Webinar: The Four Points In Your Medallion Architecture Where Data Testing Really Matters

On-demand webinar: the four points in a medallion architecture where data testing actually changes outcomes, with live TestGen demos against real bronze layer data.

Written by Chris Bergh on March 31, 2026

DataOpsData QualityOn-Demand WebinarOpen Source
Webinar: The Four Points In Your Medallion Architecture Where Data Testing Really Matters

Key points

  • The four points are the source data landing in bronze, the ingestion step itself, the transformation run through bronze, silver, and gold, and the deployment of code changes. Chris frames them as persuade, poll, protect production, and protect the push.
  • The same test does four different jobs depending on where it runs. A null check is a quality metric at the source, an anomaly signal at ingest, a tripwire that stops a fact table reaching customers, and a regression signal when your own code change caused it.
  • Each point has a blind spot, and Chris names them: source quality testing tells you nothing about silver or gold, ingest monitoring notifies rather than stops, production tripwires need tool integration, and deployment testing needs test data and a CI/CD process.
  • A dashboard that doesn't trace to a specific test result is noise. Every test result is a workflow ticket in disguise, and a score without a fix-this path leaves the engineer nothing to do.
  • Full coverage is a big number: 100 tables at 10 columns across three layers, plus the tools and jobs, comes to roughly 5,000 tests. TestGen generates about 80% of them automatically, which is the difference between minutes and an estimated 156 days.
  • The place to start is bronze. A read-only connection plus arrival frequency, volume, schema, and a couple of metrics takes about an hour on as many tables as you like, at no cost.

Most data teams sprinkle tests across their medallion architecture and hope the right ones land. A few row-count checks on gold, a freshness check on bronze, maybe a null check on silver. Then a silent schema change in the bronze layer takes down the gold dashboard on a Tuesday morning and everyone wonders where the test was.

There are four points where data testing actually changes outcomes, and the rest is decoration. The source landing into bronze is where data quality problems are cheapest to find. The bronze ingestion step is where observability earns its keep. The bronze-to-gold transformation is where you protect production from yourself. Pre-deployment is where you catch the regression that would have woken you up at 3am. Skip any of these and your architecture is one bad commit away from a long incident.

In this on-demand webinar, we walk through all four points on real data, with live demonstrations using DataKitchen TestGen. You’ll get a clear map of where the tests go, why each location pays off, and how to connect, profile, and generate tests against your own bronze layer in under an hour.

If you’re running a medallion architecture and you’re not sure whether you’re testing in the right places, this session is for you.

Watch the webinar

Fill out the form and the recording, slides, and transcript open on the next page.

Install Open Source TestGen Free, no vendor lock-in Request a Demo See TestGen Enterprise in action
Chris Bergh

Chris Bergh

CEO and Head Chef at DataKitchen. He is a leader of the DataOps movement and is the co-author of the DataOps Cookbook and the DataOps Manifesto.

LinkedIn →

Don't want to give us your email address? Go directly to the webinar recording here.