Skip to main content
BRILLIQS

Atlan

A data workspace that brings cataloguing, lineage and collaboration together where the data team already works.

Atlan is a platform for data cataloguing and governance. It records what data exists across connected systems, traces lineage between assets and supports documentation and discussion against them. Its emphasis is on integrating with the tools teams already use, so context reaches people where they work rather than only inside the catalogue.

What Atlan covers

Atlan is a data catalogue and governance platform. It connects to the systems an organisation uses, records what assets they contain, traces how those assets relate and provides a place to document and discuss them.

Two aspects distinguish how it approaches that: lineage, and where the resulting context appears.

Lineage, and the two questions it answers

Lineage records where an asset came from and what depends on it.

That sounds abstract until you have faced either of the questions it answers.

What breaks if I change this? A table needs a column removed. Somewhere downstream, an unknown number of models, reports and dashboards read it. Without lineage, the honest answer is that nobody knows, so either the change does not happen or it happens and something breaks.

Where did this number come from? A figure on a dashboard looks wrong. Establishing what produced it means tracing back through transformations and tables, which without lineage is an afternoon of investigation.

Lineage turns both into a question you can look up.

Where context appears

Data catalogues have a well documented failure: nobody uses them.

The reason is placement. People work in a query tool, a notebook or a pipeline. A catalogue is a separate application they have to remember exists, open, and search. When they are mid task, they do not.

Atlan's emphasis on integrating with the tools teams already use addresses this. Context surfaced where somebody is already working is context they will see. Context in an application they have to visit is context they will not.

That is a difference in adoption rather than in features, and adoption is what determines whether a catalogue is worth having.

Documentation and discussion in place

Alongside automated discovery, people add documentation, ownership and discussion against assets.

Keeping discussion with the asset matters. Questions about data are normally asked in message threads, answered once, and lost. The next person asks again. Attached to the asset, the answer is found by whoever needs it next.

Ownership

Recording who owns an asset answers a question that otherwise consumes time: who do I ask about this.

Without it, questions circulate until they reach somebody who knows. With it, they go directly, and the responsible person is identified when something needs fixing.

Who uses Atlan

Atlan is used by data teams in organisations with data spread across warehouses, pipelines and reporting tools, particularly where the volume of assets has outgrown informal knowledge.

Points to consider

Automated discovery and lineage stay current on their own. Documentation and ownership do not, and a catalogue whose human contributions go stale becomes something people stop trusting.

Lineage coverage depends on the connections configured and what each system exposes. Where a step in a pipeline is opaque, the lineage through it may be incomplete, and knowing where those gaps are matters.

Connecting systems requires access, and what can be recorded is bounded by what the platform can read.

Getting started

The documentation covers connecting sources, how lineage is derived, documenting and assigning ownership to assets, and the available integrations. Connecting one warehouse and following lineage from a dashboard back to its sources demonstrates the capability most people find useful first.

Key features of Atlan

Capabilities described in the official documentation.

Assets discovered from connections

Connected warehouses, tools and pipelines are read to record the assets each contains.

Lineage between assets

Relationships are traced so where a table came from and what depends on it can be followed.

Collaboration on assets

Documentation, ownership and discussion attach to an asset rather than living in separate channels.

Integrations with working tools

Context can surface in the tools teams already use rather than requiring the catalogue to be opened.

Advantages of Atlan

Factual advantages that follow from the features above.

Impact of a change is visible

Lineage shows what depends on a table, so the effect of altering it can be assessed beforehand.

Context reaches people in place

Surfacing information in the tools already in use avoids relying on people visiting a separate system.

Ownership is recorded

Knowing who is responsible for an asset means questions reach the right person rather than circulating.

Discussion stays with the data

Questions and answers attached to an asset remain findable rather than being lost in message threads.

Common use cases for Atlan

Situations the official documentation describes this tool as being used for.

Data platforms

Assessing the effect of a schema change

Lineage shows which reports and models depend on a table before it is altered.

Analytics

Tracing a figure back to its source

Following lineage upstream from a dashboard shows which tables and transformations produced it.

Enterprise operations

Recording who owns what

Ownership assigned to assets means questions and issues reach the responsible person directly.

Technology

Keeping context near the work

Integrations surface documentation in the tools people use rather than in a separate application.

Official website

Everything on this page is based on the official documentation for Atlan. You can read the source here.

Atlan official documentation

Frequently asked questions about Atlan

Answers taken from the official documentation for this tool.

Lineage records where an asset came from and what depends on it. It answers two frequent questions: what will break if this table changes, and where did the number on this dashboard originate.

No. It records metadata about assets in connected systems: what exists, how it relates, who owns it and what has been documented. The data stays where it is.

Because catalogues are commonly ignored. People work in their query tool or their pipeline, not in a governance application. Surfacing context where they already are removes the requirement to visit a separate system.

Adoption and ownership. Automatically discovered assets and lineage stay current on their own. Documentation and ownership do not maintain themselves, and someone has to be accountable for them.