Segmentation

How we estimate audience sizes across millions of users in seconds using sampling and Bayesian math.

Inside the static analyzer that powers type inference, contradiction detection, and reactive evaluation in CQL.

Implementing personalization in a headless architecture is significantly harder than simple content delivery. But it shouldn't be.

CQL just got a memory. With event-based expressions, you can now segment users by past behavior and create audiences that react to real journeys.

New system event fired whenever a user matches an audience, enabling deeper analysis, cleaner segmentation, and upcoming analytics upgrades.

New automatically processed e-commerce events give you deeper insights into customer behavior for smarter segmentation and real-time personalization.

We've just rolled out a major upgrade to our location data, making it more structured, detailed, and versatile for location-based segmentation and personalization.

Feature flags (or feature toggles) allow product teams to control the rollout of new features without redeploying code. They are widely used to enable AB testing, perform gradual rollouts, and reduce risk when releasing new product features.

A customer data platform, or CDP, helps companies create a single, unified profile for each customer to understand their needs and provide personalized experiences that feel relevant and timely.

If you ask 10 marketers, 8 will say third-party data is crucial for website personalization. However, first-party data can provide a far more effective and sustainable solution.