The task
Scenario. Next week you are renaming customer_id to cust_id in dim_customers. Somewhere downstream, pipelines and dashboards read that column by name, and unless you warn them, they break quietly at 2am on release night. You want to find everyone affected and give them a week's notice.
The ask
- Every downstream table and dashboard that uses the column, and their owners
- A warning announcement on each
- The change and the timeline spelled out
Hero prompt
The customer_id column in dim_customers is being renamed to cust_id next week. Find all downstream tables and dashboards that use it and add a warning announcement explaining the change and timeline.
Step by step
-
Find the changing asset and confirm what depends on it.
The <column> in <table> is being renamed to <new name> next week — pull <table> and confirm it has downstream lineage.
-
Trace the downstream consumers.
Trace the downstream tables and dashboards that use <table> — direct consumers first, then expand.
-
Resolve the owner of each impacted asset.
For each downstream asset, show its owner (resolve the owning team where it's a group), so every warning reaches the right people.
-
Post a warning announcement on each impacted asset.
Add a warning to each explaining the change from <column> to <new name> and the timeline.
What you'll give it
Required context
- The asset and the change (customer_id to cust_id in dim_customers)
- The timeline (next week, with a date)
- Lineage populated, and permission to post announcements
Optional context
- The migration guidance to include (what consumers should change and by when)
- A cutoff, if you only want to warn assets touched in the last 90 days
What Atlan creates
Outputs
- The full downstream consumer list for the changing column
- The resolved owner of each impacted asset (the owning team where ownership is a group)
- A warning announcement on each impacted table and dashboard
- The change and the date, stated where each owner will see it
Summary. The break is prevented instead of documented after the fact. Every owner in the path gets the same notice with the same deadline, on the asset they actually open.
Changed: a warning announcement on each impacted asset. Left untouched: the column, which you rename on your own schedule; this recipe only warns.
Follow-up prompts
Keep going once the first answer lands.
Give them the fix
Add a line to each announcement telling consumers to update to cust_id before the release date.
Size the blast radius
Where usage data exists, rank the impacted assets by how heavily they're queried, so I warn the busiest ones first.
Confirm before the change
A few days before the change, re-trace the downstream consumers and show which still have no warning announcement, so I can chase them.
Tips & troubleshooting
Send it a week ahead. The point is to prevent the 2am break, not document it after.
- Match the tool to the work. Run it where you can review the consumer list before it posts; a missed consumer is the whole risk, so check the list is complete.
- Automate or chain it. Chain it to your PR flow: when a migration is merged, fire this recipe so the warnings post automatically with the release date.
- Ask what it changed. Add "list every asset you posted a warning on." That list doubles as your notified-parties record.