How Recommendations Studio works

Recommendations Studio lets you create personalized products, content, and offer recommendations that adapt to each customer based on their activity, purchase history, affinities, and your catalog data. Recommendations work across email, push, SMS, in-app, websites, and other channels — for both known and anonymous customers.

Blueshift provides a library of ready-to-use recommendation recipes organized by industry and lifecycle stage. Select a recipe for your use case and use it as-is or customize it.

You can compare the performance of different recommendation schemes by running A/B tests, reviewing email click heatmaps, and measuring engagement with the Insights-Recommended items report.

Prerequisites

Before you can use recommendations, you need:

• An item catalog uploaded to Blueshift.
• Customer interactions with catalog items sent via an event feed or API.

The recommendation engine uses this catalog metadata, customer data, and behavioral signals to dynamically generate relevant recommendations.

External recommendations

If you have your own recommendation models, you can upload them as recommendation feeds and combine them with Blueshift's AI-powered recommendations.

How it works

You configure a recommendation scheme in Recommendations Studio and associate it with a messaging template. When a campaign runs, the scheme applies its logic to each user in the segment, delivering a personalized message with items tailored to that user.