Adverity Atlas

Adverity Atlas is a marketing knowledge layer. It connects to your existing data warehouse and gives any AI a governed understanding of what your marketing data means.

Ask questions about your data in plain language, and Atlas writes the SQL, runs it, and returns a visual answer. Your data stays in your warehouse: Atlas connects with read-only access and does not replicate your tables.

To set up Atlas, follow Getting started. For a tour of what Atlas does, see Introduction.

How Atlas works

Atlas is built on three parts:

• Glossary: a centrally governed map of your fields, metrics and calculations.

• Context: the business knowledge that explains why your data changed, captured once and reused in every answer.

• Agents and tools: investigations and answers, through chat, the LLM of your choice, or agents and apps you build.

Your warehouse and your LLM

Atlas connects to Google BigQuery, Snowflake, Databricks, Amazon Redshift, Amazon Athena, PostgreSQL and Trino/Starburst, and you can add CSV files and spreadsheets alongside them (see Data sources). It connects read-only, runs SQL in your warehouse and sends only query results to the AI, never raw data. You can use your own LLM provider: OpenAI, Anthropic, Azure OpenAI or Google Gemini.

Atlas and Connect

Atlas and Connect are separate products, and neither requires the other. Atlas works on any warehouse, whatever loaded the data. Connect collects and harmonizes marketing data and loads it into your warehouse, so Connect customers already have data that Atlas can work on.

To get started with Connect, see Guides.

Was this page helpful? Yes No