Methodology

A custom scorecard for every idea, built from the idea itself.

IdeaForge starts by understanding what kind of decision the idea creates. Then it builds a scoring surface around that decision instead of forcing the idea through a static startup checklist.

Idea fingerprint

The engine extracts market, buyer, workflow, novelty, channel, timing, and defensibility signals from the submitted idea before any scoring starts.

Adaptive dimension set

Dimensions are selected and generated around that fingerprint, so a consumer marketplace and a biotech workflow tool are not judged by the same rubric.

Weighted pressure test

Each dimension receives a score, confidence level, and reason. Risk-heavy dimensions can drag the verdict down even when the concept sounds exciting.

Decision translation

The report turns raw scores into a Dimension Signature, rival ideas, risks, action levers, and a clear build, refine, or kill posture.

What the engine looks for

The visible report is a small slice of the internal signal map. The important part is not that an idea gets a number. The important part is which signals made that number move.

Market urgencyBuyer pain intensityDistribution frictionDifferentiation clarityRetention mechanicsData moat potentialExecution complexityRegulatory exposureCapital intensityTiming advantageFounder fitFailure reversibility

How to read a score

A high score means the idea currently shows strong evidence across the selected dimensions. It is not a guarantee that the company will work.

A low score is useful when it identifies a fixable constraint: weak distribution, unclear buyer urgency, shallow differentiation, or a business model that needs a different wedge.

The best reports are challenged. IdeaForge exposes the reasoning layer so founders can spot missing context, add evidence, and run a sharper second pass.

Run an Analysis View Trust Ledger