Understanding the problem

What is design thinking?

20 min

Most founders start with a solution they already believe in. Design thinking flips that around: you start with the people you want to help, and let their problem decide what you build. It is a way of working, not a recipe — a way of thinking that makes you spend your time and money on what actually matters.

The five phases

Design thinking is often described in five phases. They are not a staircase you climb once, but a loop you move back and forth through.

  • Empathize: You get to know the people you want to help. You observe and listen instead of assuming.
  • Define: You gather what you have learned into one clear problem statement. What is the real problem, and for whom?
  • Ideate: You open up and generate many possible solutions, without judging them right away.
  • Prototype: You make a simple, cheap version of an idea — just enough to show and learn from.
  • Test: You put the prototype into the hands of real users and watch what happens.

The point is not to do the phases in the "right" order, but to keep switching between understanding and trying. Often you discover during testing that you understood the problem wrong, so you loop back to empathy. That is not failing — that is the whole idea.

Why you start with the problem

Picture Maria, who wants to build an app for dog owners. She has already imagined the features: a walk log, a map and a social feed. But she does not really know what bothers dog owners day to day. Maybe the biggest problem is finding a trustworthy dog sitter during holidays — something her app does not solve at all.

When you start with the problem, you build something people need. When you start with the solution, you risk building something elegant that nobody asked for. A well-framed problem is worth more than a clever solution, because the problem points you toward many possible solutions — while a solution you have fallen for locks you into just one.

Divergent and convergent thinking

Design thinking alternates between two mindsets:

  • Divergent thinking opens up. You explore widely, collect many impressions and generate many ideas. Anything goes, and nothing is discarded yet.
  • Convergent thinking narrows down. You choose, prioritize and decide.

A common model calls this the "double diamond": first you open up to understand the problem, then narrow to one problem statement. Then you open up again to find solutions, before narrowing to what you will actually build. The mistake many make is jumping straight to narrowing — they decide too early, about both the problem and the solution, and miss the good ideas that sit a little further out.

The most common trap

The biggest trap in the early stage is falling in love with your own solution. You have spent evenings and weekends on the idea, told friends about it and maybe bought the domain. Then it hurts to hear that it does not land — so you unconsciously start looking for confirmation instead of truth.

Jonas, who is building an accounting tool for small firms, asked his first customers: "Wouldn't this be useful?" Everyone said yes, out of politeness. He built for six months before discovering that nobody would pay. Had he instead asked "How do you solve this today?" and "What is the worst part of that solution?", he would have gotten honest answers much sooner.

The cure is easy to say and hard to live by: hold on to the problem for a long time, and hold the solution loosely. Be willing to be wrong early, while it is still cheap.

Do this now

Write your idea down in two sentences. Then split a sheet of paper into two columns. In the left column write "The problem I think I solve." In the right column write "What I actually know about this problem — from real people, not from myself." If the right column is empty or thin, you have just found the most important thing to work on in the next lessons. Keep the sheet; you will use it throughout the course.

What you'll learn in this lesson

  • The five phases of design thinking
  • Why you start with the problem, not the solution
  • Divergent versus convergent thinking
  • Not falling in love with your own solution

Stay updated

Subscribe to our newsletter for news about open source, AI, and digital innovation.