Questions, answered with evidence.

An evidence-backed AI product-idea engine. We maintain rolling top-100 leaderboards of product opportunities across more than 40 business domains. Every opportunity is discovered from real market signals, backed by citations you can open, and scored on productizability and revenue potential.

One formula, applied to every opportunity. Each idea is scored on four pillars — Demand (30%), Monetization (30%), Buildability (20%) and Openness (20%). Each pillar is the average of four dimensions — sixteen in all — scored 0–10 with cited evidence, and a dimension whose evidence can't be verified has its score capped. Verification is scoped per dimension: a URL only lifts the cap for the claims its kind of evidence can actually prove. Popularity metrics never enter the formula: GitHub stars, for example, are only used as a provenance filter during discovery, not as a quality signal. The full method is on the Methodology page.

It's a comparison, not a certainty. The formula produces an internal raw score; the number you see is half the idea's percentile position on the board and half a direct map of that raw score onto 1–100, so the gap between two scores reflects both where they rank and how far apart they really are. Every score also carries a separate confidence value and band (low, moderate or high) that is deliberately never folded into it — a single number can't tell you the difference between a good idea with thin evidence and a weak idea with airtight evidence.

Yes — the Compare page plots any two of 36 axes against each other on an x–y chart, across one domain or several. The axes include the overall score, the four pillars, every scored dimension, and the PM metrics (payback months, LTV/CAC, weeks to MVP, competitor count and more). It's available on every plan over the domains your plan covers.

Continuously, on a rolling basis. Each discovery run sweeps the sources again, re-scores what it finds, and re-ranks the leaderboards. Nothing is ever deleted: ideas that fall off a leaderboard are archived and re-compete if the evidence resurfaces, and score history is kept as an append-only record.

The Free plan lets you follow a couple of domains and browse their leaderboards for opportunities scored up to 35 — including the full analysis on each visible opportunity (market and competitive analysis, technical feasibility, business viability, the complete scoring breakdown and a build roadmap). Paid plans (Personal and Pro) remove the score ceiling and let you follow more domains. Personal opens every domain's full top-100 board with no score ceiling, plus saved-search alerts, CSV/JSON exports and monthly quotas of deep-research reports, MVP build prompts and build playbooks; Pro raises every quota and adds read-only developer API access. Current limits and prices are on the Pricing page.

No. The Free plan needs no credit card — create an account and you can browse evidence-backed opportunities in minutes.

Yes. Once signed in, use the My Ideas page to submit an idea. It is first checked against the existing leaderboards for duplicates; if it's new, it goes through the same AI evaluation pipeline and scoring formula as every discovered opportunity, and you can track its status from My Ideas throughout.

Billing is region-aware: customers in India are billed in INR, everywhere else in USD. The Pricing page shows the prices that apply to you.

Email support@aiproductopportunity.com, or use the contact form once you're signed in.