Most problem-solving starts with the wrong problem

Imagine walking into a doctor's office. You tell them:

"I don't feel well."

The doctor immediately prescribes medication.

No questions.
No tests.
No diagnosis.

You'd probably walk out. Not because the doctor isn't smart. Because they're solving a problem they don't yet understand.

Oddly enough, that's exactly how many of us approach work: a project is behind schedule, sales are slowing, and customer complaints are increasing.

Someone asks: "What's the solution?"

And the room immediately starts generating answers. The irony? The better your team is at brainstorming, the faster you'll arrive at solutions for the wrong problem.

One of the biggest misconceptions about AI is that it helps us solve problems. Not quite. AI helps us solve the problem we give it. If the problem is poorly defined, AI won't magically fix it. It will simply produce better answers to the wrong question. That's why the quality of the solution depends on the quality of the problem definition.

Instead of asking:

How do we fix this?

they ask:

• Who is affected?
• What exactly can't they do?
• What's preventing them?
• What's the real impact?

Sometimes just forcing yourself to express the problem in a single sentence changes everything. Frameworks like the 4 Ws, Jobs to be Done, or User-Need-Insight aren't paperwork, they're ways to expose assumptions before you invest time solving the wrong thing.

Imagine trying to solve a thousand-piece puzzle… without separating the edge pieces first. That's what complex business problems feel like. Experienced consultants rarely attack the whole problem at once. They break it into smaller, independent pieces.

One useful way is the MECE principle: divide the problem into categories that don't overlap but collectively cover the whole picture. Suddenly, one overwhelming challenge becomes a handful of manageable questions.

Once we've broken down a problem, our instinct is to grab the first plausible solution. But good problem solvers deliberately generate alternatives before choosing one. Think of it as exploring several trails before committing to the hike.

Techniques like Tree of Thought encourage divergent thinking: create multiple hypotheses, compare them, and only then decide which path deserves your attention.

This is where AI becomes genuinely valuable. Not because it gives you answers, because it asks questions you forgot to ask.

After defining a problem, ask AI:

What assumptions am I making?

After creating categories:

Have I missed anything?

After generating ideas:

What alternatives haven't I considered?

Instead of acting as an answer machine, AI becomes a thinking partner that exposes blind spots and strengthens your reasoning.

Many teams celebrate when the action plan is complete. The best teams celebrate when they've learned something. After acting, measure the results. Compare them to your original problem definition.

Then ask: 

Did we solve the right problem? Or did we simply execute a well-designed plan for the wrong one?

That final reflection is where real improvement happens, and it's where AI can help you inspect your process and refine it for the next challenge.

AI is incredibly good at accelerating work. But speed only matters if you're heading in the right direction.

The best problem solvers don't start with solutions, they start with understanding. Because once the problem becomes clear, the solution often becomes much clearer too.

If you want AI to become more than just another productivity tool, learn how to make it a thinking partner. Our Collaborating with AI Agents training will help you collaborate with AI to think more clearly, make better decisions, and solve problems with greater confidence.