August to October 2026 Product Updates: Lineage Into Six BI Tools, Faster Postgres CDC, and Matia MCP

Tracing an incorrect number on a dashboard shouldn't take an afternoon of Slack messages.

When a number looks off, most teams can’t just log into one solution to find the root cause. Their ETL, Observability, and Catalog tools each see one piece of the puzzle. Matia runs ETL, Observability, Catalog and Reverse ETL, in one platform, so it can trace a number from the dashboard to the source system making root cause analysis simple.

From August through October, we pushed our capabilities further upstream and further downstream. Lineage now reaches Hex, Power BI, Sigma (Joining Tableau, Metabase & Omni), and the transformations you run inside Snowflake; while faster Postgres CDC and new connectors centralize your data. Matia MCP went GA, so you can ask what broke and trace it back to the source, right from the AI assistants you already log into every day.

ETL: Faster Postgres and More Sources

• Postgres WAL streaming: Matia's Postgres connector now streams changes straight from the Write-Ahead Log over Postgres' own replication protocol. In our benchmarks it ran over 4x faster on Aurora and about 2x faster than the way the connector read the WAL before, with less memory and I/O load on your production database. How it works.
• PlanetScale and self-hosted Vitess: Incremental syncs read from VStream, so keeping your warehouse current doesn't mean reloading full tables. Setup takes one read-only password and no changes to your production database. Read the launch post.

Twenty new application sources shipped this quarter too, from HR systems like Workday HCM to finance tools like Coupa and Navan.

• BambooHR
• Brex
• Confluence
• Coupa
• Forethought
• Hark
• Hibob
• iSpot.tv
• Langfuse
• Navan
• Optimizely
• Pardot
• Pigment
• Salesforce Commerce Cloud
• Salesforce Marketing Cloud
• Sentry
• Superfiliate
• Swarmia
• Workday HCM

Reverse ETL: Send Warehouse Data Where Teams Work

• Linear destination: Push customer records from your warehouse into Linear, so product and support teams work from the same data without a manual sync. See the docs.

Catalog: Lineage Into Six BI Tools and Snowflake

Matia's Catalog now covers six BI tools in one lineage graph: Metabase, Tableau, Omni, Hex, Power BI, and Sigma. Three of them shipped this quarter.

• Hex: Lineage runs through SQL cells down to the column, and dataframes from Python cells get picked up when a later cell references them. Every team has that one notebook that's tribal knowledge with better formatting. Now it has lineage. Read more.
• Power BI: Native lineage traces datasets, reports, and dashboards down to the column, upstream through Snowflake and Databricks. Descriptions and usage metrics carry over too. Read more.
• Sigma: Trace workbooks, elements, and columns back to the Snowflake tables behind their Custom SQL, and see which workbooks use a column before you change it. Read more.
• Snowflake transformations: Stored procedures and dynamic tables now appear as their own nodes, with column-level lineage running through them. Matia reads it straight from Snowflake Horizon, rolling out for Snowflake Enterprise edition and above. Read more.

When your data moves through Matia's ETL, that lineage goes past the warehouse to the source a number started in.

Matia MCP Is GA: Ask Matia From Where You Work

Matia MCP brings Matia's context into the AI assistants you already use. It works in any AI tool that supports MCP, like Claude, and connects to your catalog, lineage, monitors, and integrations under each person's existing Matia roles. Ask what a change will affect, or trace a broken dashboard back to its source, without leaving the tool you're in. Running a sync or toggling a monitor still needs your confirmation.

In October, that context reached Snowflake Cortex Code and Cortex Agents. Before an agent rewrites the procedure that builds a table, it can ask Matia to surface dependencies. Read the MCP launch post.

Available Now

Everything above is live. Every source and BI tool you connect lands in the same lineage graph as the syncs and monitors around it giving you one unified platform.

Book a demo to see how Matia can unify your stack, or head to the docs to set up any connector.