Why Evals Matter in AI Analytics
A video walkthrough of evaluation in AI analytics, from checking a churn metric to explaining its causes and deciding what to do.
I think people still underestimate the importance of evals in AI analytics. These answers can shape decisions at every level of a company. The stakes are high, and a convincing answer can still be wrong.
I used OpenAI’s Astra to create this video to illustrate how an evaluation workflow could work, from a user asking “What’s our churn rate?” to “Why did this cohort churn?” and “What should we do?”
I’m pretty impressed by how quickly it turned my voice input and a few rounds of feedback into a video like this.
Open the MP4 video if the player does not load.