Customer Health Score Designer

A quantitative architect who transforms scattered customer signals into a single, defensible health number that predicts churn and expansion before the account team feels it—building models that CS teams actually trust enough to act on.

- The VITALS methodology — 5-pillar framework from outcome definition to continuous model stewardship

- Signal audit template mapping every data source with reliability, latency, and predictive contribution ratings

- Feature engineering playbook converting raw events into adoption depth, engagement velocity, and relationship strength

- Composite score architecture with transparent sub-dimensions for usage, relationship, outcome, and commercial health

- Threshold definition using ROC analysis and outcome distribution — explicit green/yellow/red band rationale

- Score decomposition design showing sub-score contributions and top signal movers for every account

- New-account scoring rules that handle sparse behavioral data and progressive confidence mechanics

- Quarterly model retrospective framework comparing predicted bands against actual renewal and expansion outcomes

- CSM feedback loop design — disagree buttons, override logs, qualitative context fields feeding model improvement

Drop into Claude, ChatGPT, Cursor, or any AI tool. Bring your real customer success problem — accounts churn without warning, you can't identify expansion opportunities early enough, or your current health score is a black box nobody trusts. It thinks like someone who's rebuilt health scoring systems through multiple product maturity stages.

Bundles or prompts related to customer success operations and predictive analytics.