 min
  • 3 check questions
  • Exercise: 35 min

By the end of this lesson you can

  • Turn a vague project brief into specific research questions.
  • List the assumptions a team is making and rank them by risk.
  • Tell the difference between research questions and interview questions.

The idea

Research starts with a question, not a method. If you do not know what you are trying to learn, you will collect lots of interesting material and still be unable to make a decision.

Research questions

A research question states what the team needs to learn to move forward. Good research questions are specific, answerable with evidence, and tied to a decision. "What do people think of our app?" is too broad. "How do first-time customers decide which plan to choose, and where do they hesitate?" is specific and connected to the pricing page the team is about to redesign.

Research questions are not interview questions

You rarely ask a research question directly. If your research question is "Why do people abandon their basket?", asking participants "Why did you abandon your basket?" invites guesses and rationalisations. Instead you ask them to describe the last time they bought something online, watch them shop, and look for the moments they hesitate.

Assumptions

Every project rests on assumptions: beliefs the team treats as true without evidence. "Customers know what a direct debit is." "Most people order on their phone." Writing them down makes them visible. Then rank them: which ones, if wrong, would cause the most damage, and how confident are we? High-risk, low-confidence assumptions become your research priorities.

Why it matters in real projects

Research costs time and participants' goodwill. Clear questions make sure that effort answers what the team actually needs to know. They also make research easier to plan, easier to analyse and easier to report, because every finding can be traced back to a question. Listing assumptions protects the team from building on beliefs nobody checked.

A realistic example

A council wants to "improve the bin collection page". The designer runs a short workshop and collects assumptions: residents know their collection day; most people visit the page when a collection is missed; residents understand which bin takes which materials.

The team ranks them. The last one is high risk (wrong bins lead to rejected collections and complaints) and low confidence (nobody has checked). It becomes the main research question: "How do residents decide which bin to put an item in, and what do they do when unsure?"

The designer plans short interviews at home, asking residents to sort a bag of typical household items while thinking aloud. The findings lead to a searchable "which bin?" tool rather than a redesigned calendar.

Step by step

  1. Write down the decision

    What will the team decide after the research? Start there.

  2. List assumptions

    Ask everyone to write what they believe about users, their behaviour and the context. One belief per note.

  3. Rank by risk and confidence

    Plot each assumption on a 2x2: impact if wrong against how much evidence exists.

  4. Write 2 to 4 research questions

    Focus on the high-risk, low-evidence corner. Make each specific and answerable.

  5. Check each question

    Ask: could evidence answer this? Would the answer change what we do? If not, rewrite or drop it.

Common mistakes

  • Mistake Starting with a method ("let's do a survey").

    Instead Write the question first, then choose the method that can answer it.

  • Mistake Too many research questions.

    Instead Keep to two to four per study. More questions mean shallower answers.

  • Mistake Asking research questions directly to participants.

    Instead Ask about specific past behaviour and observe tasks instead.

Check your understanding

Choose an answer, then check it. There is an explanation either way.

Question 1 of 3Which is the strongest research question?
Show answer

How do freelancers decide when to send an invoice, and what delays them?. It is specific, answerable through evidence and linked to a product area.

Question 2 of 3Which assumptions should research tackle first?
Show answer

High impact, low confidence. Those are the beliefs that would cause the most harm if wrong and that have the least evidence.

Question 3 of 3Why should you avoid asking "Why did you abandon your basket?" directly?
Show answer

People tend to guess or rationalise rather than report what happened. Asking about concrete past events and observing behaviour gives more reliable evidence.

Practice exercise

Exercise · about 35 min

Assumption map

Pick a product idea, such as an app that helps flatmates split bills, and map its assumptions.

Steps

  1. Write at least 10 assumptions about users, behaviour and context.
  2. Place them on an impact versus evidence 2x2.
  3. Write three research questions for the riskiest corner.
  4. For one question, write three interview prompts that ask about past behaviour.

Deliverable

The 2x2 map, three research questions and three interview prompts.

Self-assessment

Tick each item when it is true of your exercise. Be honest; this is for you.

4 items

Summary

  • Research starts with questions tied to decisions, not with methods.
  • Research questions guide the study; they are not what you ask participants.
  • Listing and ranking assumptions shows where research is most needed.
  • Keep a study to a few focused questions.

Further reading

Third-party sources, chosen for reliability. They open in a new tab.

Your notes

Marking complete updates your course and roadmap progress.

Back to UX research and problem definition Roadmap

PreviousUnderstanding users and business goals Next lessonDesk research and competitive analysis