Propensity to Buy model

What is propensity to buy?

The Propensity to Buy score (0.0 to 1.0) estimates the probability that a customer will make a purchase within the next 30 days. A score of 0.72 means the model predicts a 72% chance of a purchase in the next month.

How it's calculated

The model combines recency signals (how recently they browsed or bought), frequency patterns, and product affinity with external factors like email engagement, seasonal patterns from similar customers, and promotional response history.

Category-level propensity

In addition to an overall propensity score, LiftSignals calculates per-category affinity scores (e.g., Footwear: 0.41, Accessories: 0.12). This helps you personalise the content of outreach, not just the timing.

Optimal send time

The model also predicts each customer's optimal send time — the day of week and time of day when they are most likely to engage with marketing messages. This is visible on the Customer Profile page.