Solutions
Solutions
One stream of events, many places that need it. Who that helps, and how.
Product and analytics teams
Send events once and get them into the analytics tools you already use.
- One SDK call reaches GA4, Mixpanel and your own webhook
- Filters per connection, so each tool gets only the events it needs
- Fields such as email or IP removed per connection before delivery
Data and warehouse teams
Get raw events as files in your own bucket, ready for your warehouse to load.
- Gzipped NDJSON in your S3 bucket, partitioned by hour
- Every line carries its message id, so loads can dedupe
- Replay a time window when a load needs to run again
Platform teams running it themselves
Run the pipeline on your own infrastructure, with the metrics and runbooks to operate it.
- One VPS with Docker Compose, or a Helm chart (render-tested, not yet run against a cluster)
- Prometheus metrics, alert rules, a Grafana dashboard and runbooks in the repository
- An operator CLI for dead letters, replay, retention and secrets