[30-September-2025]
Snowplow Signals is now generally available. Signals is a real-time customer intelligence system that computes and serves user context. With Signals, you can power in-session personalization, recommendations engines, adaptive UIs, and agentic experiences like AI copilots and chatbots — without spending months or years building and maintaining the underlying infrastructure yourself.
The new release includes:
Multi-cloud and multi-warehouse support
- Cloud platforms: AWS and GCP deployments with identical developer experience
- Data warehouses: Native integration with Snowflake and BigQuery for batch processing (Databricks support coming soon)
No-code UI in Snowplow Console
Product managers and engineers can manage Signals directly in the Snowplow UI without writing code. This includes:
- Self-service deployment flows
- User attribute management interface
- Configuration and monitoring tools
The Python SDK is available for teams that prefer code-based workflows.
Production-grade performance
The Profiles API delivers 45ms p50 response times, enabling in-session personalization use cases like:
- Dynamic pricing and offers
- Contextual product recommendations
- AI agent responses with real-time user context
Live debugging with Snowplow Inspector
Debug and validate your personalization logic in real time using the Snowplow Inspector integration.
Components
Signals includes several core components:
- Streaming engine: Calculates real-time user attributes from live event streams (for example, current page views or cart status)
- Batch engine: Computes historical user attributes from warehouse data using ML models (for example, lifecycle stage or propensity scores)
- Profiles store: Stores and serves user profiles at low latency to your applications
- Interventions: Triggers real-time actions in your application based on user behavior
Getting started
If you'd like to trial Signals, contact your Customer Success Manager or reach out to support@snowplow.io.
For detailed implementation guidance, see the Signals documentation.