Identifying Real Problems
Find problems worth solving through systematic research
Your Progress
Section 2 of 5How to Find Problems
Problems are everywhere—you just need to know where to look and what to listen for. The best problems come from direct user contact combined with behavioral data.
Use multiple sources. No single method reveals the full picture. Triangulate insights from conversations, behavior, and market signals.
Problem Discovery Sources
Where to Find Problems
🎤User Interviews
Direct conversations with users
- •Repeated frustrations
- •Time-consuming tasks
- •Workarounds they built
- •Things they avoid
Ask "Tell me about the last time you..." and listen for pain points
Every week I spend 3 hours copying data between systems. I hate it but there is no other way.
Signs of a Real Problem
✓ Good Signs
- • Users mention it without prompting
- • They've built workarounds
- • It costs them time or money
- • Multiple users have the same problem
- • They express urgency or emotion
✗ Warning Signs
- • Only mentioned when you ask directly
- • "Nice to have" language
- • No current workaround attempted
- • Each user describes it differently
- • Theoretical "would be cool if..."
Qualifying Problems
Problem Qualification Framework
Not all problems are worth solving. Evaluate problems on three dimensions:
Users find it hard to remember their passwords
How often it happens
How painful it is
Monetization potential
Frequent but low willingness to pay (password managers are free/cheap)
Quality Over Quantity
Better to deeply understand 5 users' problems than superficially collect feedback from 100. Go deep, ask follow-ups, and understand the context and impact of each problem.
Key Takeaways
- •Use multiple sources: user interviews, support tickets, analytics, sales calls, and social listening.
- •Look for frequency, intensity, and willingness to pay—all three should be high for strong problems.
- •Real problems surface unprompted, have existing workarounds, and show urgency from users.
- •Depth beats breadth—deeply understand a few users rather than superficially survey many.