← Tools

Problem-Solution Fit Canvas

Before you build anything, define the problem.

This canvas walks you through understanding your problem deeply before proposing a solution. Work through the sections in order. The solution fields are intentionally at the bottom - don't jump there first.

⚠ Rule: Do not fill in "Proposed Solution" until every section above it is complete.
0 of 9 sections filled
✓

Canvas complete - well done.

Run your idea through the Adversarial Evaluator before building your prompt. It will catch gaps you haven't noticed yet.

Zone 1 - Understand the Problem
The Problem
Describe the problem in one or two sentences. Be specific. "Things are slow" is not a problem. "Our sales team manually copies data from emails into our CRM three times a day" is a problem.
Who Feels It
Who specifically experiences this problem? What is their role? How often do they encounter it?
Pain level: ★ ★ ★ ★ ★
When Does It Happen
What triggers this problem? Is it a recurring event, a specific process step, or a particular situation?
Current Workaround
What do people do right now to deal with this? (Hint: if the answer is "nothing" - this problem may not be painful enough to build for.)
The Real Cost
What does this problem actually cost - in time, money, errors, frustration, or missed opportunities? Be as specific as you can.
Ideal Outcome
If this problem were perfectly solved, what would be different? Describe the outcome - not the tool. What would people be doing (or not doing) that they aren't doing now?
Zone 2 - Design the Solution
Proposed Solution
Now you can describe what you want to build. What should the tool do? Describe its core function in 2–3 sentences.
Minimum Viable Version
What is the smallest version of this that would prove the idea works? If your solution has 5 features - what is the 1 feature that matters most?
What It Must NOT Do
What would make this solution unacceptable or untrustworthy? What scope is out of bounds for this prototype?
Zone 3 - Define Success
How Would You Know It's Working?
What specific, observable change would tell you this tool is succeeding? Avoid vague answers like "people like it" - describe what you would actually see or measure.
Saved