IdeaScroll

After 7 half-built projects in 6 months (Maybe More), I finally built something I think could help me focus down which to actually put full effort into.🤦‍♂️

Hey IH! 👋

The brutal truth: AI made me a serial project starter, not a finisher. it's bad...

In the last 6 months, I started 7 different projects. A productivity app (abandoned after 2 weeks), an ai confidence builder mobile app (got bored after the MVP), a budgeting tool (realized I hate budgeting GO FIGURE), and 4 others I can barely remember. The pattern was always the same: exciting idea → rapid prototyping with AI → lose interest → move to the next shiny object.

Newsflash: Regardless of what the new AI gurus tell you, It doesn't take a weekend to build a great product, even with AI. But AI makes starting feel so effortless that I confused "easy to start" with "worth finishing."

The $0 Revenue Reality Check

After 6 months and exactly $0 in revenue from any project, I had to ask myself: Am I building things people want, or just things AI makes possible?

That's when I built IdeaScroll.io - not because I thought it would be successful, but because I desperately needed to stop building random stuff. It's a community-driven startup idea validation platform where entrepreneurs get real feedback before wasting months building. They get early feedback, launchlists, pricing discovery and more. AND ITS FREE. (With some paid options, wantrepreneurs gotta eat too.)

Here's What Actually Took 4 Months (Not a Weekend)

My Stack Reality Check:

• Bolt.new: Perfect for rapid design iteration (don't trust it for auth or complex logic. but it gives you a start.)

• Roo Code: My go-to for complex architecture decisions and planning. its the bees knees of vibe coding. not gonna lie. if you're not using it. then use. roocode.com thank me later.

• VSCode: Still irreplaceable. gotta love the OG. Contrary to most people. i hate cursor. SORRY.

• Supabase: I mean come on. what more could you ask for. USE IT.

Timeline Reality: 4 months of actual work even with AI. buttons break, database queries break. big projects that can be something break. so it took me time. but im finally at a launchable point in the build where i can start getting some beta first users. so more things can break and i can fix them along the way.

The Validation Framework I Wish I Had Earlier

After studying 20+ failed side projects (mine and others), here's the pattern I see:

1. Excitement Phase: "This AI tool makes building so easy! I can build anything!"

2. Building Phase: Actually fun and rewarding

3. Launch Phase: "Crickets... why isn't anyone using this?"

4. Reality Phase: You solved a problem that didn't exist

IdeaScroll tries to catch ideas in Phase 1, before you fall in love with building them.

What's Different About Community Validation

Most validation tools give you AI analysis or surveys. But AI doesn't know if your B2B SaaS idea sucks, and surveys can't capture the nuance of "this solves a problem, but not one I'd pay for."

What works: Real users voting on your idea to build it or not. telling you how much theyd pay for it, what they're currently using to solve their problem, and what they hate about their current solution. Plus give you their email to join the launch list after its built.

The Features That Actually Matter

• Voting/scoring system: See which ideas the community thinks are worth pursuing

• Category filtering: SaaS, e-commerce, etc. (because context matters)

• Real feedback: Not AI analysis, actual human insight from people who've been there or WANT YOUR PRODUCT AND TELL YOU HOW MUCH THEYD PAY! Plus you build a launchlist.

• Free to use: No credits, no paywalls, just community value

I'm Launching With Bugs (And Here's Why)

There are still bugs. The UI isn't perfect. Some features are rough around the edges. But waiting for "perfect" was exactly the trap that kept my previous 7 projects in development hell.

Better to have 100 people using something imperfect than 0 people using something polished.

Try It (And Break It) For Me

Check out ideascroll.io and please:

• Share an idea you're considering building

• Vote on ideas that resonate with you

• Tell me what's broken (it's definitely something made it easy with a feedback widget in the bottom right corner)

• Roast my validation framework if it's wrong

Specific Questions for Fellow Serial Starters

1. How many projects have you started vs. finished in the last year?

2. Do you validate ideas before building or during/after building?

3. Would you trust community feedback over your own gut feeling about an idea?

4. What would make you actually use an idea validation platform consistently?