Predictive analytics uses historical and current data with statistical or machine-learning techniques to estimate future outcomes or probabilities. Within business intelligence, it extends analysis beyond describing what happened toward anticipating what may happen next.
Examples include forecasting demand, identifying customers at risk of churn, estimating sales probability, predicting inventory needs, detecting unusual activity, or identifying which prospects are most likely to convert.
Predictions are not certainties. Their usefulness depends on data quality, model design, changing market conditions, and whether the organization can take an appropriate action based on the result. Models should be monitored because performance can deteriorate over time.
Predictive analytics can strengthen data-driven decision-making when predictions are combined with business judgment, transparent assumptions, and measurable actions.
