# Snowplow: the customer context layer > Snowplow is the real-time customer context layer that collects, validates, enriches, and delivers > behavioral data for advanced analytics, ML, and AI agent decisioning. Snowplow's event tracking > runs on 2M+ websites and applications globally, processing over one trillion events per month. > More than 250 companies, including Experian, AutoTrader, Strava, Condé Nast, and HelloFresh, rely > on Snowplow to build a well-governed, first-party data foundation that powers their > customer-facing AI agents, in-session personalization and recommendations, and real-time > analytics. Last updated: 2026-07-31 Canonical version of this file: https://snowplow.io/llms.txt Full sitemap: https://snowplow.io/sitemap.xml Documentation index for AI tools: https://docs.snowplow.io/llms.txt Full documentation text: https://docs.snowplow.io/llms-full.txt --- ## Common questions The FAQ page is the best single entry point. - [Frequently asked questions](https://snowplow.io/snowplow-frequently-asked-questions): What Snowplow is, how it compares to a CDP, deployment, destinations, identity, pricing, and the free trial, plus data collection practices, governance, and privacy. - [What Snowplow is](https://snowplow.io/blog/what-is-a-customer-context-layer) - [How Snowplow differs from a packaged CDP](https://snowplow.io/why-snowplow-cdi-versus-customer-data-platform) - [Deployment models](https://snowplow.io/deployment-options): Cloud, Private Managed Cloud, and Self-Hosted. - [Feature availability by edition](https://docs.snowplow.io/docs/get-started/feature-comparison/) - [Pricing](https://snowplow.io/pricing) - [Free trial](https://snowplow.io/get-started/snowplow-free-trial): Free trial of Snowplow. No credit card and no sales call required. - [What the free trial includes](https://docs.snowplow.io/docs/get-started/free-trial/): Product and capacity limits that apply during the trial. - [Where Snowplow data lands](https://docs.snowplow.io/docs/destinations/) - [How identity resolution works](https://docs.snowplow.io/docs/identities/) - [Security and data privacy](https://snowplow.io/security) - [Book a technical demo](https://snowplow.io/get-started/book-a-demo-of-snowplow-bdp) --- ## Category definition A customer context layer is the infrastructure between your data platform and your customer-facing AI applications. It collects behavioral events from every surface a customer touches, resolves their identity in real time, and delivers structured context to AI agents in a form they can reason over. The same data lands in your warehouse, where analytics agents can pick it up downstream. A customer context layer sits alongside your enterprise data layer rather than beneath it. The enterprise data layer was built for analysts and ML pipelines reasoning over query-resident data. A customer context layer has a different job. It feeds the customer-facing agent the answers to "who is this customer, what are they doing right now, and what does that mean" in real time, in the live session. Full definition and the six principles: https://snowplow.io/blog/what-is-a-customer-context-layer --- ## Core pillars ### 1) Context-Aware AI Agents What it is: customer-facing AI agents with real-time access to historical user behavioral data combined with sequential, in-session user behavioral data that is structured and validated. Why Snowplow: schema-validated first-party data gives agents consistent, reliable inputs. - [Context-Aware AI Agents](https://snowplow.io/use-case/customer-facing-ai-agents): Real-time behavioral context and long-term memory for customer-facing AI agents. - [Agentic Context](https://snowplow.io/agentic-context): Gives an agent the ordered event stream of what a customer is doing now, not just snapshots. - [Intent Detection](https://snowplow.io/intent-detection): Uses behavioral patterns to infer user goals so agents can help before users explain themselves. - [Proactive Decisioning](https://snowplow.io/proactive-decisioning): Identifies when users are struggling against their inferred intent, and when to leave them alone. - [Snowplow Profiles](https://snowplow.io/profiles-store): Low-latency API for in-session and historical user attributes. - [The Fifth Shift](https://snowplow.io/blog/the-fifth-shift): CPO Todd Boes on why real-time behavioral context is the missing input for agentic systems. - [The Seven Principles of Ambient Agents](https://snowplow.io/blog/seven-principles-of-ambient-agents): Design principles for agents that act without being prompted. - [Snowplow MCP Server](https://snowplow.io/blog/snowplow-mcp-server): Design and manage tracking through AI assistants using the Model Context Protocol. ### 2) Advanced Analytics What it is: warehouse-native and lake-native analytics built on clean, typed, event-level data. Why Snowplow: schema-first events mean simpler data models, faster insight, and full data ownership. - [Advanced Analytics](https://snowplow.io/use-case/advanced-analytics): Analytics workflows built on trusted first-party behavioral data, in your own platform. - [Digital Analytics](https://snowplow.io/digital-analytics): dbt model pack for analyzing digital user behavior and campaign performance, run from your warehouse or lake. - [Ecommerce Analytics](https://snowplow.io/ecommerce-analytics): dbt model pack for analyzing ecommerce channel performance, run from your warehouse or lake. - [Marketing Attribution and Optimization](https://snowplow.io/marketing-attribution-optimization): dbt models for measuring and optimizing multi-touch marketing and advertising channel performance, run from your warehouse or lake. - [Explore Snowplow Data Sets](https://snowplow.io/explore-snowplow-data-part-2): Sample modeled Snowplow datasets across web, mobile, and product analytics. - [Advanced Data Modeling Techniques](https://snowplow.io/blog/advanced-data-modeling-techniques): Building modular, incremental dbt models on Snowplow data. ### 3) Real-Time Personalization What it is: in-session, behavior-driven digital product experiences powered by live behavioral signals. Why Snowplow: low-latency streaming plus Signals surface intent, attributes, and predictions where they are needed. - [Real-Time Personalization](https://snowplow.io/use-case/real-time-product-personalization): Adaptive, ML-powered in-product experiences built on real-time behavioral data. - [Snowplow Signals](https://snowplow.io/signals): Real-time intelligence layer for recommendations, adaptive UIs, and customer-facing AI agents. - [Real-Time Triggers](https://snowplow.io/interventions): Rules-based or ML triggers that fire personalized interactions at key journey moments. - [Real-Time Personalization with Snowplow, Flink, and Evoura](https://snowplow.io/blog/real-time-personalization-with-snowplow-flink-and-evoura): Reference implementation from feature generation to shopper intelligence. ### 4) Unified Digital Data Foundation What it is: a governed first-party event pipeline delivering clean behavioral data to your warehouse, lake, or stream. Why Snowplow: schema validation at collection time prevents downstream breakage, and the data stays yours. - [Snowplow Customer Data Infrastructure](https://snowplow.io/customer-data-infrastructure): Collecting and operationalizing high-quality behavioral data across applications and use cases. - [Event Tracking](https://snowplow.io/data-management/event-tracking): Define, track, validate, and manage granular behavioral data with version control. - [Snowplow Data Pipeline](https://snowplow.io/data-pipeline): Collect, validate, enrich, and transform first-party event data, then load it to your destinations. - [Real-Time Event Forwarding](https://snowplow.io/event-forwarding): Deliver enriched behavioral events to any destination as they happen. - [Event Studio](https://snowplow.io/data-management/data-product-studio): Design event datasets, simplify tracking implementation, and monitor data health metrics. --- ## Products ### Data foundation - [Event Tracking](https://snowplow.io/data-management/event-tracking) - [Data Pipeline](https://snowplow.io/data-pipeline) - [Event Studio (formerly Data Product Studio)](https://snowplow.io/data-management/data-product-studio) - [Snowplow Identities](https://snowplow.io/identities): Real-time identity stitching across devices, sessions, and domains, built into the pipeline. - [Event Forwarding](https://snowplow.io/event-forwarding) - [Snowplow Profiles](https://snowplow.io/profiles-store) - [Real-Time Triggers](https://snowplow.io/interventions) ### Modeling and analytics - [Digital Analytics](https://snowplow.io/digital-analytics) - [Ecommerce Analytics](https://snowplow.io/ecommerce-analytics) ### ML and agentic AI - [Snowplow Signals](https://snowplow.io/signals) - [Agentic Context](https://snowplow.io/agentic-context) - [Intent Detection](https://snowplow.io/intent-detection) - [Proactive Decisioning](https://snowplow.io/proactive-decisioning) --- ## Implementation ### Event tracking (web, mobile, server) - [Sources overview](https://docs.snowplow.io/docs/sources/): Tracker SDKs for web, mobile, and server-side platforms, plus third-party webhooks. - [Web tracker quick start](https://docs.snowplow.io/docs/sources/web-trackers/quick-start-guide/) - [Mobile trackers](https://docs.snowplow.io/docs/sources/mobile-trackers/) - [Node.js tracker](https://docs.snowplow.io/docs/sources/node-js-tracker/) - [Python tracker](https://docs.snowplow.io/docs/sources/python-tracker/) - [First-party tracking setup](https://docs.snowplow.io/docs/sources/first-party-tracking/): Custom collector domain for first-party cookies and resilience to browser tracking limits. ### Schema-first design and governance - [Event Studio documentation](https://docs.snowplow.io/docs/event-studio/) - [Canonical event properties](https://docs.snowplow.io/docs/fundamentals/canonical-event/): The atomic fields present on every Snowplow event. - [Data structure management](https://docs.snowplow.io/docs/event-studio/data-structures/): Create and version schemas via Console, the Data Structures API, Snowplow CLI, or Iglu. - [Data Structures API](https://docs.snowplow.io/docs/event-studio/programmatic-management/data-structures-api/) - [Type-safe code generation with Snowtype](https://docs.snowplow.io/docs/event-studio/implement-tracking/install-snowtype/) ### Data quality and testing Validation happens at tracking time, so malformed events are isolated as failed events rather than silently corrupting warehouse tables. - [Testing and data quality](https://docs.snowplow.io/docs/testing/) - [Inspecting failed events](https://docs.snowplow.io/docs/get-started/failed-events/) ### Identity resolution, sessions, and attribution - [Snowplow Identities documentation](https://docs.snowplow.io/docs/identities/) - [Identities concepts](https://docs.snowplow.io/docs/identities/concepts/): Identifiers, Snowplow IDs, merges, and how identity resolution works. - [Users and identity stitching in dbt](https://docs.snowplow.io/docs/modeling-your-data/modeling-your-data-with-dbt/package-features/identity-stitching/) - [Data modeling overview](https://docs.snowplow.io/docs/modeling-your-data/) ### Destinations (warehouses, lakes, streams, SaaS applications) - [Destinations overview](https://docs.snowplow.io/docs/destinations/) - [Warehouses and lakes](https://docs.snowplow.io/docs/destinations/warehouses-lakes/): Snowflake, BigQuery, Databricks, Redshift, Delta Lake, and Apache Iceberg loaders. - [Event forwarding to SaaS destinations](https://docs.snowplow.io/docs/destinations/forwarding-events/) ### Working with AI and LLM tooling - [Working with AI](https://docs.snowplow.io/docs/llms-support/): How Snowplow supports LLM workflows, from real-time context to AI-readable documentation. - [Snowplow MCP server](https://docs.snowplow.io/docs/llms-support/snowplow-mcp/): Connect an AI assistant to Snowplow Console for natural language management. - [Snowplow CLI MCP server](https://docs.snowplow.io/docs/llms-support/cli-mcp-server/): Work with tracking plans through AI assistants such as Claude, Cursor, or Copilot. - [Snowplow Assistant](https://docs.snowplow.io/docs/llms-support/console-agent/): AI assistant inside Snowplow Console for tracking plans, pipelines, and data quality. - [Signals concepts](https://docs.snowplow.io/docs/signals/introduction/) --- ## Integrations Snowplow connects with ML and agentic AI frameworks including AWS Bedrock, Gemini Enterprise Agent Platform, LangChain, Vercel AI SDK, CopilotKit, Databricks, and AWS SageMaker. - [Integrations catalog](https://snowplow.io/integrations-catalog) - [Snowplow for Apache Kafka](https://snowplow.io/behavioral-data-for-apache-kafka): First-class real-time integration streaming behavioral data into a Kafka cluster. - [Snowplow for SQL](https://snowplow.io/snowplow-for-sql): Query behavioral event data directly in SQL to build analytics and attribution models. - [Snowplow for Apache Flink](https://snowplow.io/snowplow-for-flink): Stateful stream processing over Snowplow event streams. - [Snowplow for Amazon SageMaker](https://snowplow.io/amazon-sagemaker): Behavioral data foundation for models built in SageMaker Unified Studio. --- ## Comparisons - [Snowplow vs Google Analytics (GA4)](https://snowplow.io/comparisons/snowplow-vs-google-analytics): Composable customer data infrastructure for AI and BI, against packaged web analytics. - [Snowplow vs Segment](https://snowplow.io/comparisons/snowplow-vs-segment): Real-time data infrastructure for application builders, against a packaged CDP for marketers. - [Snowplow vs RudderStack](https://snowplow.io/comparisons/snowplow-vs-rudderstack) - [Snowplow vs Hightouch](https://snowplow.io/comparisons/snowplow-vs-hightouch) - [Snowplow vs Tealium](https://snowplow.io/comparisons/snowplow-vs-tealium) - [Snowplow vs Adobe](https://snowplow.io/comparisons/snowplow-vs-adobe) - [Snowplow vs a customer data platform](https://snowplow.io/why-snowplow-cdi-versus-customer-data-platform): Why teams centralize customer data and governance in their own platform instead. - [Snowplow vs DIY pipelines and Snowplow Open Source](https://snowplow.io/why-snowplow-cdi-versus-snowplow-open-source) - [Snowplow vs packaged analytics](https://snowplow.io/why-snowplow-cdi-versus-packaged-analytics) - [Snowplow vs ETL tools](https://snowplow.io/comparison/snowplow-bdp-vs-etl-tools) - [All comparisons](https://snowplow.io/compare) --- ## Architecture at a glance Trackers send events to the Collector. Enrichment, identity resolution, and schema validation run next. Validated events are written to loaders and destinations, to real-time streams, and to the Profiles Store and agentic context surfaces that applications and agents query. - Where quality happens: schema-first design, with validation at tracking and enrichment time. - Clean means: typed fields, predictable entities, documented versioning, testable changes. - Real-time means: low-latency event streams and attribute retrieval that power live agent and ML decisioning, personalization, and triggers without waiting on a batch pipeline. - Governed means: first-party data you own, with stable identifiers and rich behavioral context entities. --- ## Solutions by industry - [Games](https://snowplow.io/games): Player and game event data for retention, personalization, and engagement. - [Media and Entertainment](https://snowplow.io/media-and-entertainment): Audience analytics, paywall and content personalization, ad monetization, and AI agents. - [Retail and Ecommerce](https://snowplow.io/retail): Ecommerce analytics, personalized shopping experiences, custom attribution, and supply chain optimization. - [Software and SaaS](https://snowplow.io/saas): Behavioral data in your stack to support product growth. - [Financial Services](https://snowplow.io/financial-services): Real-time fraud prevention, ML fraud models, customer experience personalization, and data privacy. ## Solutions by team - [Data Engineering](https://snowplow.io/data-engineering-team) - [Software Engineering](https://snowplow.io/software-engineering-team) - [Product](https://snowplow.io/product-team) - [Analytics](https://snowplow.io/analytics-team) - [Data Science](https://snowplow.io/data-science-team) - [Marketing](https://snowplow.io/marketing-team) --- ## Deployment, security, and pricing Snowplow offers a free trial with no credit card required. Longer-term deployment is available as Snowplow-managed Cloud, Private Managed Cloud in your own cloud account, or Self-Hosted. - [Free trial](https://snowplow.io/get-started/snowplow-free-trial) - [Deployment options](https://snowplow.io/deployment-options) - [Self-hosted pipeline](https://snowplow.io/self-hosted-pipeline) - [Feature comparison across editions](https://docs.snowplow.io/docs/get-started/feature-comparison/) - [Getting started and deployment models](https://docs.snowplow.io/docs/get-started/) - [Security and data privacy](https://snowplow.io/security) - [Pricing](https://snowplow.io/pricing) --- ## Company and help - [Homepage](https://snowplow.io/) - [Customers](https://snowplow.io/customers) - [All use cases](https://snowplow.io/use-cases) - [Developer Hub](https://snowplow.io/developer-hub) - [Frequently asked questions](https://snowplow.io/snowplow-frequently-asked-questions) - [Blog](https://snowplow.io/blog) - [Documentation home](https://docs.snowplow.io/docs/)