Real-Time Upstream Change Detection in Snowflake: Alerts That Actually Matter

Last time, we explored how to simplify the process of finding upstream logic changes for any reporting period you care about — usually after business users come to us with complaints.

But ideally, business users should never encounter broken reports at all (We all prefer to see them only when absolutely necessary 🙂).

So the real question is:

Can we detect upstream changes right when they happen?

With Dwh.dev from the Snowflake Marketplace, the answer is yes.

Pattern-Based Monitoring for Any Object Group

We've added functionality that lets you define a LIKE-pattern and automatically detect:

• changes to the objects matching that pattern
• changes in their upstream (sources)
• changes in their downstream (consumers)

This enables both sides of reliability:

• Upstream changes give you early warning that something may break.
• Downstream changes act as contract monitoring — who is using your data, and how that usage evolves.

Powered by Native Snowflake ALERTS

For notifications, Dwh.dev uses Snowflake's native ALERT feature.

This means:

• all checks run inside your Snowflake account
• no external services
• no data leaving your environment
• fully operational even on Snowflake Standard Edition

Every 5 minutes, Dwh.dev evaluates all relevant objects using QUERY_HISTORY, ensuring you always see changes as soon as they occur — not hours or days later.

How it works