// the find
Multiwoven/multiwoven
🔥🔥🔥 Open source Reverse ETL - alternative to hightouch and census.
Multiwoven is an open-source Reverse ETL platform — you define SQL models against your data warehouse (Redshift, BigQuery, Snowflake, Databricks, Postgres) and it syncs the results to business tools like Salesforce, HubSpot, or Klaviyo on a schedule. It's a self-hosted alternative to HighTouch/Census for teams that can't or won't pay SaaS pricing for data activation. Target audience is data engineers at mid-sized companies running their own cloud infrastructure.
The connector framework is well-structured: each integration is an isolated Ruby gem with its own spec.json, catalog.json, and client.rb, so adding a new destination doesn't require touching core sync logic. CI is split per concern (server-ci, integrations-ci, ui-ci), which means a broken connector doesn't block backend deploys. Docker Compose local setup is genuinely one command. Helm chart support is there if you want to run it on Kubernetes, not bolted on as an afterthought.
1,657 stars for a project that's been around this long is a weak signal — HighTouch and Census are both well-funded, and Multiwoven hasn't demonstrated meaningful community traction compared to peers like Airbyte or Meltano in adjacent spaces. The connector list has significant gaps: Google Ads, Amplitude, Mixpanel, Braze, Intercom are all 'coming soon' — exactly the tools data teams most commonly need for activation. AGPL-3.0 is a business risk if you want to embed this in a product rather than run it internally. The Ruby on Rails backend is fine but means your team needs Ruby expertise alongside whatever stack runs your warehouse queries, which is an operational cost.