Stage 2 / Discover

Find what to build.

Most first-time builders start with a technology and go looking for a problem. The Discover stage flips this. Start with the problem, find the person, then build.

“Without going outside, you may know the whole world. Without looking through the window, you may see the ways of heaven.” The Way of Code, 47 — Rick Rubin, after Lao Tzu
The Five Questions

Before you write a single line of code, answer these five questions. They take twenty minutes and save you twenty weekends.

01
What problem do you see?
Not "what app do you want to build." What problem have you personally witnessed? A lawyer who spends 4 hours drafting the same notice. A pool company losing leads to a broken form. If you cannot name a specific person with a specific problem, go talk to people first.
02
Who has this problem?
Not "small businesses." Name the person. Job title, tools they use, how much they spend today, where they hang out online. A product for "lawyers" will fail. A product for "solo practitioners in Florida who handle residential evictions" can work on day one.
03
How do they solve it today?
Every problem already has a solution — spreadsheets, paper, manual email, hiring a person. You need to know what you are replacing. If the current solution costs $500/month in time or money, you can charge $49 and they will say yes immediately.
04
What does 10x better look like?
Not 2x. What would make someone switch without hesitation? Usually it is speed (4 hours becomes 4 minutes), cost ($500/month becomes $49), or capability (what was impossible is now trivially easy). Claude Code + Cloudflare gives you all three.
05
Would they pay for it?
Before you build, ask 5 people: "If I built a tool that did X, would you pay $Y/month?" The answers will surprise you. Some problems people happily live with. Others they would pay to solve right now. You want the second kind.
Distribution channels

Direct outreach is the most underrated channel. Send 10 messages to people who have the problem. You will learn more in 2 days than 2 months of building in isolation.

ChannelBest forTime to signalCost
Direct outreachB2B, services1–3 daysFree
LinkedInProfessional SaaS1–2 weeksFree
Reddit / communitiesNiche tools1 weekFree
Twitter / XDev tools, AI2–4 weeksFree
Product HuntConsumer, dev tools1 day (launch)Free
Cold emailB2B, specific verticals3–7 daysFree
Paid adsValidated product only1–2 weeks$100+
Your output: a product brief

After answering the five questions, write a product brief. One page. This is what you hand to Claude Code when you build.

# Product Brief ## Problem [One sentence. Who has the problem and what it costs them.] ## Solution [One sentence. What your product does.] ## Target Customer [Specific person: title, industry, size, pain level.] ## Current Alternative [What they use today and why it fails.] ## Price [Anchor: what they pay now. Your price: what you will charge.] ## Distribution [Top 3 channels to reach this person.] ## Success Metric [How you know it is working. Usually: paying customers.]
Choose your path
I have an idea
Answer the five questions above, then scan your product to get ranked channels and a launch sequence.
I've already built something
Skip straight to the scanner. Describe your product and selling style to get a distribution strategy.