We've all been there. You're in the shower, a brilliant idea hits you. "A task manager for people who hate task managers!" You spend three months building it. You launch. Crickets. Two people sign up and one is your mom.
The fix is embarrassingly simple: validate before you build. Here's the framework I use to test ideas in a single weekend. It costs $0 and has saved me from at least five "brilliant" ideas that nobody wanted.
Saturday Morning: The Problem Sniff Test
Before anything else, answer these three questions honestly:
- Who has this problem? Name a specific person or community. "Everyone" is not an answer.
- How are they solving it now? If no one's even trying to solve it, the problem might not exist.
- Why would they switch to your solution? "Because mine is better" is not compelling enough.
If you can't answer all three, stop. The idea needs more cooking.
Saturday Afternoon: The Reddit & Forum Dive
Go where your target users complain. Search Reddit, Hacker News, Indie Hackers, and niche forums for your problem keywords:
- Search for the problem, not your solution
- Look for phrases like "I wish there was…" or "Does anyone know a tool that…"
- Count the upvotes and replies on those posts
- Note what existing solutions people mention (and their complaints about them)
If you find 5+ posts with 20+ upvotes asking for something like your idea, you might be onto something. If you find zero? Either the problem is too niche or it doesn't exist.
Saturday Evening: The Landing Page Test
Build a one-page landing site. You don't need your app — you need a page that explains what it does and has an email signup form. That's it.
Use your existing static site skills. One HTML file, a headline, three bullet points, and a Formspree/Web3Forms email capture. Total build time: 1-2 hours.
The page should answer:
- What is it? (One sentence)
- Who is it for? (Specific person)
- What's the main benefit? (One sentence)
- "Notify me when it launches" (Email capture)
Sunday Morning: Get Eyeballs
Post your landing page where your target audience hangs out:
- Reddit: Relevant subreddits (follow their self-promo rules!)
- Hacker News: Show HN post
- Indie Hackers: Product launch section
- Twitter/X: Quote-tweet someone discussing the problem
- Niche Discord servers: If appropriate
Don't say "check out my product." Say "I'm thinking about building X because I noticed Y problem. Would this help you?" Authentic curiosity beats marketing every time.
Sunday Evening: Read the Signals
After 24-48 hours, evaluate the response:
| Signal | What It Means |
|---|---|
| 10+ email signups | Strong interest. Build an MVP. |
| 3-9 signups | Mild interest. Refine the pitch, test again. |
| 0-2 signups | Weak signal. Pivot or abandon. |
| Comments: "I'd pay for this!" | Gold. Build it immediately. |
| Comments: "Cool idea" | Polite rejection. They won't use it. |
| Dead silence | The idea is invisible. The problem isn't felt. |
The Counter-Intuitive Truth
The best outcome of this weekend is discovering your idea is bad. That's not failure — that's saving yourself 3 months of building something no one will use. The worst outcome is a "maybe" signal that keeps you in limbo.
Ideas are free. Time is not. Spend your weekends validating, and your months building things that already have an audience waiting.