G2

Mohamed A.

IT System & Application Analyst

Mid-Market (51-1000 emp.)

9/25/2026

"API-First Governance That Fits Microsoft Fabric Seamlessly"

4/5

What do you like best about OpenMetadata?

We run most of our analytics on Microsoft Fabric, and honestly finding a governance tool that didn't fight the Fabric ecosystem was harder than expected. OpenMetadata's API-first approach made it possible to pull column-level lineage straight from Fabric pipelines without installing anything custom on the extraction side, which was a big deal for us since we didn't want another agent or connector adding overhead to production pipelines. The self-hosted deployment gave us full control, and the ingestion framework was flexible enough that we could catalog our lakehouse assets and start layering data contracts and quality checks as code on top without much friction. The part that's been genuinely useful beyond documentation is exposing that governance context through an MCP server, so an AI agent can actually query lineage and quality context instead of us guessing from table names. Review collected by and hosted on G2.com.

What do you dislike about OpenMetadata?

The learning curve on the ingestion framework config isn't small if you're doing anything beyond the standard connectors, we had to spend real time understanding the YAML structure and debugging a few ingestion pipeline failures before things ran cleanly. Documentation on advanced customization (custom connectors, MCP-specific setup) is thinner than the core product docs, so we ended up relying on community channels and trial and error more than I'd have liked. Also the UI can feel like it's doing a lot at once when you're onboarding new team members who haven't worked with a catalog tool before. Review collected by and hosted on G2.com.

What problems is OpenMetadata solving and how is that benefiting you?

We had zero visibility into lineage after migrating to Fabric, and no consistent way to enforce data quality or track it back to source when something broke. OpenMetadata gave us a single place for lineage, cataloging, contracts, and quality observability, and now that we're exposing it through MCP, it's also becoming the context layer our AI agents rely on instead of guessing from raw schema. That's cut down a lot of manual back-and-forth on "where did this number come from" and given us a real answer to data trust questions we couldn't answer before. Review collected by and hosted on G2.com.