Study · Foundations & Mindset · principle
Find the right problem
Most bad designs solve the wrong problem well. Ask why until you reach a cause you can change.
Built with Claude + NotebookLM · Updated · 3 min read
In short
A brief names a symptom. The thing you can change sits two or three "whys" below it. Before you design anything, ask why until the answer is a decision someone made — that is where you can act. Then rewrite the brief in one line and check it with whoever gave it to you.
When to use
- The brief arrives as a feature: "add a filter", "make the button bigger".
- The same complaint comes back after the fix shipped.
- Stakeholders disagree about the solution but agree about the symptom.
How
London, 1854. Cholera was killing people in Soho, and the doctors treating them were solving the problem in front of them, one patient at a time. John Snow asked a different question. He put every death on a map, saw them cluster around one water pump on Broad Street, and had its handle removed. The outbreak ended. The patients were the symptom; the pump was the problem.
Don Norman calls this finding the right problem · tìm đúng vấn đề, and the everyday tool for it is the 5 Whys · phương pháp 5 Tại sao:
- Write the complaint as the person says it. Not your translation of it.
- Ask why, and answer with a fact. Something you observed or can check, never a guess.
- Repeat until the answer is a decision. Usually three to five rounds. A decision is a place where someone chose, so someone can choose again.
- Rewrite the brief in one line. Show it to the person who wrote the first one.
- People leave the cart at the last step of checkout.symptom
- The shipping cost appears only on the last screen.
- The cost needs an address, and the address is asked for last.
- The form order was copied from the desktop site.
- The desktop flow was built for returning customers with a saved address.
- Nobody re-decided the order for a first-time buyer on a phone.root cause
Example
In the Baymard Institute's checkout studies, unexpected extra costs sit at the top of the reasons people abandon a cart. The brief that follows is usually "make the Pay button more prominent". Ask why people leave: the shipping cost shows up on the last screen. Why there? It needs an address, and the address comes last. Why last? The order was copied from the desktop site. The rewritten brief is "show the cost before asking for the address". The button never changed.

Traps
- Stopping at a person. "Because the developer did X" is not a root; the condition that let X happen is. Keep asking.
- Answering with guesses. One invented rung and the ladder points at the wrong pump. If you cannot check an answer, mark it as a hypothesis and go find out.
- Reframing alone. A brief you rewrote in private is a problem only you own. The person who gave it has to recognize the new frame.
Try this
Take the last brief you received. Write its first sentence, then ask why four times, in writing. Show the fourth answer to whoever wrote the brief.
Map
How
- Write the complaint in their words
- Ask why; answer with a fact
- Stop when the answer is a decision
- Rewrite the brief in one line
Try
- Ask why four times, in writing
When
- The brief arrives as a feature
- The complaint comes back after the fix
- Everyone agrees on the symptom, not the solution
Traps
- Stopping at a person
- Answering with guesses
- Reframing alone
Sources
- 01Human-Centered Design Considered Harmful (opens in a new tab) — Don Norman, 2005
- 02John Snow and the Broad Street pump (opens in a new tab) — UCLA Department of Epidemiology
- 03Toyota Production System (opens in a new tab) — Toyota Motor Corporation
- 04Cart abandonment rate statistics (opens in a new tab) — Baymard Institute
Related
- Shift work to the system
Every step the system takes is one the person doesn't. Count the steps, then move them.