Skip to content
SPECIFICTION
Get listed

DataHub

datahub.com

Free plan Open source WebWindowsmacOSLinuxAPISelf-hosted
Visit datahub.com
Spec score 8.8 #2 of 25 in Active Metadata Management Software

What it is

DataHub brings technical metadata, business knowledge, and documentation together for enterprise data and AI agents. It is available as self-hosted Core or as managed DataHub Cloud. Cloud provides natural-language search, an Ask DataHub chat agent, smart ranking, and a hosted MCP server that connects AI tools to its catalog. Its observability tools run automated schema, freshness, volume, and custom quality checks, with AI anomaly detection and incident workflows. Cross-platform, column-level lineage traces data from sources through transformations and AI models to downstream assets. DataHub Cloud has more than 100 pre-built connectors and names Slack, Microsoft Teams, Chrome, and BI tools among its native integrations. Cloud is described as SOC 2 compliant, with role- and attribute-based access controls and an in-VPC execution option for sensitive sources. Cloud availability is described as SLA-backed at 99.5%. DataHub Core is free to deploy, but requires users to handle installation, configuration, upgrades, uptime, and troubleshooting. Cloud pricing depends on data volume, users, and selected capabilities. A Google Cloud offer advertises a 21-day Cloud trial with a dedicated instance and full platform access.

Spec sheet

What it does
DataHub is an AI and data context platform that unifies enterprise data discovery, governance and observability.Read 14 Oct 2026 · datahub.com
Purpose
DataHub describes its platform as a context platform that unifies technical metadata, business knowledge, and documentation for enterprise data and AI agents.Read 30 Sep 2026 · datahub.com
Trial
A Google Cloud offer page advertises a 21-day DataHub Cloud trial with a dedicated instance and full platform access.Read 30 Sep 2026 · datahub.com
API
DataHub documents GraphQL and REST interfaces and provides CLI commands for GraphQL operations, search, entity retrieval and metadata changes.Read 14 Oct 2026 · docs.datahub.com
Integrations
The project advertises more than 150 integrations, including Snowflake, BigQuery, dbt, Databricks, Looker, Tableau, Power BI, Airflow, Kafka, S3 and Iceberg.Read 14 Oct 2026 · docs.datahub.com
Security
DataHub Cloud states 99.5% uptime, fine-grained access controls, AWS PrivateLink, IP restrictions, in-VPC remote execution and SOC 2 compliance.Read 14 Oct 2026 · datahub.com
Compliance
Gartner's product profile states that DataHub Cloud supports SOC II and HIPAA compliance and offers in-VPC execution for sensitive environments.Read 14 Oct 2026 · gartner.com
Support
Open-source users receive community Slack and self-service documentation, while Cloud customers receive dedicated customer success and private support channels.Read 14 Oct 2026 · datahub.com
Founded
LinkedIn and Gartner list DataHub's founding year as 2021.Read 14 Oct 2026 · linkedin.com
AI integrations
The homepage lists MCP-native integrations with Cortex, Genie, Cursor, Claude, LangChain, Agent Development Kit, CrewAI, and custom agents.Read 30 Sep 2026 · datahub.com
Availability
DataHub Cloud is fully managed and described as having SLA-backed 99.5% availability.Read 30 Sep 2026 · datahub.com
Community scale
The project reports more than 16,000 community members, over 750 contributors and adoption by more than 3,000 organizations.Read 14 Oct 2026 · github.com
Company headquarters
LinkedIn lists the headquarters at 3101 Park Boulevard, Palo Alto, California 94306, United States.Read 14 Oct 2026 · linkedin.com
Company history
The company page says the founders built DataHub from metadata work at LinkedIn and Airbnb and that the company is headquartered in Palo Alto, California.Read 30 Sep 2026 · datahub.com
Company origin
DataHub says the commercial company was formed in 2021 as Acryl Data around the LinkedIn-originated open-source project.Read 14 Oct 2026 · datahub.com
Company size
LinkedIn lists DataHub as a privately held software-development company with 51–200 employees.Read 14 Oct 2026 · linkedin.com
Context platform
Its context features ingest metadata into a unified graph, index query history, let experts approve context and expose validated context through MCP, GraphQL, APIs and SDKs.Read 14 Oct 2026 · github.com
Discovery
The platform provides natural-language search, column-level lineage and profiling for schemas, statistics, ownership and usage.Read 14 Oct 2026 · github.com
Documentation depth
The documentation site includes integration guides, API and SDK references, deployment, authentication, authorization, developer and release-history sections.Read 14 Oct 2026 · docs.datahub.com
Extensibility
The project supports custom connectors, transformers, actions and an extensible metadata model.Read 14 Oct 2026 · docs.datahub.com
Funding
Bessemer Venture Partners says it led DataHub's $35 million Series B announced on May 21, 2025, with participation from 8VC, Tru Arrow, SineWave, In-Q-Tel and Zero Prime.Read 14 Oct 2026 · bvp.com
Governance
DataHub includes a versioned business glossary, ownership and stewardship tools, access policies and compliance controls.Read 14 Oct 2026 · github.com
Ingestion
The ingestion framework offers scheduled pull connectors, programmatic push ingestion, metadata transformation and UI-managed pipelines.Read 14 Oct 2026 · docs.datahub.com
Investors
DataHub lists 8VC, LinkedIn, Bessemer Venture Partners, Next Play Ventures and other investors on its company page.Read 14 Oct 2026 · datahub.com
Licence
The open-source DataHub project is released under the Apache License 2.0.Read 14 Oct 2026 · github.com
Limit: data quality and dbt
Several G2 reviewers report missing or limited support for data-quality tests and dbt integrations.Read 14 Oct 2026 · g2.com
Limit: implementation effort
A G2 reviewer says teams must invest time and secure cooperation from data owners before the catalog delivers value.Read 14 Oct 2026 · g2.com
Limit: native integrations
A Gartner reviewer rates DataHub 4.0 and says it captures essential metadata but still has limited native integrations.Read 14 Oct 2026 · gartner.com
Limit: scale and interface
G2 feedback mentions slower or more cumbersome handling of very large datasets and occasional UI clunkiness.Read 14 Oct 2026 · g2.com
Limit: support documentation
A G2 reviewer reports difficulty finding supportive documents and occasional delays in support responses.Read 14 Oct 2026 · g2.com
Lineage
DataHub describes cross-platform, column-level lineage that traces data from source through transformations and AI models to downstream assets.Read 30 Sep 2026 · datahub.com
Notable limits
The comparison page says DataHub Core has no SSO or fine-grained permissions out of the box, while DataHub Cloud pricing depends on data volume, users, and selected capabilities.Read 30 Sep 2026 · datahub.com
Observability
DataHub supports quality assertions, freshness checks, SLA tracking, incident tracking and lineage-based root-cause analysis.Read 14 Oct 2026 · github.com
Praise: API and permissions
Gartner's displayed reviewer insight praises API-based configuration management and granular privileges.Read 14 Oct 2026 · gartner.com
Praise: connectivity
G2 reviewers praise the broad connections to third-party tools and the ability to bring metadata from many systems into one interface.Read 14 Oct 2026 · g2.com
Praise: usability
G2 reviewers repeatedly describe DataHub as easy to use for organizing datasets, cataloging data and tracking lineage.Read 14 Oct 2026 · g2.com
Production deployment
The documentation says quickstart is for local development and recommends Kubernetes with Helm charts for production.Read 14 Oct 2026 · docs.datahub.com
Rating on G2
4.4 out of 5 from 8 reviewsRead 14 Oct 2026 · g2.com
Rating on Gartner Peer Insights
4.5 out of 5 from 20 reviewsRead 14 Oct 2026 · gartner.com
Rating on TrustRadius
10 out of 10 from 2 reviewsRead 14 Oct 2026 · trustradius.com
SDKs
Official documentation covers Python and Java SDKs for emitting metadata from applications and JVM data systems.Read 14 Oct 2026 · docs.datahub.com
Security practices
DataHub says it encrypts customer data at rest and in transit and conducts third-party penetration tests and static security analysis.Read 30 Sep 2026 · datahub.com
Self-hosting
DataHub OSS can be self-hosted, while DataHub Cloud is a managed service with no infrastructure for the customer to run.Read 14 Oct 2026 · github.com
Supported operating systems
The quickstart documentation provides Docker instructions for Windows, macOS and Linux.Read 14 Oct 2026 · docs.datahub.com
System requirements
The quickstart requires Docker and Docker Compose v2, Python 3.10 or newer, and a tested allocation of 2 CPUs, 8 GB RAM, 2 GB swap and 13 GB disk.Read 14 Oct 2026 · docs.datahub.com

Plans and prices

DataHub Core Free open source · self-hosted · core discovery, lineage, observability and governance Read 14 Oct 2026 · datahub.com
DataHub Cloud Priced on request managed SaaS · pricing varies by data volume, users and capabilities · 99.5% SLA Read 14 Oct 2026 · datahub.com

Every DataHub plan and price →

Strong points and limits

Strong points

  • Core is free to deploy.
  • Cloud offers more than 100 pre-built connectors.
  • Lineage traces data at column level across platforms.
  • Cloud includes onboarding and a dedicated customer success team.

Limits

  • Core has no SSO or fine-grained permissions out of the box.
  • Core users handle installation, upgrades, uptime, and troubleshooting.
  • Cloud pricing depends on data volume, users, and capabilities.

Who it suits

DataHub may suit enterprise teams that need a shared catalog, lineage, and data-quality monitoring. Core is for teams prepared to operate a self-hosted deployment; Cloud is for those seeking managed service and onboarding support.

DataHub in depth

Read the full write-up

DataHub is a metadata and context platform for enterprise data and AI teams. It suits organisations that need shared discovery, lineage and business context across data assets. Core avoids licensing fees; Cloud trades custom pricing for managed operations and broader access controls.

Overview

DataHub brings technical metadata, business knowledge and documentation together so teams can find data assets and understand their relationships. Metadata discovery, a business glossary and lineage analysis make it relevant to organisations trying to connect technical catalogs with business meaning and AI workflows.

Core and Cloud address different operating preferences: Core is open source and self-hosted, while Cloud is a managed enterprise service. Core’s no-cost entry point comes with meaningful operational and access-control responsibilities, so it is best for teams equipped to run the platform themselves.

DataHub sits in Metadata Management Software. It supports API, Linux, self-hosted and web use.

Key features

Discovery and AI connections

Cloud combines natural-language search, smart ranking and an Ask DataHub chat agent with a hosted MCP server that connects AI tools to the catalog. The maker says it has more than 100 pre-built connectors and native integrations including Slack, Microsoft Teams, Chrome and BI tools. MCP-native integrations include Cortex, Genie, Cursor, Claude, LangChain, Agent Development Kit, CrewAI and custom agents. That breadth is useful when teams want catalog context to reach collaboration, analytics and AI tools, though these Cloud discovery capabilities do not make Core equivalent to the managed service.

Observability and lineage

Automated checks cover schema, freshness, volume and custom quality rules, alongside AI anomaly detection and incident workflows. These features give data teams ways to detect and manage issues rather than treating the catalog as a static inventory. Cross-platform, column-level lineage follows data from source through transformations and AI models to downstream assets, helping teams understand dependencies when investigating changes or tracing usage.

Security and service

Cloud is described as SOC 2 compliant, with role-based and attribute-based access controls and an in-VPC remote execution option for sensitive sources. DataHub says it encrypts customer data at rest and in transit and conducts third-party penetration tests and static security analysis. Cloud is fully managed with SLA-backed 99.5% availability, onboarding and adoption support, a dedicated customer success team and private Slack support. Core instead provides basic access controls, community Slack and self-service documentation; it has no SSO or fine-grained permissions out of the box. That gap makes Cloud the more appropriate choice where those access controls matter.

Pricing

DataHub Core

free, billed Free to deploy; open source. Core is self-hosted and includes manual installation, configuration, upgrades, uptime and troubleshooting, with basic access controls and community support. It fits teams able to take on that work and accept the lack of built-in SSO and fine-grained permissions. The zero-cost plan’s trade-off is operational ownership, not merely a smaller feature bundle.

DataHub Cloud

Custom pricing, scoped to data volume, users and selected capabilities; contact sales. Cloud is managed enterprise SaaS, with pricing shaped around the use case and data environment. It is the better fit for teams seeking managed operations, Cloud’s discovery features, stronger access controls and included onboarding and support. A Google Cloud offer advertises a 21-day trial with a dedicated instance and full platform access.

Platforms

DataHub supports API, Linux, self-hosted and web platforms. Core is self-hosted; Cloud is managed enterprise SaaS. The choice is therefore not just about where users access the catalog, but whether the team wants to operate the service itself.

Who it's for

DataHub is strongest for enterprise data teams that need metadata discovery, a business glossary and lineage across data systems, and that want catalog context connected to observability and AI workflows. Core suits teams with the capacity to manage installation, upgrades and uptime, and whose access-control needs fit its basic controls. Cloud better serves teams that need managed availability, SSO-free? no: Core has no SSO, so Cloud is preferable when stronger role- and attribute-based controls are needed, as well as onboarding and dedicated support.

Pros and cons

Pro: Discovery, glossary and cross-platform column-level lineage bring asset context and dependencies together, including paths through AI models.

Pro: Cloud combines more than 100 pre-built connectors with observability checks, anomaly detection and incident workflows.

Pro: Free, open-source Core gives self-hosting teams a no-cost way to deploy the platform.

Con: Core requires teams to handle installation, upgrades, uptime and troubleshooting, and lacks SSO and fine-grained permissions out of the box.

Con: Cloud pricing varies with data volume, users and capabilities, so buyers must obtain a use-case-specific price.

Con: Core’s basic access controls make it a poor fit where more granular permissions or SSO are required.

Alternatives

Progress Semaphore is a paid option with a trial and a development plan for demos, development and capability evaluation; consider it when that plan structure better matches evaluation needs.

Ab Initio Data Platform is a paid alternative that requires a free proof of concept before purchase, making it a fit for buyers who want that step before committing.

Aurelius Atlas offers a free, open-source self-hosted plan with optional consulting, so it is worth considering when a no-license-cost self-hosted option is the priority.

MetaKarta is a paid alternative whose Data Lineage Starter plan costs $50 a year and caps usage at five concurrent users and five pre-selected connectors; it may suit smaller, tightly bounded lineage needs.

Dawiso starts at €445 a month for Standard, with five user seats, five contributor licenses, 20 viewer licenses and unlimited connectors. Consider it when those stated seat counts and unlimited connectors align with the team’s needs.

Alation Data Intelligence Platform uses a pool of consumption units for AI capabilities, with the pool size estimated with Alation; it may suit buyers seeking that AI-capability model.

Aristotle Metadata Registry has paid Micro and Secure hosting plans; a reader comparing metadata registry options may want to consider it.

SemanticWorx Affirma offers paid, inquiry-priced annual subscription plans; it is another option for buyers comparing custom-priced software.

Verdict

Choose DataHub if your team needs a connected catalog, lineage and observability for enterprise data and AI, and can match the deployment model to its operational capacity. Core’s free self-hosting is compelling for capable operators, while Cloud is the stronger choice for managed availability, access controls and support. Look elsewhere if Core’s control limits are unacceptable and Cloud’s use-case-based pricing is not a fit.

Verdict

DataHub offers a free self-hosted option and a managed Cloud edition with catalog, observability, and lineage features. Teams should weigh Core’s operational and access-control limits against Cloud’s use-case-based pricing.

Where Specifiction places it

Alternatives to DataHubAll alternatives →