ICP scoring on raw account data is guesswork. The same account can score Tier A in one framework and Tier C in another depending on which signals were even checked. Three patterns matter for any team running a real scoring pass.
The framework is transparent and tunable. Every tier assignment shows the reasoning per dimension — size fit, industry, geography, active signal, stage. When the AE or marketer disagrees with a tier, they adjust the weight and re-run. The framework is not a black box. The verified data underneath it is the fixed input; the scoring weights are the variable.
Active signal separates Tier A from Tier B. Two accounts with identical firmographics can sit in very different buying windows. A company that closed a Series E last quarter has budget, mandate, and stack-rebuild appetite. A company that has been quiet on funding for three years has none of those things. The prompt surfaces the last round date and IPO history so the signal layer is visible, not assumed.
Stage indicator changes the sales motion. A public company runs a replacement sale — they have a stack, they are evaluating vendors against incumbents, the conversation is about TCO and switching cost. A growth-stage company runs a rebuild sale — they are scaling beyond their original tooling, the conversation is about future-fit and integration depth. The same product gets pitched differently. The prompt flags stage so the rep picks the right script.
Data drawn from 300M+ verified contacts and millions of company records under GDPR, CCPA, SOC 2, ISO 27701, ISO 31700, and TRUSTe.
