offdataai
Microsoft Purview

Purview Data Governance is retiring — keep your glossary working

Microsoft ends support for Purview Data Governance on November 12, 2027. Bring your glossary terms, data products and owners into offdataai, live from your tenant or from a CSV export, and turn them into definitions AI agents use to answer questions.

  • From $1/month
  • 8 SQL targets
  • 174 templates
Import from Microsoft Purview · previewtext
1source Purview Unified Catalog (read-only)
2read 412 terms · 18 products · 96 CDEs
3
4domains → owning teams
5 Finance 138 terms owner finance-data@
6 Sales 97 terms owner sales-ops@
7 Supply 121 terms steward supply-team@
8
9proposals 401 concepts · 18 data products
10skipped 11 expired terms
11flagged 29 drafts to review
12conflict "Active Customer" Sales ≠ Finance
13
14nothing changed yet · review, then release
What Microsoft announced

Twelve months to move what your teams wrote down

  1. Nov 12, 2026

    Retirement announced

    Purview narrows to data security and compliance. Governance moves to Fabric's OneLake catalog.

  2. Now – mid 2027

    Your window to move

    Inventory terms and owners, settle conflicting definitions, and bring the glossary into its new home.

  3. Nov 12, 2027

    Support ends

    Microsoft ends support for Purview Data Governance.

“Support for Purview Data Governance will end Nov. 12, 2027.” Purview stays focused on data security and compliance; the OneLake catalog in Microsoft Fabric becomes the foundation for governance.

What carries over

Your glossary, mapped one to one

A scanner can rediscover your tables. It can't rediscover what your teams decided those tables mean, or who is accountable for each definition. That is what the import keeps.

In Microsoft PurviewIn offdataai
Glossary termA concept in your ontology, with its definition, acronyms and synonyms
Governance domainThe team that owns the definition, kept per team so Sales and Finance can differ
Owner, expert, stewardOwner name and email on the definition, with Purview recorded as the source
Related terms, parent, resourcesKept as evidence on the concept, with links preserved
Data productA concept whose fields are its critical data elements, with its business use
Critical data elementA typed field on the data product it belongs to
Draft / Published / ExpiredDrafts flagged for review, published terms proposed, expired terms skipped
How the move works

Connect, review, serve to agents

  1. 01

    Connect or upload

    Connect your Purview tenant with an Entra app registration (read-only, metadata only), or upload the glossary CSV Purview exports. Unified Catalog and the classic data map both work.

  2. 02

    Preview, then review

    See every term, data product and team definition before anything is saved. Everything arrives as a proposal; nothing already in the ontology is overwritten.

  3. 03

    Bind and serve

    Bind concepts to real tables in Databricks, Snowflake, BigQuery or Postgres, rule on conflicting definitions, release a version, and serve it to AI agents over MCP.

The CSV import is included from the Starter plan; the live Purview connection is part of Pro. See plans

After the move

From a reference glossary to answers agents can trust

Definitions per team

Finance and Sales each keep their meaning of a term, and a ruling records which one agents use.

Released, not drifting

The ontology is versioned with draft and release, with an audit log of who changed what.

Answers, not just entries

Agents ask in business terms and get SQL built from your definitions and join paths, under row filters and PII masking.

No lock-in, ever again

Export Apache Ossie, OWL, JSON-LD, dbt MetricFlow YAML, Snowflake semantic views or Databricks metric views.

Alongside Microsoft Fabric

The OneLake catalog for access and protection. offdataai for what the data means.

You don't have to choose. Fabric's OneLake catalog covers discovery, permissions and sensitivity labels in the Microsoft estate, and Purview keeps protecting sensitive data. offdataai holds the business meaning your glossary captured, ties it to the tables in any warehouse you use, and serves it to AI agents. Keep your Purview CSV export as a record either way.

Watch the demo

See it in action

A narrated walkthrough, recorded in the real app.

How it works

Three steps, minutes not months

1

Describe or import

Answer a short interview, open a template, draw on the canvas or connect a live database.

2

Review a validated model

Keys, grain, history and PII are checked before any code is written. Refine it in plain English.

3

Ship and keep it current

Native DDL, dbt, seeds, docs and an ontology. Push to Git, diff versions, generate migrations.

Overview

Why teams choose offdataai

app.offdataai.com
A validated ERD: keys, SCD2 history and PII flags on every table

A validated ERD: keys, SCD2 history and PII flags on every table

  1. 01

    Your business glossary is the part of Purview that is hardest to rebuild.

  2. 02

    Tables can be rescanned; the meaning your teams agreed on, and who owns it, cannot.

  3. 03

    offdataai imports that meaning as reviewable proposals, keeps each domain's definitions with the team that owns them, binds them to the tables in your warehouse, and serves them to AI agents with access rules applied.

Capabilities

What you get with the import

  • Live or CSV

    Read the Unified Catalog or classic data map directly, or upload Purview's glossary export.

  • Owners kept

    Governance domains become owning teams, with owner and steward names and emails.

  • Data products intact

    Each data product keeps its critical data elements as typed fields.

  • Review before release

    Preview first; everything lands as a proposal and nothing existing is overwritten.

  • Ready for agents

    Serve definitions over MCP with row filters, PII masking and a log of every query.

0

Validated templates

0

SQL targets

0

Modeling paradigms

0+

Industries covered

Generates native SQL and dbt for 8 platforms · connects live to Snowflake, Databricks, BigQuery, Redshift and Postgres

SnowflakeBigQueryDatabricksRedshiftPostgreSQLSynapseFabricDuckDBdbtSnowflakeBigQueryDatabricksRedshiftPostgreSQLSynapseFabricDuckDBdbt
FAQ

Frequently asked questions

When does Microsoft Purview Data Governance support end?
On November 12, 2027. Microsoft announced it on November 12, 2026, in the post 'A next step in the evolution of the Microsoft Data Governance strategy'. Purview continues with a focus on data security and compliance, and governance moves to the OneLake catalog in Microsoft Fabric.
What does offdataai import from Purview?
Glossary terms with their definitions, acronyms, synonyms, related terms and references; governance domains, which become the teams that own each definition; owners, experts and stewards; data products; and critical data elements, which become fields on their data product. Expired terms are skipped and drafts are flagged.
Live connection or CSV: which should we use?
Either. The live connection reads Purview's Unified Catalog or the classic data map with an Entra app registration you create (tenant ID, client ID and a client secret); it is read-only and reads metadata only, never your data. The CSV route works from a glossary export and needs no credentials. CSV import is included from the Starter plan; the live connection is part of Pro.
Will the import overwrite anything?
No. You preview the import first. Every term and data product then arrives as a proposal for review, and nothing already in your ontology changes. Re-running the import adds only what is new.
Do we have to stop using Microsoft Fabric?
No. The OneLake catalog handles discovery, permissions and protection inside the Microsoft estate. offdataai holds the business meaning, binds it to tables in Databricks, Snowflake, BigQuery or Postgres, and serves it to AI agents over MCP. The two work side by side.
Can we leave offdataai later?
Yes. The ontology exports as Apache Ossie, OWL and JSON-LD, and as dbt MetricFlow YAML, Snowflake semantic views and Databricks metric views.

Part of the offdataai platform

The business context behind accurate AI answers, built from the data model up.

See all six stages →
  1. 01DesignLogical, physical and dimensional models.
  2. 02GenerateSQL, dbt, documentation and migrations.
  3. 03DefineBusiness metrics, semantic models and ontology.
  4. 04ConnectReal warehouse schemas and metadata.
  5. 05GovernDrift, impact analysis, tests and approvals.
  6. 06ActivateMake trusted context available to AI agents.

This page covers Generate and Connect. Every stage works on the same model, so what you design here carries through to the rest.

Accuracy starts with context

Give your AI agents context they can trust.

Design the model, define what your metrics mean, and serve that context to every agent and BI tool, on the warehouse you already run.

Book a demoGet started

Self-serve plans available

Microsoft, Microsoft Purview and Microsoft Fabric are trademarks of the Microsoft group of companies. offdataai is not affiliated with or endorsed by Microsoft. Dates and statements about Microsoft products come from Microsoft's public announcement of November 12, 2026; confirm details for your tenant with your Microsoft account team.