Most candidates fail a product manager problem-solving case not because they lack the right answer, but because they jump straight to solutions before they understand the problem. Interviewers are watching how you think. This post gives you a repeatable five-step structure you can apply to almost any problem-solving prompt, so you always have a starting move.
1. Clarify the goal before touching the problem
Prompts are deliberately ambiguous. Spend the first minute narrowing scope: who is the user, what outcome matters, and what does “better” actually mean here? Ask one or two sharp clarifying questions, then state the goal back in your own words. This alone separates strong candidates from the rest.
2. Frame the problem out loud
Before solving, lay out how you'll break the problem down. A simple frame — segments, drivers, or a funnel — signals structure. Say it explicitly: “I'll look at this in three parts.” The interviewer can now follow you, and you've given yourself a map.
Structure is the product you're actually being tested on. The recommendation is secondary to how you got there.
3. Explore, don't guess
Walk through your frame one branch at a time. State assumptions as you go and flag which ones you'd validate with data. It's fine to be wrong here — what matters is that your reasoning is legible and you're not pattern-matching to a memorised answer.
Common branches worth exploring
- User segments and where their needs diverge.
- The funnel or journey, and where drop-off concentrates.
- Trade-offs between effort, impact, and risk.
4. Decide and commit
Ambiguity is not an excuse to avoid a recommendation. Pick a direction, say why, and name what you're trading off. A clear, defensible “here's what I'd do and why” beats an exhaustive list of options with no point of view.
5. Pressure-test your answer
Close by stress-testing yourself before the interviewer does: what would change your mind, what's the biggest risk, and how would you measure whether the decision worked? This shows maturity and buys you enormous credibility.
See the Real Revolut Problem Solving Interview Case
Here's a full Revolut PM problem solving case, worked through end to end with the MECE framework — free.
Get Real Case for FREE →The takeaway
You don't need a different approach for every case — you need one structure you trust under pressure. Clarify, frame, explore, decide, pressure-test. Practise it until it's automatic, and the ambiguity stops being scary.