Skip to main content

For AI agents: a Markdown version of this page is at https://docs.snowplow.io/docs/cdi.md. The index of all documentation pages is at https://docs.snowplow.io/llms.txt, and the index for this section is at https://docs.snowplow.io/docs/cdi/llms.txt.

Snowplow CDI overview

Snowplow Customer Data Infrastructure (CDI) is a pipeline for behavioral data. You define the events and entities you want, trackers in your apps send them to a Collector, and the pipeline checks each event against its schema, enriches it, and loads it into your warehouse, lake, or stream. Events that don't match their schema become failed events, so they aren't loaded and aren't lost.

Use CDI when you need event-level data you control: for product and marketing analytics, for a composable CDP, or as the source for Signals.

Start here​

You can start on a hosted environment that Snowplow provisions for you, with no infrastructure to deploy. The free trial is 14 days of Snowplow CDI Cloud, and you usually receive access instructions within moments of verifying your email address. Once the environment exists, either hand the setup to your coding agent with the quickstart, or follow the steps yourself to track your first events and query them. When you're ready for production, see Choose a Snowplow deployment.

Quickstart30 min

Quickstart with your coding agent

Get a hosted environment and send validated events from your app.

Step by step

Get access to Console, track your first events, and query them.

Already have data in your warehouse?

Query the atomic events table, then build sessions and users tables with Unified Digital.

Core concepts

Events, entities, schemas, and how the pipeline processes them.

Coming from Segment, Google Analytics, Adobe Analytics, or Mixpanel? See Migrate from another tool.

Explore by stage​

The CDI sidebar follows the path an event takes, from design to data model. Each card opens the first page of a sidebar group.

Design tracking

Plan events and entities in Event Studio, and generate tracking code with Snowtype.

Collect events

Trackers for web, mobile, server, and games, and webhooks for third-party tools.

Test your tracking

Check the events your tracking sends before they reach production.

Enrich events

Add data to events in the pipeline, such as location, parsed user agents, and campaign details, or mask personal data.

Identities

Stitch a user's identifiers into one Snowplow ID in real time, across devices and logins.

Load and forward

Load into warehouses and lakes, forward events to other tools, or stream them to your own targets.

Query and model data

Query the atomic events table, and build sessions, users, and attribution tables with dbt packages.

Monitor data quality

Find and fix failed events, and set up data quality alerts.

Reference​

Configuration options, API references, and component versions are in the Reference section, shared with Signals.

Was this page helpful?