HMW Questions: From Research Insights to Design Ideas

Why it exists

Discovery research produces rich knowledge about user problems, but teams often jump to pet solutions that ignore what was learned. HMW questions were created to bridge this gap by forcing teams to translate insights into shared problem statements before ideation begins. They keep the group focused on user needs rather than individual preferences.

The mental model

Think of an HMW question as a key that unlocks a specific door. Too narrow and it only opens one room; too broad and it does not fit the lock at all. The sweet spot is a framing that is open enough to surprise you with unexpected solutions, yet tight enough to know when an idea is off-topic. It shifts the conversation from we need a dashboard to how might we help users feel in control of their data.

How it works

Start with a concrete insight or problem from your discovery research. Rephrase it as a question beginning with How might we. Remove any embedded solution so the question is agnostic about the means. Then broaden it slightly to expand the possibility space without losing the original intent. For example, if users spend too long checking submissions for mistakes, avoid how might we add a checklist and instead ask how might we support users to efficiently draft submissions they are happy with.

When to use it

Use HMW questions at the end of a discovery phase, after the team has synthesized findings and agreed on top insights. They work best in a workshop setting where cross-functional stakeholders write questions together, then vote on which ones to take into ideation sessions.

When not to use it

Do not use HMW questions before you have research insights, or as a substitute for understanding the problem space. A generic question like how might we improve the user experience generates generic ideas because it is not anchored to anything you learned and will fail to address the root problems you uncovered.

One canonical example

A team discovers that users are unsure which tax form to complete. A poor HMW asks how might we tell users which form to complete, which limits ideas to communication tactics. A strong HMW asks how might we make users feel confident they are filing their taxes correctly, opening the door to auto-filing, tailored questions, or simplified forms.