🔧

Validate Idea Before Building

A systematic approach to validating your startup or product idea before investing time building. Avoid building something nobody wants.

4phases•20tasks•~2-4 weeks

Phases & Tasks

Talk to 10 potential users
10h
Interviews, not surveys. Listen for pain points. Don't pitch your solution.
Identify if problem is painful enough to pay for
1h
Are they actively looking for solutions? Have they paid for alternatives?
Document exact quotes and patterns
1h 30m
Write down what they said verbatim. Look for repeated themes.
Find where these people hang out online
1h
Subreddits, Twitter hashtags, Discord servers, forums. You'll need this for distribution.
Quantify: how often does problem occur?
30m
Daily problem > weekly > monthly. Frequency affects willingness to pay.

Describe your solution in one sentence
30m
If you can't, simplify. "X that does Y for Z" format.
Get reactions from interviewees
2h
Show them mockups or describe the solution. Watch their reaction.
Check for existing solutions (competitors)
1h 30m
Google, Product Hunt, G2, Capterra. Who else solves this?
Identify why existing solutions fail
1h
Too expensive? Too complex? Missing features? This is your angle.
Define your unique angle
45m
Why will you win? Simpler? Cheaper? Better for specific use case?

Create landing page with value prop
2h
Simple page: problem, solution, signup/waitlist form.
Drive traffic from target channels
3h
Post where your users hang out. Small budget ads if needed.
Measure signup/interest rate (> 5% good)
30m
Track conversions. 5% landing page conversion is decent. 10%+ is great.
Collect emails or pre-orders
1h
Email is interest. Pre-order is stronger signal. Try both.
A/B test messaging if needed
2h
Try different headlines, value props. See what resonates.

Review all evidence honestly
1h
Look at interviews, landing page data, competitor research. Be honest.
Calculate: can you reach these users affordably?
45m
What's the CAC? Do the unit economics work at scale?
Decide: build, pivot, or kill
30m
Strong signals → build. Mixed signals → pivot. Weak signals → kill.
If building: start with smallest possible version
1h
What's the 1-week version? Ship that first.
If killing: document learnings, move on
45m
Write down what you learned. Don't sink-cost yourself. Next idea.

Ready to start this playbook?