How do you validate a startup idea?

Validating a startup idea means finding evidence that real people have the problem you're solving — before you build anything. Most failed startups built products nobody wanted. Validation is how you avoid that.

1. Check community signal first

Before talking to anyone, search for your problem on Hacker News, Reddit, and niche forums. Look for threads where people complain about the exact pain you're targeting. If founder communities have been debating your idea for years, the problem is real. If you can't find any signal, that's information too.

IdeaCheck automates this: it runs semantic search over 28,000 HN discussions and surfaces the most relevant threads in seconds — giving you a scored verdict on market demand, technical complexity, distribution, and timing.

2. Do 10 customer interviews

Talk to 10 people who match your target customer. Don't pitch — listen. Ask about the last time they faced the problem. Ask what they did about it. Ask what it cost them. If they've built a workaround, the problem is serious enough to pay for a solution.

Red flag: "I'd definitely use that" without any evidence they currently solve the problem. Green flag: "I've tried three tools and none of them work" or "we spend 8 hours a week on this manually."

3. Run a smoke test

Build a landing page that describes the product. Don't build the product yet. Drive traffic via Google Ads, Reddit, or a community post. Measure how many people click "sign up" or "get early access." A 2–5% conversion on cold traffic is a signal worth pursuing.

4. Map the competition honestly

Competitors are good — they prove the market exists. The question is whether you have a reason to win a segment. Search Hacker News for why previous attempts at your idea failed. Those failure post-mortems are the most valuable research you can do.

5. Only then, write code

If community signal is positive, 7+ of 10 interviews confirm the pain, and your smoke test converts — you have enough signal to start building a minimal version. Even then, get your first paying customer before building feature 2.