Lesson 2.2

Design Thinking/Module 2 · Define — framing the right problem

Lesson 2.2

How Might We

Turning problems into open doors

6 min Lesson 11 of 32

The hook
You have Lakshmi's sharp POV. Why not design now? Because a statement is closed — the mind reads it and looks for the answer. The best design comes from generating many answers first. So you make one grammatical move: 'Lakshmi needs to belong' becomes 'How might we help Lakshmi belong without asking?' Three little words that change a closed problem into an open door.

Why a question beats a statement

A statement invites judgement; a question invites generation. 'How might we...' shifts the mind from judging one solution to producing options. This is the bridge from Define to Ideate: converge to the right problem, then deliberately reopen that specific problem for solutions. Each word does work: 'how' assumes a solution exists and asks for method; 'might' lowers the stakes and gives permission to be wild and imperfect; 'we' makes solving a shared act, not a lone-expert verdict.

The Goldilocks problem: scope

Too broad — 'how might we make Lakshmi happy?' — no traction, every idea is technically on topic so none is aimed. Too narrow — 'how might we fit a prayer shelf in the corner?' — a solution smuggled in, generating trivial variations. Just right — 'how might we give Lakshmi a space that feels her own, without her having to ask?' — tied to the real insight, open enough for many genuinely different solutions, focused enough to aim them. Test: a good HMW lets you generate five genuinely different ideas in a minute.

When two POVs collide

A real home has multiple personas with colliding POVs — Lakshmi needs a calm prayer corner, the teenager a lively study/music space, in the same room. Don't pick a winner; frame the collision as one both/and question: 'how might the same room be a calm retreat for Lakshmi's dawn prayers and a lively space for the teenager in the evening?' The time dimension from the day-in-the-life map dissolves it — Lakshmi prays at dawn, the teenager occupies the evening; a flexible, time-shifting room serves both. The HMW turns an either/or into a both/and, and the both/and is the most interesting design problem in the house.

Generate several HMWs from one POV

From 'belong without asking': how might we give her a space that's hers by default? make her feel like family rather than a guest? let her contribute without being in the way? design for her comfort so she never has to request help? Carry several into Ideate; let them generate from multiple directions.

How Might We reopens the closed problem into a generative door — each word does real work.

Go deeper — for practitioners & students

Vary the HMW's angle deliberately — amplify the good, remove the bad, question an assumption, or use the opposite ('how might we make a shared room feel private — must privacy mean a separate room at all?'). The unexpected angles hide original ideas. The HMW is the last cheap place to change the problem — spend real time getting it right. And resist resolving the collision in the question itself ('how might we add a folding screen...' pre-decides the answer); keep the conflict live and let Ideate find the resolution.

Try it

1. Convert your POV to one HMW (check all three words do their job). Test the altitude — can you generate five genuinely different solutions in a minute? Adjust until you can. Generate three more HMWs from the same POV, each a different angle. Then write one collision HMW holding two personas' conflicting needs together without pre-deciding the answer — and notice whether time dissolves the conflict.

Check yourself

3 quick questions — pick an answer to see why.

Q1What does each word in 'How Might We' contribute?

Q2An HMW like 'how might we make Lakshmi happy?' fails because it is:

Q3When two personas' POVs collide, the lesson recommends you:

Key terms

How Might We (HMW): A grammatical move that reopens a closed POV into an open, generative question, bridging Define to Ideate.
Scope (the Goldilocks problem): The craft of pitching an HMW neither too broad nor too narrow, so it can generate five genuinely different but related ideas.
Both/and question: An HMW that holds two colliding personas' needs together without choosing a winner, turning an either/or into the richest design problem.

Recap

A sharp POV is still closed, and statements invite judgement. The How Might We question reopens it into a generative door — how (it's solvable), might (permission to be wild), we (a shared act). The craft is scope: not too broad, not too narrow, just right enough to generate five different-but-related ideas. Best of all, the HMW handles colliding POVs — frame the collision as a both/and question (often dissolved by time). This is the last cheap place to change the problem.

Carry forward →

You have a beautifully framed open question. But when you step through it your mind hands you your first idea, and you'll want to grab it. Why is that first idea almost always a trap, and how do you generate the many before you choose the one?