Snowplow to ClickHouse Integration
AI-first ETL from Snowplow into ClickHouse. Governed entities, incremental sync, typed landing tables.
How Datrise loads Snowplow into ClickHouse
Datrise syncs Snowplow's events, users, sessions, contexts, page views, and event schemas into ClickHouse as a MergeTree table per source entity. Flexible or custom fields land in JSON or Map columns, and timestamps such as created, updated, and status changes are typed as DateTime64.
Sync is incremental: Datrise uses inserts into a ReplacingMergeTree keyed on stable id, so the latest version wins on merge, so re-runs update only what changed. Partition by month and order by (entity id, updated-at) for fast range scans. ClickHouse deduplicates asynchronously on merge, so Datrise uses ReplacingMergeTree and FINAL-safe queries rather than assuming immediate upserts.
Ideal for high-volume event analytics that need sub-second aggregation.
Endpoints
Snowplow: Behavioral event pipeline with a strict schema.
ClickHouse: Columnar OLAP engine for fast aggregations.
How Snowplow entities map to ClickHouse
| Snowplow entity | ClickHouse object | Notes |
|---|---|---|
| events | snowplow_events | DateTime64 events |
| users | snowplow_users | id PK · linked to snowplow_events |
| sessions | snowplow_sessions | id PK · linked to snowplow_events |
| contexts | snowplow_contexts | id PK · linked to snowplow_events |
FAQ
How does Datrise handle Snowplow's custom fields in ClickHouse?
Flexible values are stored as JSON or Map columns, so new fields don't require a migration; strongly-typed fields — dates, numbers, and references — are promoted to native ClickHouse types.
How does the Snowplow to ClickHouse sync stay up to date?
It runs incrementally — Datrise uses inserts into a ReplacingMergeTree keyed on stable id, so the latest version wins on merge.
Related pipelines
More destinations for Snowplow
Early access
Connect Snowplow to ClickHouse the easy way
Skip brittle scripts and manual exports. Join the waitlist to get a guided setup, AI-assisted mapping, and reliable incremental sync for this integration.