How Might We 💬 Prompts

SUMMARY

Purpose: To reframe user problems or insights into open-ended design opportunities that encourage creative exploration. 

Design Thinking Phase: Define 

Time: 45–60 min session + 1–2 hours analysis 

Difficulty: ⭐⭐ 

When to use:    After synthesising research into clear findings   To align cross-functional teams around actionable problem statements   Before ideation or sprint planning

What it is

“How Might We” (HMW) is a problem framing technique used to translate user insights into structured prompts for ideation. It reframes challenges as opportunities by asking, “How might we…” followed by a targeted problem or potential.

📺 Video by DAN Innovation Council. Embedded for educational reference.

Why it matters

HMW questions enable product teams to shift from problem-focused mindsets to solution-focused thinking. By framing insights in an open but bounded way, teams can spark collaborative ideation without jumping to solutions too early. It aligns everyone—from stakeholders to engineers—around purposeful design problems.

When to use

•  Right after affinity mapping or thematic research synthesis
•  When reframing vague stakeholder problems into user-centred opportunity areas
•  Before ideation workshops, sprints, or co-design sessions

Benefits

•  Rich Insights: Helps uncover user needs that aren’t visible in metrics.
•  Flexibility: Works across various project types and timelines.
•  User Empathy: Deepens understanding of behaviours and motivations.

How to use it

1.  Gather key user insights or pain points from research (e.g., interview quotes, usability data).
2.  Identify the core challenge behind each insight — what problem is it pointing to?
3.  Reframe that problem into a question using “How might we…” language. Keep it optimistic and open-ended, but specific enough to tackle in ideation.
4.  Write multiple variations if needed — different angles can lead to different solution spaces.
5.  Cluster or prioritise HMW questions based on feasibility, impact, or strategic fit before moving into ideation.

Example Output

From the insight “Users abandon the checkout when shipping options are unclear,” we might generate:

•  How might we help users feel more confident about shipping timelines?
•  How might we make delivery expectations clearer during checkout?
•  How might we reduce uncertainty around shipping costs early in the journey?

Common Pitfalls

•  Too vague: “How might we improve the experience?” lacks direction and won’t guide ideation effectively.
•  Too narrow: “How might we add a chatbot?” forces a specific solution, limiting creative options.
•  Not rooted in research: Skipping the insight phase can result in poorly framed or irrelevant questions.

10 Design-Ready AI Prompts for How Might We – UX/UI Edition

How These Prompts Work (C.S.I.R. Framework)

Each of the templates below follows the C.S.I.R. method — a proven structure for writing clear, effective prompts that get better results from ChatGPT, Claude, Copilot, or any other LLM.

C.S.I.R. stands for:

•  Context: Who you are and the UX situation you're working in
•  Specific Info: Key design inputs, tasks, or constraints the AI should consider
•  Intent: What you want the AI to help you achieve
•  Response Format: The structure or format you want the AI to return (e.g. checklist, table, journey map)

 

Level up your career with smarter AI prompts.    Get templates used by UX leaders — no guesswork, just results.   Design faster, research smarter, and ship with confidence.   First one’s free. Unlock all 10 by becoming a member.

Prompt Template 1: “Generate How Might We questions from a research insight”

Generate How Might We questions from a research insight

Context: You are a Product Designer synthesising findings from a usability study on [checkout experience].  
Specific Info: One key insight shows that users [abandon the process] when they encounter [surprise costs at the final step].  
Intent: Transform this insight into a variety of HMW prompts suitable for a design sprint.  
Response Format: Return a list of 5–7 “How might we…” questions with a note on what aspect each one frames (e.g., clarity, trust, timing).

Ask for clarification if the insight lacks user behaviour detail.
Offer one idea for reframing the insight if better HMW questions could emerge.