Comparison| AIpedia Editorial Team

AI Data Catalog and Data Governance Platforms Comparison 2026: Atlan, Alation, Collibra and 7 More (Pricing & Free Plans)

We compared 10 AI data catalog and data governance platforms, including Atlan, Alation, Collibra and DataHub, on pricing, free plans, Japanese support, platforms, strengths and weaknesses, with picks by goal and an FAQ (vendor websites checked in November 2026).

Choosing an AI data catalog or data governance platform is harder than it looks. Most vendors do not publish prices, the tools overlap in scope, and each one describes its AI features in its own terms. Japanese-language support is often not stated at all.

This article compares ten platforms: Atlan, Alation, Collibra, DataHub, Collate, Dataedo, Ataccama ONE, Select Star, Informatica Cloud Data Governance and Catalog, and Decube. For each tool we cover pricing, free plans or trials, Japanese support, supported platforms, strengths, and weaknesses. We also give picks by goal and tips for buying without regrets.

We compared the tools using each vendor's official website, checked in November 2026. Details may change, so check the official website for the latest pricing and features before you decide.

What Are AI Data Catalog and Data Governance Platforms? What They Do and What to Know Before Choosing

AI data catalog and data governance platforms help organizations find, describe, and control their data. Many now also pass that context to AI agents. Typical uses, based on what the vendors list, include the following.

  • Data discovery and documentation: a catalog records business definitions, owners, and usage. Some tools, such as Atlan and Collibra, generate asset descriptions with AI.
  • Lineage and impact analysis: tools such as Select Star and Decube track column-level lineage. Dataedo offers object-level and column-level lineage.
  • Governance and policy management: business glossaries, role-based access control, ownership, and automated policy checks, as in Collibra. Informatica Cloud Data Governance and Catalog also lists policy automation.
  • Data quality monitoring: Ataccama ONE, Dataedo, and Decube include quality checks, anomaly detection, or both.
  • Context for AI agents: many tools offer an MCP server or MCP integration so agents can use metadata and lineage. Examples are Atlan, DataHub, and Collate.
  • AI governance: Collibra, Collate, and Decube list features to manage AI assets, agents, or models.

Before choosing, check what a free plan or trial actually covers. Collate lists a Free Tier with 5 users and 500 data assets. Dataedo offers a 14-day free trial with all features. Several other tools list no free trial and direct you to a demo.

Pricing models differ widely. Dataedo and Decube publish prices, per editor or per user. Informatica calculates pricing from IPU consumption, and Ataccama ONE sets tiers by users, data objects, and DQ configurations. Several others list no prices and ask you to request a demo or a quote.

Also check Japanese support, deployment options, and size fit. Informatica has a Japanese-language version of its official website, but none of the ten states Japanese UI support on the official pages checked. Minimums such as Dataedo's 3 editors or Decube's 10 users matter for small teams. Check the official website for the latest pricing and features.

Comparison Table of Top 10 Tools (Pricing, Japanese Support, Platforms)

ToolMain plansJapanese supportPlatforms
AtlanContact salesNot statedWeb, API/SDK, MCP server, SQL interface
AlationCheck the official pricing pageNot statedWeb, API/SDK, MCP, Open Connector Framework
CollibraCheck the official websiteNot statedWeb, API, Slack and Microsoft Teams integration
DataHubDataHub Core: open source / DataHub Cloud: check the official websiteNot statedWeb, API/SDK, MCP, Open source (DataHub Core)
CollateFree Tier (5 users, 500 data assets) / Premium, Enterprise: contact salesNot statedWeb, API, AI SDK, MCP and more
Dataedo14-day free trial / Essentials: $18,000/year / Data Lineage: $24,000/yearNot statedWindows, AWS, Docker (self-hosted), Cloud setup (on request), Web, SQL scripts
Ataccama ONEContact salesNot statedWeb, API, MCP server, Ataccama Cloud
Select StarNo prices listed (part of Snowflake)Not statedWeb, API, MCP server, Slack integration
Informatica Cloud Data Governance and CatalogIPU-based pricing: check the official pricing pageJapanese site availableWeb (Intelligent Data Management Cloud), API, Cloud Connectors
DecubeStarter: $175/user/month / Growth: $225/user/month (billed annually) / Enterprise: customNot statedWeb, API and MCP, Private cloud (Enterprise), Login regions in the US, EU and APAC

AIpedia has not rated these tools, because they do not have their own pages on AIpedia yet. Pricing and plans change; we checked each vendor's official website in November 2026. Check the latest information on the official website before you sign up.

Features, Pricing, and Best For: 10 Tools

1. Atlan

Atlan is a data catalog and governance platform that the official website calls the "context layer for AI." It brings lineage, business terms, governance, and access policies together and delivers them to AI agents. Context Agents generate asset descriptions and link business terms, and a human reviews the context before it is certified.

  • Strengths: Lineage, glossary, governance, and context for AI sit in one platform. The official website lists more than 80 native connectors, including Snowflake, Databricks, BigQuery, dbt, Tableau, and Power BI.
  • Strengths: AI agents can reach it through an MCP server, SQL interface, API, and SDK. It is described as supporting MCP and A2A.
  • Strengths: The official website reports more than 690K descriptions generated across more than 50 customers, with 87% rated equal to or better than human-written ones. These are vendor figures.
  • Things to note: Pricing is not public, and the official pages checked do not mention a Japanese UI or availability in Japan. They also do not name security certifications such as SOC 2 or ISO 27001, so ask Atlan directly.
  • Things to note: There are many new products for AI agents, so confirm with sales what is included as standard.

Pricing: Contact sales (no pricing listed on the official website). Visitors are directed to request a demo.

Best for: Teams that want one platform for lineage, glossary, and governance and plan to connect AI agents through MCP. Japanese support should be confirmed before adoption.

2. Alation

Alation provides what the official website calls the "Alation Intelligence Operating System (AIOS)." It bundles Catalog, Lineage, Quality, Curation, Data Products, Marketplace, Ontologies, Agents, and more. It records business definitions, owners, trust signals, quality scores, lineage, and usage.

  • Strengths: The official website says more than 500 global companies use it, including hundreds of the Fortune 1000. It connects to more than 120 sources without moving data.
  • Strengths: Catalog, lineage, quality, and governance run on the same platform. AI-drafted content is scored by quality checks and refined by stewards.
  • Strengths: Agents can reference the catalog through MCP and API. The Open Connector Framework lets you connect to non-standard sources.
  • Things to note: For pricing, check the official pricing page, which offers a 30-minute consultation. The official pages checked do not mention a Japanese UI or availability in Japan, although Japan Airlines and NTT docomo appear among the customer logos.
  • Things to note: The deployment model (cloud, on-premises, hybrid) is not stated on the product page checked. The many modules make the necessary scope and cost hard to estimate.

Pricing: Check the official pricing page. The site offers a 30-minute consultation.

Best for: Larger organizations with many data sources that want catalog, lineage, and quality in one platform. Confirm deployment options and Japan availability in advance.

3. Collibra

Collibra describes itself on the official website as the "Enterprise AI Control Plane." It manages context and governance across data, models, and AI agents. Its core is Data Governance and the AI Command Center, which detects AI agents, sets what they can read and write, enforces rules at runtime, and stores audit trails.

  • Strengths: According to the official website, 78 of the Fortune 500 use it, and it manages more than 2 billion data assets. Data governance and AI governance come from the same vendor.
  • Strengths: It offers a business glossary, role-based access control, ownership, automated policy checks, and audit trails.
  • Strengths: Collaboration flows are embedded in Slack and Microsoft Teams, and a Data Helpdesk handles data issues from request to resolution.
  • Things to note: Check the official website for pricing and trial options; it offers an interactive tour and a demo request. The official pages checked do not mention a Japanese UI or availability in Japan.
  • Things to note: The messaging leans toward AI governance, so look at the product pages for standalone catalog features. The official pages checked do not name certifications such as SOC 2 or ISO 27001, so ask Collibra directly.

Pricing: Check the official website for pricing. The site offers an interactive tour and a demo request.

Best for: Enterprises that need data governance and AI governance together, with audit trails and role-based access control.

4. DataHub

DataHub Cloud is an enterprise-grade service built on DataHub Core, an open-source metadata platform. The official website calls it "The Context Platform for AI Agents." Modules cover Discovery, Observability, Governance, Lineage, AI Data Management, and Context Management.

  • Strengths: It is built on open-source DataHub Core. The official website cites more than 3 million downloads per month and more than 3,000 organizations.
  • Strengths: Ask DataHub, native MCP integration, and Context Kit connect it to agent tools such as Claude, Cursor, Cortex, Genie, CrewAI, and LangChain.
  • Strengths: Lineage, quality, data contracts, and policies are handled on the same platform. Experts review and approve metric definitions that agents propose.
  • Things to note: Check the official website for DataHub Cloud pricing. The official pages checked do not mention a Japanese UI, availability in Japan, or specific certification names such as SOC 2, so ask the vendor directly.
  • Things to note: There are several product layers (Activation, Layer, Store), so confirm the required configuration with sales.

Pricing: Check the official website for DataHub Cloud pricing. DataHub Core is open source.

Best for: Teams that value an open-source foundation and want to connect AI agent tools through MCP. Japanese support should be confirmed.

5. Collate

Collate, which the official website calls "the Agentic Data Platform," is a managed service based on the open-source OpenMetadata. It adds AI features such as Collate AI, AskCollate, AI Studio, and AI Governance Studio. Core features include Data Discovery, Lineage, Quality, Observability, Governance, and Data Contracts.

  • Strengths: The pricing page offers a Free Tier (free sign-up) with 5 users and 500 data assets. Premium and Enterprise tiers raise the limits.
  • Strengths: It lists more than 130 connectors. External agents can connect through MCP and the AI SDK.
  • Strengths: The official website lists SOC 2, SCIM 2.0, GDPR, and CCPA, and describes a 99.9% SLA for the managed version. Its pages mention both SOC 2 Type II and SOC 2 Type 1, so ask Collate which report is current.
  • Things to note: The Free Tier is multi-tenant SaaS only, refreshes weekly, and is limited to 5 users and 500 assets. The pricing page says inactive clusters are reclaimed after 4 weeks of inactivity.
  • Things to note: Premium and Enterprise prices are not public, and the official pages checked do not mention a Japanese UI or availability in Japan.

Pricing: Free Tier: free sign-up (5 users, 500 data assets, weekly data refresh, multi-tenant SaaS only). Premium: contact sales (25 users, 5,000 data assets, refresh every 8 hours, single-tenant or hybrid SaaS). Enterprise: contact sales (50 users, 10,000 data assets, hourly refresh, private BYOC also available, limits can be expanded). Additional users and data assets cost extra (no prices listed).

Best for: Teams that want to start free on a managed catalog built on open-source OpenMetadata and move to a paid tier later.

6. Dataedo

Dataedo is a lightweight data governance platform that combines a data catalog, data quality, lineage, and governance. The official website positions it as easy to use for both data engineers and business analysts. It has more than 50 native connectors.

  • Strengths: Annual prices are published, from $18,000 to $32,000 per year. Viewers and connectors are unlimited, and the price depends on the number of editors.
  • Strengths: A 14-day free trial lets you try all features with your own data.
  • Strengths: Data quality includes more than 90 built-in rules and custom SQL rules. It can be self-hosted on-premises or in your own AWS or Azure environment, and the official website states it is strong in Microsoft-based environments.
  • Things to note: It requires a minimum of 3 editors and a contract from $18,000/year, which is expensive for small teams.
  • Things to note: The AI features are described only in a limited scope, such as automatic cataloging and PII detection. The official pages checked do not mention a Japanese UI, availability in Japan, or certifications such as SOC 2 or ISO.

Pricing: Essentials: $18,000/year (standard support). Data Lineage: $24,000/year (premium support). Data Quality: $32,000/year (premium support). The minimum is 3 editors, and the price depends on the number of editors. Viewers and connectors are unlimited. A 14-day free trial includes all features and works with your own data. Confirm with Dataedo which modules each plan includes.

Best for: Teams that want published prices and a trial with their own data, and that may prefer self-hosting.

7. Ataccama ONE

Ataccama ONE is described on the official website as "An agentic platform, with data quality at its core." It puts data quality, a catalog, lineage, reference data management, observability, and master data management (MDM) on one platform. The ONE AI Agent supports steward work.

  • Strengths: Data quality is the core, with automated DQ checks, monitoring, anomaly detection, and remediation. The catalog, MDM, and observability sit on the same platform.
  • Strengths: The official website says it was named a Leader five times in Gartner's Magic Quadrant for Augmented Data Quality. The Data Trust Index shows whether a dataset can be used for agent actions.
  • Strengths: Read-only users are unlimited at every tier. All limits are "soft": the pricing page says exceeding one does not interrupt operations or trigger retroactive charges.
  • Things to note: Prices are not public, and a quote is required. No free trial is mentioned, so you start with a demo or evaluation consultation.
  • Things to note: The product scope is broad, so work out which modules you need. The official pages checked do not mention a Japanese UI or availability in Japan.

Pricing: Request pricing (no prices are listed on the official pricing page). Your tier is set by three factors: named users, data objects, and active DQ configurations (checks run at least once in the past 90 days). Read-only users are unlimited at every tier. All limits are "soft": exceeding one does not interrupt operations or trigger retroactive charges, and long-term overages are addressed collaboratively rather than back-charged. Each named user gets an annual AI query allocation, and more can be added as an add-on.

Best for: Organizations where data quality is the main concern and where MDM or observability may be needed on the same platform.

8. Select Star

Select Star is described on the official website as a "Metadata Context Platform for Data & AI." It provides a data catalog, column-level lineage, Ask AI, an MCP server, ER diagrams, and access control. Documentation is generated automatically.

  • Strengths: It provides column-level, cross-platform lineage and automatic documentation. The MCP server can pass metadata and lineage to AI tools.
  • Strengths: The official website lists broad connections to warehouses such as Snowflake, BigQuery, Redshift, and Databricks, and BI tools such as Tableau, Looker, and Power BI.
  • Strengths: It can analyze BI dashboard logic to create semantic models for use with Snowflake Cortex Analyst and others. One listed case says nib identified about 600 unused tables.
  • Things to note: Pricing is not listed. The site shows the notice "Select Star is now a part of Snowflake," so confirm current availability and how the product will be offered before you plan a purchase.
  • Things to note: On SOC 2, the official website says it treats data according to AICPA SOC 2 standards for security, confidentiality, and availability, so ask for the audit report if you need one. The official pages checked do not mention Japanese support or availability in Japan.

Pricing: No prices are listed on the official website. Confirm current availability and pricing with the vendor.

Best for: Teams that rely on column-level lineage and BI integrations, especially those already working with Snowflake.

9. Informatica Cloud Data Governance and Catalog

Informatica Cloud Data Governance and Catalog (CDGC) is a product of the Intelligent Data Management Cloud. It adds lineage and business context to a centralized catalog of data and AI assets. AI-based classification links business terms with technical assets, and lineage across source systems is created automatically.

  • Strengths: The official website has a Japanese version (/ja/), and the data catalog and data governance product pages are published in Japanese.
  • Strengths: Catalog, lineage, policy automation, and marketplace sit in a single cloud. It can be combined with data integration, quality, and MDM products in the same cloud.
  • Strengths: The official website says more than 5,000 companies use Informatica and that it is recognized in six Gartner Magic Quadrant reports.
  • Things to note: Pricing is based on IPU consumption, so check the official pricing page or request a quote. The pricing page does not show a free trial for Catalog or Governance.
  • Things to note: The official pages checked do not state whether the product UI is available in Japanese, so confirm the UI language and local support with Informatica. As part of a larger suite, the scope needs to be confirmed if you use only the catalog.

Pricing: Check the official pricing page. According to the official website, pricing is calculated from consumption of Informatica Processing Units (IPUs), and sample price ranges are provided when you request a quote. No free trial for Catalog or Governance is mentioned.

Best for: Organizations that want Japanese-language product information and may combine the catalog with other Informatica data management products.

10. Decube

Decube is a platform that the official website calls a "trusted data context layer." It combines a catalog, governance, column-level lineage, data quality monitoring, and AI governance. Trusty AI (Beta) converts English questions into SQL and answers based on lineage, owners, and metadata.

  • Strengths: Per-user pricing and the annual minimum are published, so costs are easy to estimate. All plans are billed annually in US dollars.
  • Strengths: Quality monitoring covers schema changes, freshness, and anomalies, with no-code and SQL tests and data contracts. MCP integration is available.
  • Strengths: The official page lists SOC 2, ISO 27001, HIPAA, and GDPR. Login regions include the US, EU, and APAC.
  • Things to note: The minimum is 10 users and from $21,000/year, which is heavy for small teams. No free trial is mentioned; you start with a demo or the Sandbox.
  • Things to note: Trusty AI is in Beta and is included at no additional cost for now; the pricing page says its pricing will be communicated before general availability. The official pages checked do not mention a Japanese UI or availability in Japan.

Pricing: Starter: $175/user/month (billed annually, from $21,000/year, minimum 10 users, up to 3 data sources, 1,000 monitors, 25 Trusty AI queries per user per month). Growth: $225/user/month (billed annually, from $54,000/year, minimum 20 users, up to 10 data sources, 3,000 monitors, 100 Trusty AI queries per user per month). Enterprise: custom pricing (volume pricing available, unlimited data sources and monitors). Users beyond the plan minimum are billed at the same per-user rate. Add-ons: additional monitors $0.59 each, additional data sources $100 per source per month (billed annually), single-tenant SaaS +$1,000/month (billed annually, Growth and Enterprise only).

Best for: Mid-sized teams that want transparent per-user pricing with catalog, lineage, and quality monitoring in one product.

Recommendations by Goal

GoalRecommendedWhy
Start with a free planCollate, DataHubCollate lists a Free Tier (5 users, 500 data assets), and DataHub Core is open source.
Try the product before buyingDataedoIt offers a 14-day free trial that lets you try all features with your own data.
Japanese-language informationInformatica Cloud Data Governance and CatalogThe official website has a Japanese version at /ja/, with data catalog and data governance product pages in Japanese.
Published prices for budgetingDataedo, DecubeDataedo lists $18,000/year to $32,000/year, and Decube lists Starter at $175/user/month, billed annually.
Enterprise scale and AI governanceCollibra, AlationAccording to their official websites, 78 of the Fortune 500 use Collibra, and more than 500 global companies use Alation.
Data quality as the core needAtaccama ONE, DecubeAtaccama ONE is built with data quality at its core, and Decube includes monitoring for schema changes, freshness, and anomalies.
Access from AI agents through MCPAtlan, Select StarAtlan provides an MCP server, and Select Star offers an MCP Server for Data that passes metadata and lineage to AI tools.
Column-level lineageSelect Star, DecubeBoth list column-level lineage as a core feature on the official website.

Most tools here do not publish prices, so request quotes before you shortlist. Japanese UI support is not stated for any tool on the official pages checked, so confirm it with each vendor. Select Star's site says it is now part of Snowflake, so confirm its availability too.

How to Choose Without Regrets

Check what the free plan or trial really covers

Only a few tools list a way to try the product without a sales call. Collate has a Free Tier limited to 5 users and 500 data assets, with weekly refresh. Dataedo offers a 14-day trial of all features with your own data.

Ataccama ONE and Decube do not mention a free trial on their pricing pages. Ataccama offers demos and a guided evaluation, Decube offers a product tour and a sandbox, and Collibra offers an interactive tour and a demo request. Plan for a demo and a structured evaluation instead.

Understand the pricing model before asking for a quote

Models differ. Dataedo charges by editor, Decube by user, and Informatica by IPU consumption. Ataccama ONE sets tiers by named users, data objects, and active DQ configurations.

Atlan and Select Star list no prices. For Alation, Collibra, and DataHub, check the official websites for current pricing. Ask each vendor what is included as standard and what is a paid add-on.

Do not assume Japanese support

Most of these vendors do not mention a Japanese UI or availability in Japan on the official pages checked. Informatica Cloud Data Governance and Catalog has a Japanese-language version of its website, but the product's UI language is not stated.

If your team works in Japanese, ask each vendor in writing about UI language, documentation, and support hours.

Match the minimums to your team size

Dataedo requires at least 3 editors and starts at $18,000/year. Decube requires at least 10 users on Starter and starts at $21,000/year. These can be heavy for small teams.

Collate's Free Tier may fit a small pilot. Enterprise-oriented tools such as Collibra and Alation may suit larger organizations, but ask about scope and cost.

Confirm deployment, security, and product direction

Dataedo can be self-hosted, Collate offers SaaS, hybrid, and BYOC options, and Decube offers private cloud on Enterprise. Alation's product page does not state its deployment model.

Certifications also need checking. Collate and Decube name specific ones on their websites; for the others, ask each vendor for current certifications and audit reports. Select Star's site says it is now part of Snowflake, so ask how it will be offered. Confirm the final decision with a qualified professional where legal or compliance questions arise.

FAQ

Q. Which tools publish their prices?

A. Dataedo and Decube publish prices. Dataedo lists Essentials at $18,000, Data Lineage at $24,000, and Data Quality at $32,000 per year. Decube lists Starter at $175/user/month and Growth at $225/user/month, billed annually. For the other tools, check each official pricing page or ask the vendor for a quote.

Q. Is there a free plan or free trial?

A. Collate lists a Free Tier with 5 users, 500 data assets, and weekly data refresh. DataHub Core is open source; for DataHub Cloud pricing, check the official website. Dataedo offers a 14-day free trial with all features. Check the official website for the latest pricing and features.

Q. Which tools offer Japanese-language information?

A. Informatica Cloud Data Governance and Catalog has a Japanese version of the official website at /ja/. Its data catalog and data governance product pages are published in Japanese. For the other tools, the official pages checked do not mention Japanese support. Confirm UI language and local support with each vendor.

Q. Which tools support MCP for AI agents?

A. According to the official websites, Atlan, Alation, DataHub, Collate, Ataccama ONE, Select Star, and Decube mention MCP. Atlan also lists an SQL interface, API, and SDK. DataHub lists integrations with agent tools such as Claude and Cursor. Collibra's AI Command Center manages AI agents; ask Collibra about MCP support.

Q. What should I confirm before choosing a tool?

A. Start with pricing, since many vendors require a quote or demo. Then confirm Japanese support, deployment options, and security certifications. Collate and Decube name certifications such as SOC 2 on their websites, so ask the other vendors for theirs. Select Star has a notice that it is now part of Snowflake, so confirm how it will be offered.

Summary

Most of the ten platforms combine a catalog, lineage, and governance, and many add MCP access for AI agents. The clearest differences are in pricing and trials. Dataedo and Decube publish prices, Collate lists a Free Tier, and Dataedo offers a 14-day trial. For the other seven, check each official pricing page or ask the vendor for a quote.

Japanese support is another gap. Informatica Cloud Data Governance and Catalog has a Japanese-language website, while the official pages checked for the others do not mention Japanese support. Check minimums, deployment options, and certifications against your team size and requirements. Check the official website for the latest pricing and features.

  • Want to start without a sales call → Collate (Free Tier with 5 users and 500 data assets)
  • Want published annual pricing and a trial with your own data → Dataedo (from $18,000/year, 14-day free trial)
  • Need Japanese-language product information → Informatica Cloud Data Governance and Catalog (Japanese site and product pages)

View all AI data analytics tools (14 items)

📊Browse all tools in AI Data Analytics →

Comparisons for Other Categories

Is your tool missing from this article? Vendors can add it with a paid section written by our editors (JPY 50,000, one-time), or take the first position, labeled “Sponsored” (JPY 80,000). If a fact about a tool here is wrong, we correct it for free.

Written & verified by

A

AIpedia Editorial Team

The AIpedia Editorial Team specializes in researching, comparing, and hands-on testing AI tools. We create accounts and use the tools we cover, verifying pricing, key features, and real-world usability before writing. Articles are reviewed regularly to keep the information up to date.