Benefits of Lakeflow Connect for data ingestion

The future of data engineering on Databricks is with Lakeflow: the unified solution for ingestion, transformation and orchestration. You can use Lakeflow Connect to ingest data from any source, Apache Spark™ Declarative Pipelines to transform your data, and Lakeflow Jobs to orchestrate any workflow. Together, these enable customers to deliver fresher and higher-quality data for their business.

Lakeflow Connect provides 100+ native connectors for ingestion from enterprise applications, databases, enterprise knowledge sources, and events. It offers a point-and-click UI and unified governance via Unity Catalog, feeding AI agents with complete enterprise context. Lakeflow Connect leverages efficient incremental reads and writes to make data ingestion faster, more scalable and more cost-efficient, while your data remains fresh for downstream consumption. Under the hood, Lakeflow Connect pipelines are built on Apache Spark™ Declarative Pipelines. Moreover, many of the connectors read and write only the data that's changed in the source system. Finally, we leverage Arcion's source-specific technology to optimize each connector for performance and reliability while also limiting impact on the source system.

Across the fast growing list of Lakeflow Connect connectors are several connector types: SaaS connectors, Database connectors, File source connectors, and Streaming connectors. For the complete, up-to-date list, refer to the documentation. Managed connectors are in various release states, and standard connectors offer higher levels of ingestion pipeline customization. Note that Lakeflow Connect uses serverless compute for Lakeflow Declarative Pipelines. For guidance, refer to the Lakeflow Connect documentation.

Yes. We offer a permanent Free Tier where every workspace receives 100 DBUs per day at no cost, allowing you to ingest up to 100 million records per day across eligible managed sources. For more details, check out the pricing page.

Lakeflow Connect is built for data engineering and data professional teams who need to bring enterprise data into the Data + AI Platform without heavy custom coding. Its no-code UI and simple API let practitioners self-serve data ingestion, while deep integration with Unity Catalog and Lakehouse Monitoring gives governance and observability teams the controls they need. Organizations apply it across use cases such as customer 360 analytics, centralized HR workforce dashboards, manufacturing digital twins, intelligent document processing, and consumer analytics. Because ingestion is fully integrated with the platform, it also feeds AI agents with the complete business context they need for trusted reasoning.

Customers moving off legacy ETL stacks and custom middleware report meaningful efficiency gains after adopting Lakeflow Connect. Panasonic's BI team cut Power BI refresh times by 50% after moving from a rigid legacy ETL stack to the Databricks Data + AI Platform, according to its case study. Insulet eliminated fragile, problematic middleware between Salesforce and Databricks SQL, reducing data latency from days to minutes, and Kotahi cut the time to extract and prepare Salesforce data from roughly three hours to 30 minutes after replacing a third-party ingestion tool. Because connectors are fully managed, teams also avoid the ongoing pipeline maintenance that hand-built ETL or middleware layers typically require.

You can start using Lakeflow Connect for free by following the getting-started documentation, which walks through setting up your first ingestion pipeline. Every Databricks workspace also includes a permanent Free Tier with 100 DBUs per day, enough to ingest up to 100 million records daily across eligible managed sources. If you'd like to see the product first, take the Lakeflow Connect feature tour, and when you're ready to scale, review the usage-based pricing details.