Marcus Andrey Study
Ler em português Library

UX Foundations · Step 2 · 4 min read

Problem statements and "How might we"

Before you design, write down the problem: who has it, what they need, why it matters, without a solution hidden inside. How to write a problem statement, turn it into "How might we" questions that open up ideas, and check it with your team.

Five objects on a board, five ways to frame a design problem: the problem itself, the need, the people, the journey and how you'll measure success. Point at one and a tray slides under it to frame it. Illustration: Marcus Andrey, drawn with Hairline (MIT, Lucas Marques).

Most design arguments are really arguments about the problem. One person is solving "the checkout is ugly", another "people abandon their basket", a third "we need more sales". Each design looks wrong to the others because they're answering different questions. Writing the problem down, in one sentence everyone agrees with, is the cheapest design tool there is.

What you'll be able to do after reading:

  • Write a problem statement with no solution inside it
  • Turn it into "How might we" questions that open up ideas
  • Tell a good problem statement from a disguised solution
  • Check the statement with your team before designing

1. What a problem statement is

NN/g defines it simply: "A concise description of the problem that needs to be solved" (NN/g, Rosala, Problem Statements in UX Discovery, 2021). A good one says who has the problem, what they're trying to do, what gets in the way, and why it matters.

1Who

A specific group of people.

First-time buyers on phones
2Trying to

What they want to get done.

Finish a purchase
3But

What gets in the way.

Can't see the total cost
4Which matters because

The cost for them and for the business.

They give up at the delivery step
Four parts of a problem statement. None of them is a solution. Diagram: Marcus Andrey, CC BY 4.0.

Stanford's d.school teaches a similar "point of view": combine "user, need, and insight" into "an actionable problem statement" (d.school, An Introduction to Design Thinking: Process Guide).

A triangular road sign with a large question mark, against a blue sky.
A good problem statement works like a sign before a junction: it makes everyone stop and ask the same question before they choose a road. Photo: Colin Kinner, Wikimedia Commons, CC BY 2.0.

2. Solutions in disguise

The most common mistake is a solution dressed as a problem:

Disguised solution Problem statement
"We need a chatbot for support" "Customers wait two days for answers to simple questions about their order"
"Add a dark mode" "People who use the app at night find the bright screen tiring"
"Redesign the dashboard" "Managers can't see which team is behind schedule without opening five reports"

If your statement contains a feature, a screen or a technology, take it out and ask "why would that help?" The answer is usually the real problem.


3. "How might we" questions

A "How might we" question turns the problem into an invitation to ideas. NN/g traces the phrase to Procter & Gamble in the 1970s, later adopted by IDEO (NN/g, Rosala, How Might We Questions, 2021). Harvard Business Review tells the story: Min Basadur's team at P&G, trying to compete with a rival soap, asked "How might we create a more refreshing soap of our own?", and the answer became a new brand (HBR, Berger, 2012).

The three words each matter: how assumes it can be solved, might allows ideas that may not work, we makes it shared. See How Might We.

Good HMWs are neither too wide nor too narrow:

Too wide Just right Too narrow
How might we improve shopping? How might we show first-time buyers the total cost before the delivery step? How might we add a cost banner to the cart?

4. Check it with the team

Before designing, share the statement and ask three questions: Is this the problem we agreed to solve? Do we have evidence it's real? Would solving it matter to users and to the business? If anyone hesitates, you've just saved weeks.


5. Common mistakes

  • A feature hidden in the statement.
  • No specific user ("users", "people", "everyone").
  • A statement nobody checked with evidence.
  • HMW questions so wide that any idea fits, or so narrow they dictate one.

6. Vocabulary

Term Meaning
Problem statement A short description of who has a problem, what gets in the way, and why it matters
Point of view (POV) The d.school's version: user, need and insight in one sentence
How might we (HMW) A question that turns a problem into an invitation for ideas
Reframing Rewriting a problem from another angle to find better solutions

7. Try it (no tools needed)

  1. Take a request you heard recently ("we need a new filter", "add notifications").
  2. Ask "why?" until you reach the user's problem. Write it in one sentence: who, what they're trying to do, what's in the way, why it matters.
  3. Write three HMW questions from it: one too wide, one too narrow, one just right.

Reflection questions: Did the real problem surprise you? Who in your team would describe it differently?


8. If you already practise

  • Put the problem statement at the top of every design file and critique deck; refer back to it when feedback drifts.
  • Write two or three competing problem statements for the same situation and test which one the evidence supports best.
  • Gause and Weinberg's Are Your Lights On? is a short, funny classic on what a problem really is.

Checklist

Learn more

Books, courses and videos to go deeper

Book links go to Amazon. As an Amazon Associate I earn from qualifying purchases, at no extra cost to you.

User needs and jobs to be done · Defining success · How Might We · The design process · Assumption mapping

Saved only on this device.

↑ ↓ to move · Enter to open · Esc to close