Data catalog tools have been on the enterprise data team agenda for many years. The combination of cloud data platform adoption, regulatory pressure for data lineage documentation, and AI-driven data preparation requirements has continued to push data catalog adoption forward through 2024 and into 2026. A working read of where the practice has actually landed.

The framing question.

A data catalog is a tool that documents what data assets exist in the organisation, where they are, who owns them, how they relate to each other, and what they mean in business terms. The catalog supports data discovery, data lineage understanding, data quality measurement, and data governance workflows.

The honest framing is that the catalog tool by itself is not the goal. The goal is the operational outcome that the catalog supports — analysts finding the right dataset for their work, data engineers understanding the downstream impact of a schema change, governance teams documenting the data assets under their accountability, AI engineers finding training data sources with provenance information. The tool is the means, the operating practice is the end.

Where data catalog adoption sits in May 2026:

Adoption of data catalog tools in mid-market and enterprise Australian organisations has continued to grow through 2024-25. Most large Australian organisations now have a data catalog tool deployed at some level of operational maturity. The maturity ranges widely — from organisations using the tool seriously as part of the data governance workflow to organisations where the tool was deployed but is not actively used.

The mature deployments share several characteristics:

Active ingestion. The catalog is being kept current through automated ingestion from the source systems — databases, data warehouses, data lakes, BI tools, ETL platforms. The metadata is current and the catalog represents what is actually in the data estate rather than what was there a year ago.

Active stewardship. Data stewards are working in the catalog as part of their role. They are confirming ownership, updating business glossary terms, approving access requests, and documenting data quality observations.

Integration with data workflows. The catalog is integrated with the broader data workflows — the data engineering pipeline tools, the BI tool catalogues, the access management workflows, and the data quality tools. The catalog is not a standalone destination but is connected to the systems that the data team works in.

User adoption. The analyst and data scientist user base is actively using the catalog for data discovery. The integration is making it easier to find appropriate datasets than the previous tribal-knowledge approach.

The less mature deployments share different characteristics:

The catalog was deployed as a project rather than as a continuing operational capability. The initial ingestion was done, the initial metadata was captured, but the ongoing maintenance and stewardship work has not happened.

The catalog is not integrated with the systems the data team works in. The catalog is a separate destination that users have to remember to visit, rather than something integrated into the daily workflow.

User adoption is low. Analysts continue to ask each other for help finding data because the catalog has not become the reliable resource it was supposed to be.

The tool landscape.

The data catalog tool landscape in 2026 is more consolidated than it was three years ago but still has meaningful variety. The major categories include:

The cloud-native catalogs from the hyperscalers (Microsoft Purview, AWS Glue Data Catalog, Google Cloud Dataplex). These have grown significantly in capability through 2024-25 and have been the natural choice for organisations standardising on a single cloud platform.

The independent enterprise catalogs (Alation, Collibra, Informatica, Atlan) that have grown across the multi-cloud and hybrid environments. The independent vendors have continued to add capabilities specifically for the AI and machine learning workflow patterns.

The open-source catalogs (Apache Atlas, DataHub, OpenMetadata) that have grown adoption at organisations with engineering teams comfortable with operating open-source data infrastructure. The open-source options have matured significantly through 2024-25.

The data lakehouse-integrated catalogs (Unity Catalog from Databricks, Snowflake’s native cataloging) that have grown alongside the broader adoption of the lakehouse pattern.

The choice between these has been shaped by the existing data platform commitments. Organisations standardising on Microsoft Fabric have generally been using Purview. Organisations on Databricks Unity Catalog have used the native Databricks cataloging. Organisations across multi-cloud environments have more often used the independent vendor tools.

What the operational practice has matured into:

Business glossary management. The work of defining business terms, mapping them to data assets, and maintaining the glossary over time has matured. The organisations that have invested in this work have made the catalog meaningfully more useful for non-technical users.

Data ownership clarity. The work of assigning data ownership for each significant data asset has been one of the operational priorities through 2024-25. The catalogs support ownership documentation but the work of getting senior business stakeholders to accept accountability for specific data assets is organisational work that the catalog cannot do alone.

Data lineage documentation. The lineage capabilities of catalog tools have continued to improve through 2024-25, with most major tools now supporting column-level lineage across heterogeneous source systems. The operational use of lineage information has expanded from compliance documentation to active impact analysis for change management.

Data quality integration. The integration between data quality measurement and the data catalog has continued to develop. The pattern of attaching data quality scores to catalog entries is now common, with several tools supporting active data quality assertions and the routing of quality issues to data stewards.

Access management integration. The integration between catalog tools and the access management workflow has continued to mature. The data access request workflows that route through the catalog have become more common.

What is still difficult:

Unstructured data. The cataloging of unstructured data assets (documents, images, audio, video) remains less mature than the cataloging of structured data assets. The growth of AI applications working on unstructured data is creating pressure on the catalog tools to extend their coverage.

Streaming data. The cataloging of streaming data sources and the lineage tracking across streaming pipelines has been an active development area but the practice is still maturing.

Cross-tool consistency. Organisations using multiple catalog tools (for example, the cloud-native catalog for some workloads and the independent enterprise catalog for others) have had ongoing challenges with consistency of the metadata across the tools. The interoperability work continues to be one of the operational frictions.

For data teams in Australian organisations in mid-2026, the working read is that the catalog tooling is mature enough to deliver meaningful value, but the implementation requires sustained operational investment rather than a tool deployment alone. The organisations that have committed to the operational practice are getting value. The organisations that have deployed the tool without committing to the operational practice are typically not getting the return they expected.

The next 12 months will likely bring continued AI-related extensions to the catalog tools, continued integration with the broader data platform stacks, and continued operational maturation at the organisations that have committed to the practice. The catalog category is well past the early-adopter phase and into the operational discipline phase.