Mira Solène 15 min readFor fifteen years, "we run Informatica" was the safest sentence a data leader could say in a budget meeting. In 2026 it has become the start of a longer conversation. Salesforce now owns the company, the on-premises PowerCenter line is past its standard support date, and the AI projects every board is asking about have turned governance from a compliance chore into a delivery dependency. The result is a wave of teams typing the same search: what are the alternatives to Informatica?
This guide answers that question as an independent publisher. The Curator has no commercial relationship with any vendor named here, nobody paid for placement, and the order of the platforms below is not a ranking. Every date and product claim links to the source we read it from, checked on 20 November 2026. Where a fact comes from a vendor about itself, we say so.
Short answer: Most teams are not replacing "Informatica"; they are replacing one or two of its services. If you need a governance process, look at Collibra. If you need adoption and cross-platform context, look at Atlan or Alation. If most of your data already lives in one cloud platform, its native catalog may cover more than you expect. If you have engineers and want control, DataHub and OpenMetadata are serious options. Master data management and pipeline integration are the two pieces almost nobody on this list replaces.
Key takeaways
- The ownership change is real, and so is the product investment. Salesforce completed its acquisition of Informatica on 18 November 2025, and the first major post-acquisition release in May 2026 added headless, MCP-accessible data management and a generally available CLAIRE Data Quality Agent. Staying is a legitimate option, not a failure.
- PowerCenter is the urgent case. Standard support for PowerCenter 10.5.x ended on 31 March 2026, and published migration guides put the end of paid extended support at 31 March 2027. That is a pipeline problem first and a governance problem second.
- Even the alternatives are moving. Microsoft has stopped new sign-ups for Purview Unified Catalog and is consolidating data governance into Microsoft Fabric's OneLake catalog, with existing capabilities in maintenance mode through 12 November 2027.
- The industry is splitting into two layers: governance built into the data platform (Unity Catalog, Snowflake Horizon, OneLake, Google's Knowledge Catalog, Amazon SageMaker Catalog) and cross-estate governance layered on top (Collibra, Atlan, Alation, DataHub, OpenMetadata).
- Culture beats tooling. In a Gartner survey of 223 data and analytics leaders, cultural resistance outweighed funding as the main reason governance initiatives fail, 60% to 40%. No platform on this page fixes that on its own.
Why 2026 is the year everybody is re-evaluating
Three separate clocks started ticking at roughly the same time, and together they explain the surge of interest.
1. A new owner. Salesforce's own release describes bringing Informatica's catalog, integration, governance, quality, privacy, metadata and MDM services "to the Salesforce platform" to serve as a data foundation for its Agentforce AI agents. In May 2026 the product, now branded "Informatica from Salesforce", announced headless data management exposed through the Model Context Protocol, autonomous data management agents and what it calls a unified agent and context catalog. Customers with heavy Salesforce estates may read that as good news. Customers whose data lives mostly elsewhere often want to understand their options before the next renewal.
2. An end-of-support date. PowerCenter, the on-premises ETL product that brought many organisations to Informatica in the first place, reached end of standard support for version 10.5.x on 31 March 2026. Informatica's own path forward is its cloud platform, IDMC, and the company markets reuse of "up to 100%" of PowerCenter assets during modernisation. Its knowledge base also notes that the PowerCenter end-of-support date does not apply to Cloud Data Integration for PowerCenter. Any migration conversation therefore forces a second question: if we are moving pipelines anyway, do we keep the governance stack too?
3. AI made governance a delivery dependency. Gartner's data governance guidance warns that by 2027, 60% of organisations will fail to realise the anticipated value of their AI use cases because of incohesive data governance frameworks. Regulation is adding pressure: following the EU's Digital Omnibus on AI, the AI Act's rules for stand-alone high-risk systems in Annex III now apply from 2 December 2027, and those for high-risk AI embedded in regulated products from 2 August 2028. Article 10 of the Act sets data and data governance requirements for high-risk systems, so lineage, documentation and quality evidence stop being nice-to-haves.
Calculated by The Curator from the dates in the sources below. Your own contract dates may differ.
Step one: decide which Informatica you are replacing
The most expensive mistake in this category is shortlisting catalogs when the thing you actually depend on is masking, MDM or pipelines. Informatica is a suite. Before you take a single vendor call, map what you use today against what an alternative would really need to do.
| What you use today | What it does | Typical replacement | How hard |
|---|---|---|---|
| Data Governance & Catalog | Glossary, ownership, search, policies, governance workflow | Any platform in this guide; this is the contested core | Moderate |
| Lineage (incl. CLAIRE-assisted) | Column and table lineage across sources, ETL and BI | Most platforms; quality varies sharply by connector | Moderate to hard |
| Data Quality | Rules, cleansing, profiling, scorecards | Native in some platforms, separate licence in others, warehouse-scoped in others | Moderate |
| Data access & privacy policies | Masking and filtering enforced in the data platform | Often the warehouse or lakehouse itself (Unity Catalog, Snowflake Horizon) | Hard |
| MDM and 360 applications | Golden records, match and merge | Very few governance platforms; Ataccama is a notable exception | Very hard |
| Data Integration / PowerCenter | Moving and transforming data | Not a governance product at all; ELT tools, dbt, cloud-native pipelines or IDMC | Project-sized |
If your answer is "mostly the first three rows", you have a wide field of options. If the access-policy or MDM rows are load-bearing, expect a partial replacement, and say so in your first vendor call rather than in month four.
The five families of alternatives
Comparing twelve products feature by feature produces a spreadsheet nobody reads. It is more useful to sort them by design philosophy, because the philosophy predicts how a product will behave in your organisation far better than a checklist.
1. Governance-workflow suites: Collibra
Collibra is the option built around governance as a managed process: named owners, stewardship, approval workflows and an audit trail of who decided what. In November 2026 it made Collibra Control Tower generally available, a no-code engine that continuously checks the catalog against your governance standards, ships with more than 100 pre-built controls covering BCBS 239, AI governance, GDPR, critical data elements and data products, and routes failures to the right person by email, Slack or Teams. Collibra also offers two-way metadata sync with Amazon SageMaker Catalog.
Best fit: regulated organisations with a real governance function that needs evidence, not just a search box. Watch out for: the operating model. Collibra rewards organisations that already have stewards; it does not create them. Check whether data quality and observability sit in a separately licensed product for your deployment.
2. Active-metadata and context platforms: Atlan and Alation
These two compete on adoption: making the catalog something analysts and engineers open voluntarily because it is where context lives.
Atlan leans into the modern cloud stack and, in November 2026, announced a collaboration with Microsoft to extend OneLake catalog governance across the rest of the data estate. Atlan says Purview glossaries, classifications and lineage can carry over on the Apache Atlas model, which makes it a natural shortlist entry for teams affected by the Purview retirement. Its in-warehouse data quality engine is documented for Snowflake, Databricks and BigQuery, so confirm coverage if your critical data lives elsewhere.
Alation started as the analytics-first catalog and has spent 2026 expanding into AI governance: a dedicated AI Governance offering in May that registers models, agents and tools, maps them to regulations and generates model cards, followed in September by six new components for its "AIOS" platform, including ontologies and semantic model mastering. A published vendor comparison that quotes Alation's documentation notes its data quality monitoring is a separate purchase and that anomaly metrics enter a 30-day warm-up period, which matters for pilots.
Best fit: organisations whose main problem is that nobody can find or trust data, and that run several platforms at once. Watch out for: treating search adoption as governance. Ask how policies are enforced, not just displayed.
3. Platform-native catalogs: govern where the data lives
The biggest structural shift of the last two years is that every major data platform now ships a serious catalog of its own, and the hyperscalers are pulling governance closer to storage.
- Databricks Unity Catalog was open-sourced in June 2024 under Apache 2.0 and hosted at the LF AI & Data Foundation. It supports the Iceberg REST catalog and Hive Metastore interfaces and governs tables, files, functions and AI models together.
- Snowflake Horizon Catalog is built on Apache Polaris, the Iceberg REST catalog Snowflake donated to the Apache Software Foundation in 2024. In October 2026 Snowflake announced a public preview of fine-grained access control for reads and writes on Iceberg tables via the Iceberg REST Catalog API, so policies can follow data to external engines.
- Google Knowledge Catalog is the new name for Dataplex Universal Catalog. Google says existing Dataplex deployments and APIs keep working and that discovery, lineage, data quality and business glossary features are unchanged, while adding a context graph and MCP integrations for AI agents.
- Amazon SageMaker Catalog, built on Amazon DataZone, provides a business catalog, fine-grained access policies, data quality monitoring, sensitive data detection and lineage, and since March 2026 syncs metadata with Atlan, Collibra and Alation.
- Microsoft Fabric OneLake catalog is now Microsoft's primary experience for governing Fabric data, while Purview continues as the security and compliance layer. Purview Unified Catalog is closed to new accounts and in maintenance mode through 12 November 2027.
Best fit: estates where 70% or more of the important data sits on one platform. Watch out for: the edges. Native catalogs govern their own platform brilliantly and everything else less so; ask how BI tools, SaaS applications and legacy databases appear.
4. Open-source metadata platforms: DataHub, OpenMetadata and friends
DataHub began at LinkedIn and is developed in the open under Apache 2.0; the company behind it, formerly Acryl Data, rebranded as DataHub in 2025. Independent comparisons describe the core project as including a large connector library, column-level lineage, glossary and data contracts, with observability monitors and some workflow features in the paid DataHub Cloud.
OpenMetadata is also Apache 2.0 at its core and builds data quality tests and profiling into the open-source product, with Collate offering a managed service. Alongside them, Apache Polaris and Unity Catalog OSS give you open, vendor-neutral table catalogs that pair well with either.
Best fit: engineering-led teams that want to own their metadata model and avoid licence lock-in. Watch out for: the bill you do not see. Self-hosting means running a multi-service production platform, and the governance workflow layer is usually thinner than in commercial suites.
5. Quality-first suites with MDM: Ataccama
If master data is the part of Informatica you cannot live without, the field narrows dramatically. Ataccama ONE combines data quality, catalog capabilities and a separate MDM product, which is why it is one of the few names that comes up when teams need a fuller like-for-like swap. Expect a longer evaluation and a deeper implementation than a pure catalog.
Side-by-side comparison
The table below summarises the families. It is deliberately qualitative: published pricing in this category is rare, and feature lists change every quarter.
| Option | Family | Strongest at | Main watch-out | Licence model |
|---|---|---|---|---|
| Informatica from Salesforce (IDMC) | Full suite | Breadth: integration, quality, MDM, catalog in one | Roadmap tied to the Salesforce ecosystem | Commercial, consumption-based |
| Collibra | Governance workflow | Stewardship, controls, audit evidence | Needs an existing governance function | Commercial, quoted |
| Atlan | Active metadata | Cross-platform context, modern stack, Fabric partnership | Quality engine scoped to specific warehouses | Commercial, quoted |
| Alation | Active metadata | Analyst adoption, AI governance inventory | Quality monitoring sold separately; warm-up period | Commercial, quoted |
| Unity Catalog | Platform-native / open | Lakehouse governance for data and AI assets | Best inside the Databricks ecosystem | Open source plus managed |
| Snowflake Horizon | Platform-native | Policies that follow Iceberg data to other engines | Cross-engine access control still in preview | Part of Snowflake |
| Google Knowledge Catalog | Platform-native | Google Cloud estates, AI agent context | Recently renamed; check feature naming in contracts | Google Cloud pricing |
| Amazon SageMaker Catalog | Platform-native | AWS estates, data and ML assets together | Strongest inside AWS | AWS pricing |
| Microsoft Fabric OneLake catalog | Platform-native | Fabric estates, Purview security and compliance | Purview governance migration needed by Nov 2027 | Part of Fabric |
| DataHub | Open source | API-first metadata graph, connectors, lineage | Monitoring in paid tier; self-hosting effort | Apache 2.0 plus Cloud |
| OpenMetadata | Open source | Catalog, lineage and quality tests in one OSS product | Operating it yourself | Apache 2.0 plus Collate |
| Ataccama ONE | Quality + MDM | Data quality and master data together | Heavier implementation | Commercial, quoted |
Which family should you start with?
| If this sounds like you | Start your shortlist with |
|---|---|
| "Our regulator asks who approved this definition and when." | Collibra, then Alation |
| "Analysts cannot find the right table and do not trust what they find." | Atlan or Alation |
| "Almost everything we care about is in Databricks / Snowflake / Fabric / BigQuery / AWS." | That platform's native catalog, plus a cross-estate layer only if gaps remain |
| "We were on Purview Unified Catalog." | OneLake catalog for Fabric data, plus Atlan or another cross-estate layer for the rest |
| "We have strong engineers and a small licence budget." | OpenMetadata or DataHub |
| "MDM is the reason Informatica is here." | Ataccama, or stay on Informatica MDM and replace only the catalog |
| "We are deep in Salesforce and Agentforce." | Re-evaluate Informatica from Salesforce before leaving |
A 90-day evaluation plan that avoids the usual traps
- Days 1–15: inventory. List every Informatica service you license, who uses it weekly, and what breaks if it disappears. Note your renewal and support dates.
- Days 15–30: shortlist by family, not by logo. Pick one family that fits your situation from the table above, then two or three products within or next to it.
- Days 30–75: pilot on your hardest pipeline, not the demo dataset. Test lineage through your messiest ETL job and your most-used dashboard. If a product uses anomaly detection, ask for run history or plan a pilot long enough to clear its warm-up period.
- Days 75–90: price the whole thing. Include connectors, quality monitoring, AI features, hosting and the people needed to operate it. Ask which AI model providers power assistant features and whether your metadata is used for training.
- Plan your exit on day one. Favour platforms that speak open standards such as the Iceberg REST catalog API and OpenLineage, and confirm you can export your glossary, ownership and lineage in a usable format.
Ten questions to ask every vendor
- Which of our sources does your quality engine cover, as opposed to your catalog?
- Is lineage parsed automatically for our ETL and BI tools, or assembled manually?
- Are access policies enforced in the data platform or only documented in the catalog?
- What is licensed separately: quality, observability, AI assistants, extra connectors?
- How long before anomaly detection produces useful alerts?
- Can we import our Informatica glossary and ownership, and how much is manual?
- Which AI models power your assistant, and where is our metadata processed?
- How do you support an AI inventory and evidence for EU AI Act or ISO 42001 work?
- What does a typical customer of our size need in headcount to run this?
- If we leave in three years, what do we take with us, and in what format?
Five mistakes we keep seeing
- Replacing the catalog and forgetting the pipelines. Governance platforms describe data; they do not move it.
- Buying a tool to create a culture. Gartner's 2026 survey points to cultural resistance as the leading reason governance fails. Budget for stewards and training, not just licences.
- Two-week pilots of machine-learning monitors. Many detectors need weeks of history before they are useful.
- Assuming the "alternative" is stable. Purview's retirement shows even a hyperscaler product can change direction in a single announcement.
- Ignoring the stay option. For Salesforce-heavy estates, the post-acquisition Informatica roadmap may be the cheapest path.
Frequently asked questions
Is Informatica going away after the Salesforce acquisition?
No. Salesforce completed the acquisition in November 2025 and has continued to ship major Informatica releases, including the Spring 2026 IDMC release. What changes is strategic direction, which is now closely tied to Salesforce's Data 360 and Agentforce products.
What is the best free alternative to Informatica for data governance?
OpenMetadata and DataHub are the two most mature open-source metadata platforms, both under Apache 2.0. OpenMetadata includes data quality tests in the open-source edition; DataHub keeps observability monitoring for its paid cloud edition. "Free" refers to the licence, not the cost of running it.
Can a data catalog replace Informatica MDM?
Generally not. Master data management is a different discipline from cataloguing. If MDM is critical, evaluate an MDM-capable suite such as Ataccama or keep your existing MDM and replace only the governance layer.
What should Microsoft Purview data governance customers do?
Microsoft's guidance is to govern Fabric data in the OneLake catalog and keep Purview for security, audit and compliance. Existing Purview governance capabilities remain in maintenance mode through 12 November 2027, which gives you roughly a year to plan.
How does the EU AI Act affect data governance tooling?
Article 10 sets data and data governance requirements for high-risk AI systems. Following the Digital Omnibus on AI, those high-risk rules apply from 2 December 2027 for Annex III systems. Lineage, data quality evidence and documentation become inputs to compliance, so check how each platform exports that evidence.
The bottom line
The Informatica question is really three questions wearing one coat: where your pipelines will run, where governance should live, and who in your organisation will own it. Answer them in that order. Teams that start with a clear map of what they actually use, pick a family before a logo, and pilot on real data usually end up with a stack that is smaller, cheaper and better adopted than the one they left. Teams that start with a feature matrix usually end up buying another one in three years.
Disclosure: The Curator is an independent publisher. We have no commercial relationship with any vendor mentioned, and this article contains no sponsored placements. Facts were checked against the sources below on 20 November 2026; product capabilities change often, so verify anything decisive with the vendor.
Sources
- Salesforce — Salesforce Completes Acquisition of Informatica (18 Nov 2025)
- Informatica — Informatica from Salesforce announcements at Informatica World (20 May 2026)
- Informatica — IDMC Spring 2026 release
- Informatica — PowerCenter to Cloud Modernization
- Informatica Knowledge Base — PowerCenter end-of-support and CDI for PowerCenter
- Tropos — Informatica PowerCenter End of Support: Your Options and Timeline
- Archon — Informatica PowerCenter EOL: Migration and Legacy Archiving (Nov 2026)
- Gartner — Data governance topic overview
- Gartner — Press release on data governance culture (21 Sep 2026)
- European Commission AI Act Service Desk — Implementation timeline
- EUR-Lex — Regulation (EU) 2026/1744 (Digital Omnibus on AI)
- Microsoft Learn — Data governance with Microsoft Purview (retirement notice)
- Microsoft Tech Community — From Purview Data Governance to the Fabric OneLake Catalog (13 Nov 2026)
- Collibra — Collibra Control Tower is generally available (Nov 2026)
- Business Wire — Atlan and Microsoft collaborate on OneLake governance (17 Nov 2026)
- Atlan — Purview Data Governance Retirement: What to Know
- Alation — New AI Governance offering (11 May 2026)
- Alation — AIOS expansion with six products (17 Sep 2026)
- Decube — 9 Informatica Alternatives for Modern Data Governance (vendor comparison citing product documentation)
- Databricks — Open Sourcing Unity Catalog (Jun 2024)
- Snowflake — Unified Governance for Iceberg Tables with Horizon Catalog (Oct 2026)
- Snowflake — Apache Polaris and the Iceberg REST Catalog (Apr 2026)
- Google Cloud — Knowledge Catalog overview (formerly Dataplex Universal Catalog)
- AWS — Amazon SageMaker Catalog FAQs
- AWS — SageMaker Unified Studio metadata sync with third-party catalogs (Mar 2026)
- CodeWithSQL — DataHub vs OpenMetadata (verified Nov 2026)
- DataCatalog.net — DataHub vs OpenMetadata: An Open-Source Comparison
From our own rounds
Measured on The Curator, from real sessions people played on this site — not a third-party dataset.
- Rounds played here
- 282
- Questions per round
- 1.7
Most-played topics right now: Product Validation (1).
Play a round and add to these numbersSubscribe to The Curator updates
New articles, site messages and notifications — straight to your inbox. Unsubscribe anytime and your details are deleted.
Rate this article
Discussion
Comments are moderated. Read our editorial policy.