The validation pipeline

The validation pipeline

Every great product starts with a guess. idea_builder’s job is to turn that guess into evidence — cheaply, before you spend months building. The platform is built around a single five-stage loop that takes an idea from spark to decision:

Research → Score → Validate → Compare → Decide

The five stages at a glance

Stage	What it does	Where you see it
Research	Collects evidence about the market, the audience, and the competition	The Research section of each idea
Score	Turns that evidence into numbers you can act on	The PMF score and Scorecards pages
Validate	Puts your riskiest assumptions in front of real people	The Validate section: surveys, discovery, fake doors
Compare	Lines your ideas up side-by-side so you can be honest about trade-offs	The Comparison dashboard
Decide	Commits to go, no-go, or iterate — and keeps your pipeline honest	The Decide section and the idea status workflow

How the stages feed each other

The loop is deliberately sequential, but you can jump in at any point and loop back as many times as you need.

1. Research feeds Score. Every research run — trends, market sizing, competitor analysis, and the rest — produces structured evidence. The scoring engines read that evidence automatically: your PMF score picks up proxy signals like search demand and market growth, and your scorecards give you a weighted view of which idea is strongest.
2. Score points to Validate. When a score is middling or built on assumptions, the validation tools let you test the specific assumption that matters most. Write a survey about willingness to pay, run a customer discovery interview, or publish a fake-door page and see if anyone clicks and pays.
3. Validate sharpens Score. Real responses flow back into your PMF score through the automatic bridge, so your numbers reflect what actual people told you — not just what the research suggested.
4. Compare forces the decision. With several ideas researched and scored, the comparison dashboard puts them side-by-side so the strongest one becomes obvious.
5. Decide and move on. Commit to a path, update the idea’s status, and export the evidence you need — for your team, your investors, or yourself.

Where each stage lives in the app

Everything hangs off an individual idea. Open any idea and you’ll see the pipeline as the main sections of that idea’s page:

• Research — run research phases and AI research agents, review the results, and set up recurring monitoring.
• Score — see your PMF score and the proxy signals behind it, plus weighted scorecards and prioritization views.
• Validate — surveys, audience panels, customer discovery, fake-door tests, and landing pages.
• Compare — the side-by-side dashboard across all your ideas.
• Decide — the go/no-go toolkit and exporting your work.

Tips for getting the most out of the loop

• Don’t wait for perfect research. Run the phases that answer your riskiest question first, then validate that specific answer.
• Validate the killer assumption. If your whole idea hinges on “people will pay $29/month,” a fake-door test with a real payment button answers that better than another research report.
• Loop, don’t linear. Founders who go back and forth between Score and Validate make better decisions than those who force a one-way march.
• Use the status workflow to stay honest. Move each idea from draft to researching to validating and beyond — it stops dead ideas from quietly eating your time.

Tip

You don’t have to run every stage on every idea. Use the loop to spend the least amount of effort needed to kill a bad idea — or gain confidence in a good one.

Next up: understand what research runs are and start collecting evidence.