How would you improve Atlan's column-level lineage for data engineers doing impact analysis before schema changes?

Introduction

Column-level lineage is one of Atlan's most valued capabilities. It traces how each column flows from source systems through transformation tools like dbt and warehouses like Snowflake and Databricks into BI dashboards and, increasingly, AI applications. For data engineers, the most important use of lineage is impact analysis: before renaming a column, changing a data type or deprecating a table, they need to know what will break downstream and who to warn.

The question asks how to improve Atlan's column-level lineage for data engineers doing impact analysis before schema changes. Today, many engineers still change schemas, wait for something to break, and then fix it. If Atlan can make impact analysis fast and trustworthy, it prevents incidents and earns a place in the engineer's daily workflow.

Step 1

Clarifying Questions (5 mins)

• Looking at the product context, how do engineers use lineage for impact analysis today?

• Hypothetical answer: They open Atlan's lineage graph, search for the table and expand downstream nodes manually.
• Impact: The graph may be powerful but slow for large estates.

• Considering user behaviour, how often do schema changes cause downstream breakages for our customers?

• Hypothetical answer: Customers report several incidents per month from upstream changes breaking dashboards or pipelines.
• Impact: High-value problem.

• Examining external factors, which tools must lineage cover for impact analysis to be trusted?

• Hypothetical answer: dbt, Snowflake, Databricks, Airflow, Tableau, Power BI, Looker, and reverse-ETL tools.
• Impact: Coverage gaps undermine trust; we must show where lineage may be incomplete.

• Thinking about company alignment, is the goal engagement among engineers or reduced incidents for customers?

• Hypothetical answer: Reduced incidents, which drives retention and expansion.
• Impact: We will measure prevented incidents.