You did it!
In this tutorial, you created a deployment marker, and then found an issue monitoring services after a deploy. You successfully went from issue discovery to having a valid, data-backed hypothesis using the Service Health for Deployments view. You were able to see the latency, see correlated attributes with the latency (for later verification), find the downstream operation causing the problem, and then found the actual issue looking at the details of a trace.
You are now safe to deploy (even on a Friday)!