Ten-dimension scoring: how the score is computed

A deterministic rule engine scores ten dimensions — same input, same score, replayable on the public verify page. This page covers what each dimension reads and how the total maps to verdicts.

The ten dimensions

• User specificity (out of 10) — is your "who" a findable group of people? Narrower audiences score higher; danger words zero it out.
• Scene clarity (out of 10) — does the scene have timing and triggers (every time / when / before deadline)?
• Pain & payment (out of 15) — is the problem real + how do they pay today; concrete numbers beat vague claims.
• Evidence strength (out of 20, heaviest) — strongest single item + independent-source cross-validation + traceable links.
• Supply gap (out of 15) — why current solutions fall short, and which link you cut into (supply questions 3/4).
• Repeatability (out of 10) — recurring demand, or one-off that dies after the hype.
• 48-hour feasibility (out of 10) — is step one light enough: landing page / form / manual delivery yes; training models / full pipeline no.
• Traffic path (out of 5) — where the first users come from; a named channel counts, a shrug does not.
• Product-shape fit (out of 5, bonus) — is the recommended shape "closest to the user's entrance": plugin inside the host app, single page catching search terms, form + manual delivery, etc.; an unnarrowed idea caps low.
• Compliance risk (deduction, up to −30) — trademark / copyright / black-hat / privacy / regulated / grey-hat / impersonation wordlists; grey-hat and impersonation also carry a hard-rule stop.

Supply-demand imbalance also produces a parallel five-level conclusion (none / weak / medium / strong / alpha) from the supply-gap questions.

Five verdicts

• Stop — score or compliance risk hits the floor; do not invest.
• Rescope — the direction may hold but the claim/audience needs rewriting; back to the inbox.
• Get more evidence — there is signal but not enough proof; bring originals and payment records, then re-screen.
• Validate MVP — enter the 48-hour real-world validation.
• Alpha sprint — demand surging + supply short; ship first, charge first, conclude the same day.

Which band your score lands in is decided by the engine's current rule configuration; input and configuration are archived with every verdict snapshot and replayable byte-for-byte on the public verify page (match: true).

Above the score: hard rules

No matter how high the score, a hit on the seven hard rules caps the verdict: danger words stop outright, zero evidence never turns green, unverified willingness to pay blocks the green light, and more.