Choose between two design directions without just picking your favorite
Liking one better and solving the actual problem are different questions.
- Use it for
- Designers and PMs stuck between two genuinely different directions.
Choosing between two design directions often collapses into personal taste dressed up as a decision — "I just like this one better" — because nobody explicitly reconnected the options to the actual problem being solved. Both directions might be well-executed and still trade off differently against what actually matters (learnability vs speed, flexibility vs simplicity), and that tradeoff, not aesthetic preference, is what should decide it.
This prompt forces the comparison back to the actual problem before letting taste weigh in at all.
When not to use this
For a decision that's genuinely just aesthetic (brand color, illustration style) with no functional tradeoff behind it, this framework adds nothing — trust taste there. This is for decisions where the two directions actually behave differently for users.
Did this work?
You can articulate the specific tradeoff between the two directions in terms of the actual problem, not just aesthetic preference — and know what would make one clearly right.
Tested on claude-opus-5. Evidence status is draft; it moves to battle-tested only on recorded runs, never by hand.