At a glance
| Source connectors | 150+ listed: database CDC, SaaS apps and files |
| Destination connectors | 60+ listed: Snowflake, BigQuery, Databricks, Iceberg, PostgreSQL, Elasticsearch, Kafka-compatible APIs and more |
| Batch or streaming | Both: real-time streaming and scheduled batch loads |
| Change data capture | Yes, log-based for PostgreSQL, MySQL, SQL Server and Oracle |
| Transformations | Derivations in SQL (SQLite), TypeScript, or Python on private and BYOC data planes |
| Deployment | Public SaaS; private or BYOC data plane on Enterprise |
| Licensing | Managed service per GB plus per connector instance; core source-available under BSL |
Estuary suits teams that need low-latency change data capture from operational databases into warehouses, lakes or streaming systems, with batch SaaS loads on the same platform. Its published per-GB pricing is easy to model, but the monthly fee per connector instance adds up when you have many small sources.
Overview
Estuary moves data between systems continuously. A capture ingests data from a source into collections, which are datasets stored as JSON documents in a cloud storage bucket and validated against a JSON schema on every read and write. A materialization pushes collections to a destination and keeps it updated as new documents arrive. Databases are captured with log-based CDC, and SaaS sources and files with scheduled batch loads.
The home page lists 150+ source and 60+ destination connectors (elsewhere it says 200+ in total), including CDC for PostgreSQL, MySQL, SQL Server and Oracle, MongoDB, and SaaS sources such as Salesforce and HubSpot. Destinations include Snowflake, BigQuery, Databricks, Apache Iceberg, PostgreSQL, Elasticsearch and Kafka-compatible APIs. Transformations, called derivations, can be written in SQL (SQLite), TypeScript or, on private and BYOC data planes, Python.
Key features
- Log-based change data capture from PostgreSQL, MySQL, SQL Server and Oracle
- Streaming and batch pipelines on one platform, with collections stored in your cloud storage bucket
- Derivations for continuous transformations in SQL (SQLite) or TypeScript, and Python on private and BYOC data planes
- Schema validation of every document against a JSON schema
- flowctl command-line tool for inspecting, testing and publishing pipelines as code
- Kafka-compatible API and destinations such as Snowflake, BigQuery, Databricks and Iceberg
- Private and bring-your-own-cloud deployment on the Enterprise plan
Pros and cons
Pros
- Real-time CDC and batch loads on the same platform
- Published pay-as-you-go prices and a free Developer plan without a time limit
- Pipelines can be managed as code with flowctl
- BYOC option keeps the data plane in your own cloud account
Cons
- Per-connector-instance fees become significant with many small sources
- Private and BYOC deployment are Enterprise-only and custom-priced
- The core is under the Business Source License, not an OSI open source licence
- Estuary's own pricing example and calculator gave different figures when checked
Pricing
The free Developer plan covers up to 10 GB a month and 2 connector instances, with no credit card. The pay-as-you-go Cloud plan is USD 0.50 per GB plus USD 100 a month for each connector instance, falling to USD 50 a month per instance after the sixth. Estuary's FAQ example counts data at both source and destination (100 GB moved from Postgres to Snowflake counts as 200 GB, or USD 300 a month), while the page calculator showed a different figure for the same case, so check the calculator with your own volumes. Exceeding the Developer limits starts a 30-day Cloud trial. Enterprise, with private and BYOC deployment, SSO, a SOC 2 report and a HIPAA BAA, is custom-priced on annual contracts.
Pricing checked on the vendor's official page on 7 November 2026. Prices change; confirm before buying.
Who it is best for
Teams that need real-time CDC from operational databases into a warehouse, lake or streaming system, and want published per-GB pricing.
Who should look elsewhere
- Teams with many low-volume SaaS sources, where per-connector fees dominate: see Fivetran alternatives.
- Teams that require a fully self-hosted, OSI open source tool: see Airbyte vs Meltano.
Verdict
Estuary is a strong option when latency matters and most of your data comes from databases that support log-based CDC. Its pricing is transparent but counts connector instances as well as data, so it is less economical for long lists of small SaaS sources, and private deployment requires an Enterprise contract.
Frequently asked questions
Is Estuary Flow the same as Estuary?
Yes. Estuary's documentation says "Estuary was formerly Estuary Flow". The Flow name remains in the GitHub repository and the flowctl CLI.
How much does Estuary cost?
The Developer plan is free up to 10 GB a month and 2 connector instances. The Cloud plan is USD 0.50 per GB plus USD 100 a month per connector instance (USD 50 after the sixth). Enterprise is custom-priced.
Is Estuary open source?
Partly. The main repository is published with a Business Source License and an Apache 2.0 licence file, so it is source-available rather than fully open source. The managed service is commercial.
Sources
- Estuary home page
- Estuary pricing
- Estuary documentation
- Estuary concepts
- Estuary derivations
- Estuary repository and licences (estuary/flow)
Checked November 2026.
How we research reviews and what star ratings mean: our editorial method.