How Do You Optimize Costs in Azure Beyond Just Using Dashboards?

I’ve been utilizing tools like Copilot and Claude for analysis. By pasting screenshots and asking why costs are high, I’ve successfully identified issues like unexpectedly high Azure backup costs. There isn’t a universal tool out there; it requires experience and meticulous monitoring of your environment.

A big part of saving money comes from understanding what’s actually necessary to log versus what’s just clutter. Teams tend to log excessively, which can cost a pretty penny. Also, many don’t realize that logs come in various tiers—using the basic tier can save you a substantial amount compared to the more expensive analytical tier. Setting a budget with alerts is crucial, though not everyone seems to follow through with sticking to it!

Using Advisor recommendations is a decent starting point, but it’s important not to stop there. Focusing on reserved instances can save a lot—if you have workloads running steadily for a few months, switching from pay-as-you-go can save up to 40%. Also, remember to clean up unused resources! Azure doesn’t do that for you.

It’s all about understanding your workloads and sizing them properly. But does that sizing stay consistent, or do you find yourself reevaluating regularly?

We rely on ProsperOps to manage our Reservations and try to keep a close eye on our environment's costs. If your organization isn't proactive about this, expenses can get out of control. Setting up scaling in K8s clusters, using Databricks compute policies, and tiering storage when possible has been essential for us.