Glossary

Signals collected from a user's actions — clicks, views, purchases, searches — that Raptor uses to build individual preference profiles and generate recommendations.

A recommendation technique that identifies items a user might like based on the behaviour of users with similar preferences — "customers like you also bought…".

Statistical algorithms that learn patterns from historical data without explicit programming. Raptor uses ML models to continually improve recommendation accuracy.

Tailoring the content, products, and messaging shown to each individual visitor based on their unique profile and real-time behaviour.

A system that predicts and surfaces items most likely to interest a specific user. Raptor's engine combines collaborative filtering, content-based signals, and popularity data.

The measurable increase in a key metric (revenue, CTR, conversion) attributable to personalised recommendations compared to a non-personalised baseline.