You're on Informatica: IDMC Governs and Moves Your Data. Data Workers Keeps It True Everywhere It Lands
Informatica CDGC catalogs and governs your data and CDI moves it. Data Workers catches what breaks downstream, fixes it with approval and proves the fix held.
Your enterprise data architecture runs on Informatica's Intelligent Data Management Cloud (IDMC). Cloud Data Governance and Catalog (CDGC) scans your sources, builds lineage, holds the business glossary and marks the critical data elements (CDEs) your policies cover; the Data Marketplace hands governed datasets to the people who ask. Cloud Data Integration (CDI) mappings and taskflows move the data from your ERP and CRM into the warehouse every night, and MDM keeps the golden records. Many estates still run PowerCenter beside it. Since Salesforce closed its acquisition on Nov 18, 2025, the site reads "Informatica from Salesforce", and in 2026 CLAIRE agents, Headless Data Management and Informatica's MCP servers bring that governed context to AI. Data Workers does the operations work that keeps it true: it catches what breaks in the systems Informatica feeds, fixes the cause behind approvals and proves the fix held, with a receipt for every change.
Key takeaways
- •Informatica keeps its job. CDGC stays the inventory of record, CDI keeps moving the data, MDM keeps the golden records and your stewards keep their policies.
- •Data Workers operates everything Informatica feeds. Your team's assistant brings CDGC lineage, glossary and IDMC job metadata in from Informatica's MCP servers as proposals, and Data Workers joins them to the runs and tables it reads natively in Snowflake and dbt.
- •CDI taskflows stay the integration owner's. Data Workers never starts an ingestion run; it proposes the change, queues the downstream rerun through your orchestrator and verifies.
- •One plan, one approval, one receipt. A named owner approves in Spellbook, and an unanswered request expires and escalates, never auto-grants.
- •Autonomy is set per domain, from L0 manual to L4 autonomous.
Informatica is the system of record and the integration engine. Data Workers is the operator for everything they feed.
Informatica records what your data is and moves it where it needs to go. Data Workers acts on what happens next. A Wednesday of quarter-end close at a medical device maker: Oracle Fusion Cloud ERP holds the general ledger, an IDMC CDI taskflow loads journal lines into Snowflake, dbt Cloud builds the finance close models, a Power BI semantic model called Close Cockpit feeds the CFO's 08:00 flash revenue report, CDGC holds lineage and the Legal Entity CDE, alerts go to a Microsoft Teams channel and tickets live in ServiceNow. An illustration, not a customer case.
| Time | System | What happens |
|---|---|---|
| Tue 17:45 | IDMC Cloud Data Integration | To match the CDGC glossary standard, an integration developer deploys a new version of mapping m_fusion_gl_lines: target column ENTITY_CD becomes LEGAL_ENTITY_CODE. The Snowflake target adds the new column and keeps the old one |
| Wed 01:30 | Oracle Fusion and IDMC | Taskflow tf_fusion_gl_close loads 412,000 journal lines for day 2 of close into FUSION_RAW.GL_JOURNAL_LINES; every new row has ENTITY_CD empty and the code in LEGAL_ENTITY_CODE |
| 02:15 | dbt Cloud and Snowflake | The finance_close job runs; stg_fusion__gl_lines still selects ENTITY_CD, the inner join in fct_gl_balances drops all 412,000 lines, and the not_null test on entity_cd is set to warn, so the job finishes green |
| 03:10 | Data Workers | Sees zero Wednesday lines for all 22 legal entities in fct_gl_balances against a 30-day baseline; run_quality_check on the raw table fails: ENTITY_CD is empty on every row since 01:31, and the IDMC job metadata shows LEGAL_ENTITY_CODE added at that load. Posts a card to the finance data channel in Teams and opens a ServiceNow ticket |
| 03:16 | Data Workers | Reads CDGC over Informatica's API: lineage maps ENTITY_CD to m_fusion_gl_lines, the mapping changed Tue 17:45, and Legal Entity is a CDE with a named finance data owner. trace_cross_platform_lineage follows the dbt manifest to Close Cockpit; blast_radius_analysis finds 7 dbt models, the 08:00 flash report and the intercompany elimination extract at risk |
| 03:30 | Data Workers | Proposes one plan in Spellbook: a dbt diff to stg_fusion__gl_lines that reads LEGAL_ENTITY_CODE with ENTITY_CD as fallback and raises the not_null test to error; a hold on the 07:00 Power BI refresh; a dbt Cloud rerun after merge; a note to the CDGC owner to rescan the mapping. No CDI rerun: the data loaded, in the new column |
| 06:40 | Spellbook | The finance data owner reviews the diagnosis, the blast radius and the plan, and approves |
| 06:45 | Power BI | The BI owner pauses the 07:00 scheduled refresh |
| 06:55 | GitHub | The analytics engineer merges the dbt diff |
| 07:00 | dbt Cloud | The finance data owner queues the approved finance_close job, recorded in Spellbook, and reports all tests passing, the not_null test now at error |
| 07:25 | Snowflake | Data Workers verifies: lines for all 22 entities within baseline, debits equal credits per entity in fct_gl_balances, and the row count matches the CDI load |
| 07:35 | Power BI | The BI owner runs the refresh; Close Cockpit shows every entity |
| 07:40 | Teams and ServiceNow | Data Workers posts the resolution card and updates the ticket summary with the receipt link; the service desk owner resolves the ticket |
| 10:00 | CDGC | The CDGC owner rescans; lineage now shows LEGAL_ENTITY_CODE feeding Net Revenue |
Informatica did its jobs: CDI loaded every line on time and CDGC knew which mapping fed the Legal Entity CDE. The break sat downstream: the rename followed the glossary, the dbt project lived outside IDMC and the test only warned. Data Workers joined those pieces before dawn, one owner approved one plan, and the CFO's numbers were right at 08:00.
| Job | What Informatica does | What Data Workers does |
|---|---|---|
| Inventory | CDGC scans sources, builds lineage, holds glossary terms, CDEs and policies | Reads that context as a first-class source and joins it to live runs, tests, usage and grants across every platform |
| Moving data | CDI mappings and taskflows load the warehouse; CLAIRE ELT Agents help build pipelines | Watches what each load does downstream and queues dependent reruns through your orchestrator |
| Detecting a break | Lineage shows what a column feeds as of the last scan | Catches the volume gap and the emptied column with warehouse quality checks, and traces both through the models built on them |
| Root cause | Shows which mapping writes which column inside IDMC | Traces from the mapping through dbt, the warehouse and BI to the report at risk |
| The fix | The integration owner edits the mapping; stewards update terms | Proposes the complete repair as one plan, each step routed to its owner: the dbt diff to merge, the refresh to hold, the rescan to run |
| Verification | The next scan refreshes lineage | Checks counts and balances on every table the fix touched, with the owner's test results |
| The record | Policies, lineage and stewardship history in CDGC | A receipt: cause, changes, approver, what they touched, how they were verified and how to undo them |
Why doesn't Informatica just do this itself?
Because Informatica built a data management suite, and it acts where it is in charge. Inside IDMC its reach is wide and getting wider: Informatica describes CLAIRE agents that "discover data, build pipelines and proactively fix data quality issues", the Data Quality Agent has been GA since Spring 2026, and on May 20, 2026 it announced a Metadata Enrichment Agent, a Data Steward Agent and Agentic Integration for Q4 2026. Its MCP servers, GA since the Fall 2025 release, connect agents to CDGC search, MDM and address verification, and the Informatica plugin for Claude is grounded in catalog metadata. We concede that surface happily: inside IDMC, Informatica is deep.
The Wednesday fix needed more than IDMC. The code to change lived in a dbt project in GitHub, the run to queue was in dbt Cloud, the refresh to hold belonged to the BI team and the record of the incident sat in ServiceNow. A governance suite that edited another team's dbt models or paused their BI refresh on its own would take on liability for systems it does not run, and the trust a catalog of record depends on. Acting across them takes blast-radius scoping, a named owner's approval, an undo written first, verification beyond the catalog and a receipt an auditor can read. That is a different product, and it is the product Data Workers is.
There is also focus. Informatica's 2026 direction is trusted data for AI agents across Salesforce's Data 360 and Agentforce: a sensible bet for a suite its size. Running the whole data lifecycle across every other vendor's tools is the job we chose.
Every tool owns a slice. Data Workers covers the whole lifecycle
Informatica goes deepest on catalog and governance, with integration close behind. Each point tool adds another console, contract and handoff. Data Workers covers the whole lifecycle with one context, one approval flow and one audit trail.
| Stage | Data Workers | Informatica | Why we scored it this way |
|---|---|---|---|
| Catalog & Context | 9 | 9.5 | Informatica's home stage: Cloud Data Governance and Catalog (CDGC) harvests metadata, lineage, glossary terms and CDEs across sources, and its catalog capabilities now publish as MCPs into Agent Fabric Context Catalog, which governs data assets and AI agents together. Data Workers joins that context to live runs, tests, usage and grants in one governed graph. |
| Analytics & Insights | 8 | 5 | CLAIRE GPT lets people ask about data in natural language. Data Workers answers data questions from governed definitions with lineage behind every number. |
| Data Quality | 8 | 7.5 | This page scores CDGC and CDI, where quality shows as scores on CDEs; Cloud Data Quality and its Data Quality Agent (GA Spring 2026) win this stage on our Informatica Data Quality page. Data Workers runs checks too and fixes the cause when one fails. |
| Observability & Incidents | 8.5 | 5 | Data Observability watches data health inside IDMC. Data Workers diagnoses across systems, fixes with approval and verifies the fix. |
| Pipelines & Ingestion | 8.5 | 7.5 | Cloud Data Integration mappings and taskflows move data at enterprise scale, and CLAIRE ELT Agents help build them. Data Workers queues reruns through your orchestrator and proposes ingestion changes for their owner. |
| Schema & Migration | 8 | 5 | CDGC shows what a changed column feeds after its next scan. Data Workers catches the change with a warehouse quality check, scores its blast radius and drafts the fix with rollback. |
| Governance & Access | 8.5 | 9 | Informatica's second home stage: CDGC policies, stewardship, data access management and the Data Marketplace. Data Workers dry-runs each warehouse access request and proposes a time-bound grant for the owner. |
| Security & Privacy | 8 | 7 | Classification and data privacy services sit in the suite. Data Workers' pull request review flags new columns whose names or annotations look sensitive, and leaves a receipt on every data change. |
| Cost / FinOps | 8 | 2 | Governance and integration are not cost tools. Data Workers reads warehouse spend next to each incident and drafts setting changes for their owner. |
| MLOps & Models | 7.5 | 5 | Trusted data for AI agents across Data 360 and Agentforce is the 2026 theme. Data Workers keeps the data under your models fresh and correct. |
Informatica wins its two home stages, and we scored the suite's integration and quality breadth where it shows. For the category view, see Data Workers vs a data catalog, data catalog vs context layer and what an agentic data catalog is.
How Informatica and Data Workers work together
Spellbook Data Catalog (in preview) is where the data team looks: each proposed change, who approved it and how to roll it back. Behind it, Data Context Wizard keeps one governed context graph across the estate with CDGC as a first-class source, the Data-Agents Swarm does the work with 20+ specialist agents, the Autonomous Data-Conductor runs each fix end to end, and per-domain guardrails hold approvals, receipts and rollback.
Informatica to Data Workers. Informatica connects over its API or MCP servers today: your team's assistant reads CDGC lineage, glossary terms, CDEs and owners, and IDMC asset and job metadata, side by side with Data Workers, and hands what matters to Context Wizard as proposals. It reads Snowflake, dbt Cloud, Teams and ServiceNow natively and reaches Power BI and Oracle Fusion over their APIs; the full list is on Data Workers integrations. A schema change or a volume gap starts a diagnosis: diagnose_incident and get_root_cause work the evidence, trace_cross_platform_lineage follows CDGC lineage into the dbt manifest and on to BI, blast_radius_analysis maps what is at risk, explain_table pulls each table's definition, lineage and documentation, and get_incident_history checks for repeats. The repair runs through remediate: code changes go to the owner as a diff to merge, orchestrator reruns are queued, dbt Cloud reruns go to the owner after approval, and CDI taskflows stay with the integration owner. Verification uses run_quality_check and get_quality_score, and every step lands in get_audit_trail.
Data Workers back to Informatica. Informatica keeps the lineage, the glossary and the policies. Catalog changes Data Workers suggests, such as a rescan after a mapping change or a term mapped to a new column, go to the CDGC owner; approved facts land in the Context Wizard graph. The receipt lives in Spellbook and the audit trail, linked from the ServiceNow ticket.
Side by side in one client. Informatica's Headless Data Management, GA since Spring 2026, exposes its services through native MCP support, and the Informatica plugin for Claude Code (informatica-for-claude-platform in the claude-plugins-official marketplace) adds a CDGC catalog discovery server that you point at your tenant. Add Data Workers beside it, one start-agent.sh entry per agent, as the client setup guide documents. Example for Claude Code's .mcp.json, Data Workers entries only:
{
"mcpServers": {
"dw-context-catalog": { "command": "/path/to/dataworkers-claw-community/start-agent.sh", "args": ["dw-context-catalog"] },
"dw-schema": { "command": "/path/to/dataworkers-claw-community/start-agent.sh", "args": ["dw-schema"] },
"dw-incidents": { "command": "/path/to/dataworkers-claw-community/start-agent.sh", "args": ["dw-incidents"] }
}
}Ask "who owns the Legal Entity CDE, and why did Wednesday's journal lines disappear?" and the client calls both: Informatica returns the owner and certified definition from CDGC; Data Workers returns the mapping change, the dbt join that dropped the lines, the blast radius and a plan. Informatica's access controls govern IDMC; Data Workers' guardrail decides whether a data change may run in that domain, with a named approver. A shared Data Workers endpoint takes an API key or OAuth tokens from your identity provider, such as Okta or Entra ID, verified through JWKS.
One renamed column, L0 to L4. The same incident at each level, set per domain.
- •L0 manual. A controller spots the gap in Close Cockpit at 08:05 and the hunt starts across four teams.
- •L1 observe. Data Workers posts the diagnosis and the flash report at risk at 03:16. Nothing changes.
- •L2 propose. Data Workers proposes the full plan; nothing runs until the finance data owner approves in Spellbook.
- •L3 act reversibly. For proven change classes, such as rerunning the close models after a merged fix, Data Workers queues the step, verifies it and records the receipt.
- •L4 autonomous. In a scoped domain with a long clean record, Data Workers runs the repeatable steps end to end; CDI mappings, CDGC and BI refreshes stay with their owners.
On safety, see is it safe to let AI agents change production data, how approvals work, autonomy levels and rollback. On where data lives: the agents run in your infrastructure, your data stays in your systems, and the hosted Conductor sees workflow metadata only.
What changes for your team
Informatica gave your governance office the map. Data Workers works from it.
- •Incidents. A change that breaks a governed CDE gets a cross-system diagnosis, a proposed fix and a receipt.
- •Data quality. A failed check on a CDE becomes a fixed cause, verified before the next run reads it.
- •Cloud spend. Warehouse spend sits next to each incident, and setting changes go to their owner drafted.
- •Access. A warehouse access request is dry-run and proposed as a scoped, time-boxed grant for the data owner to approve.
- •Audits. CDGC records the policy and the lineage; Data Workers records the fix, the approver and the undo.
- •Migrations. A move off PowerCenter runs in approved waves, with parity checks planned and tracked and the completion gate held for the owner's sign-off.
The integration developer changes most: a glossary-driven mapping change no longer breaks a team they never met. See the data incident response playbook and who owns the agents.
Keep Informatica, or consolidate?
Keep Informatica if you love it; Data Workers works with it from day one. Many teams consolidate once Data Workers runs that slice too.
Most Informatica estates keep IDMC: CDGC maps to the governance program and CDI carries years of mappings. What teams consolidate is the work around it: a second lineage tool to see past IDMC, scripts that check the warehouse after each load, and the email thread that is the only record of last quarter's fix. Thinking of building that layer yourself on Informatica's MCP servers? Read build it ourselves with Claude Code and MCP servers: the calls are easy; the context graph, approvals and rollback are the work.
For the rest of this stack, see you're on Informatica Data Quality, you're on Informatica PowerCenter, you're on Collibra, you're on dbt and what is an agentic data platform. Planning a move off legacy Informatica jobs? See automating Teradata and Informatica migrations.
The case for your CFO
The outcome: your governance program already knows what every CDE means. Data Workers makes that knowledge act. When a change upstream breaks a number downstream, the cause, the blast radius and one plan arrive before anyone opens the report, so the close, the regulatory file and the board pack run on correct data.
The risk story: autonomy is set per domain. At L1 Data Workers only reads; at L2 it changes nothing until a named person approves; at L3 it applies changes it can undo. CDI taskflows, CDGC and BI refreshes stay with their owners. An unanswered approval request expires and escalates, never auto-grants. No agent can promote its own work. Every change carries a receipt: cause, change, approver, verification and undo. An org-wide stop halts all autonomous dispatch. Zero migration: every system stays where it is.
Why now: every AI agent that reads Informatica's governed context over MCP raises the cost of a number that is quietly wrong. The first win is read-only: every schema change and volume gap on the finance domain gets a cross-system diagnosis and a blast radius. What stays the same: your glossary, policies, stewards, mappings and schedules. For the numbers, see the ROI of agentic data operations.
The sentence to repeat upstairs: "Informatica tells us what our data means and moves it; Data Workers makes sure it stays right everywhere it lands, with an approval and a receipt."
Getting started
Start with a pilot. Pick one governed domain, such as finance close, connect Data Workers to Informatica, your warehouse, dbt and BI, and run at L1 so every break gets a diagnosis and a blast radius. Then turn on a first write class at L2, such as dbt fixes proposed as diffs for the model owner. The pilot path and plans are on the pricing page, and the pilot is credited in full against the first year.
FAQ
Is Informatica now part of Salesforce? Yes. Salesforce completed its acquisition of Informatica on Nov 18, 2025, and the brand is now "Informatica from Salesforce". IDMC, CDGC, CDI and MDM continue as products, with new connections to Data 360.
How does Data Workers connect to Informatica? Over Informatica's API or MCP servers today, reading CDGC lineage, glossary terms, CDEs, owners and IDMC job metadata. Snowflake, Databricks, BigQuery, dbt, Airflow, Teams and ServiceNow are native. You can also run Informatica's MCP server and Data Workers side by side in one client.
CLAIRE agents already build pipelines and fix data. Why add Data Workers? CLAIRE agents work inside IDMC, and they do it well. A break often lands outside it, in a dbt model, a warehouse table or a BI refresh, and someone has to find it, fix it in every system it reached and prove the fix held. That is the job Data Workers does, under one approval.
Does Data Workers run our CDI taskflows, edit our mappings or write to CDGC? No. Taskflows and mappings stay with your integration owner, and catalog changes such as a rescan go to the CDGC owner. Data Workers proposes changes with the evidence attached, queues downstream reruns through your orchestrator, hands dbt Cloud reruns to the owner, and verifies the result.
Where does our data go? The agents run in your infrastructure and hold the warehouse credentials and model key. Your data stays in your systems; the hosted Conductor sees workflow metadata only, such as table names, proposals and approval records.
Sources
- •Salesforce, "Salesforce Completes Acquisition of Informatica" (Nov 18, 2025), https://www.salesforce.com/news/press-releases/2025/11/18/salesforce-completes-acquisition-of-informatica/ (checked Oct 3, 2026)
- •Informatica, Cloud Data Governance and Catalog product page ("Informatica from Salesforce"), https://www.informatica.com/products/data-governance/cloud-data-governance-and-catalog.html (checked Oct 3, 2026)
- •Informatica, CLAIRE AI (CLAIRE agents, CLAIRE GPT, CLAIRE Copilot), https://www.informatica.com/platform/claire-ai.html (checked Oct 3, 2026)
- •Informatica, Headless ("MCP servers, APIs and A2A services"; plugin for Claude "Now live"), https://www.informatica.com/headless.html (checked Oct 3, 2026)
- •Informatica, "Informatica Announces Fall 2025 Release" (Oct 29, 2025: MCP Servers GA; CLAIRE agent status), https://www.informatica.com/about-us/news/news-releases/2025/10/20251029-informatica-announces-fall-2025-release-with-latest-innovations-to-intelligent-data-management-cloud.html (checked Oct 3, 2026)
- •Informatica, "Informatica from Salesforce Delivers the Trusted Data Foundation Every AI Agent Needs" (May 20, 2026: Headless Data Management and Agent Fabric Context Catalog GA Spring 2026; Metadata Enrichment Agent, Data Steward Agent and Agentic Integration Q4 2026), https://www.informatica.com/about-us/news/news-releases/2026/05/20260520-informatica-from-salesforce-delivers-the-trusted-data-foundation-every-ai-agent-needs-now-across-every-surface-every-platform-everywhere.html (checked Oct 3, 2026)
- •Informatica plugins for Claude Code, GitHub forcedotcom/informatica-claude-plugins (Apache 2.0; created Sep 1, 2026), https://github.com/forcedotcom/informatica-claude-plugins (checked Oct 3, 2026)
- •Data Workers open-source repository, https://github.com/DataWorkersProject/dataworkers-claw-community (checked Oct 3, 2026)
- •Data Workers client setup guide, https://dataworkers.io/opensource-docs/client-setup/ (checked Oct 3, 2026)