# How Might We Question

A how-might-we question turns an evidenced user problem into an open design challenge that allows several possible solutions.

- Type: method
- Domain: Strategy & discovery
- Level: Foundational
- Last reviewed: 2026-10-20
- Original: [Soroush’s Product Wiki](<https://wiki.soroush.uk/topic/how-might-we-question/>)

## What it means

A how-might-we question, often shortened to HMW, frames an opportunity for exploration. It names a useful outcome without prescribing the implementation. Its scope should be specific enough to guide ideation and broad enough to permit alternatives; changing a feature request into question form is not sufficient.

## Why it matters

The wording of a challenge shapes the ideas a team considers. Starting with a desired user outcome can reveal simpler changes, service improvements, or different interaction approaches that a solution-first brief would exclude.

## When to use it

Use HMW questions after research or problem framing and before generating concepts. They are less useful when the problem is still an unsupported assumption or when a fixed technical correction already has a clear implementation.

## How it works

1. Select an observed problem and the people affected.
2. Describe the improvement those people need.
3. Write several questions that preserve the outcome but leave the solution open.
4. Check for hidden feature prescriptions and overly broad wording.
5. Choose a question, generate alternatives, and identify what each would need to prove.

## A practical example

A team replaces “How might we add a chatbot to returns?” with “How might we help customers choose the correct return option confidently?” The revised question allows clearer product labels, better examples, or assisted support to compete with the chatbot idea.

## Where it breaks

A well-worded question does not validate the problem or guarantee useful ideas. Business constraints and excluded groups can disappear in optimistic wording. Keep the underlying evidence and non-negotiable constraints alongside the question.

## Common mistakes

- Embedding the preferred solution in the question.
- Making the challenge so broad that any idea qualifies.
- Treating workshop agreement as proof of a user need.

## Sources

1. [Using “How Might We” Questions to Ideate on the Right Problems](<https://www.nngroup.com/articles/how-might-we-questions/>) — Nielsen Norman Group · Professional guidance
2. [How Might We](<https://www.designkit.org/methods/how-might-we.html>) — IDEO.org Design Kit · Method guide

## Related topics

- [Problem Framing](<https://wiki.soroush.uk/topic/problem-framing/>)
- [Opportunity](<https://wiki.soroush.uk/topic/opportunity/>)
- [Crazy Eights](<https://wiki.soroush.uk/topic/crazy-eights/>)
