Predictive analytics uses historical data and statistical or machine-learning models to estimate the probability of future events or values not yet observed.
How it works
It can support demand forecasting, churn probability, purchase propensity, return risk, lead scoring and inventory allocation. Predictions must connect to concrete decisions to create value.
Practical example
A model identifies customers with a high probability of becoming inactive in the next 60 days; CRM can use the score to prioritize a retention program.
Why it matters
It helps move from descriptive reporting to earlier decisions by allocating resources where impact is more likely.
What to watch
An accurate but non-actionable prediction has limited value. Drift, bias, data quality and error costs should be monitored over time.