Predictive scoring is a model's estimate of the likelihood of an outcome such as churn, conversion, purchase, or engagement.
Category
AI
Also known as
predictive score
It is a classic machine-learning use of CDP data, turning historical behavior into a forward-looking number a campaign can act on. For example, a high churn-risk score can trigger a retention journey, while a high purchase-propensity score can prioritize a product recommendation.
In practice
A score is only actionable if it reaches an activation tool. A churn score that never triggers a retention journey is a dashboard number, not a decision.
Discuss this term
A question, a bug, or outdated info? Tell us — it goes straight to the team, and we keep the glossary current.