There’s only so much you can write in a single newsletter issue before it becomes the wrong format. Often I would have a lot more to say about the topics I cover, yet I don’t want to turn the emails into something that takes an hour for you to read through. At the same time, I want to ensure those who are supporting my work with a Plus subscription get value for their money.

Time to try something new. Hello, Perspectives Deep Dives!🚀

One hour videos not only make it easier to visualize complex topics. They also allow me to show demos of tools and experiences from Microsoft, the community, and myself. Because in the age of vibe coding, everyone can be a tool maker. Merely putting things out in a GitHub repo doesn’t mean anyone would understand the “what” and “why”, though. We often need real people to explain it to us.

The idea is not to replace written content with video, because at least I much prefer reading about this technology than watching endless tutorials about something that may or may not be relevant, from a person I may not fully trust yet. Since the Deep Dives are meant exclusively for the Plus subscribers, though — I guess that trust barrier is easier to cross.

There’s already a lot of detailed licensing terms insights I share over at licensing.guide for the audience who cares about the commercial side. But when it comes to technical solution builders and architects, this won’t necessarily make it any easier to approach the topic. In fact, it may just end up causing more anxiety when you hear about the latest changes and gotchas in Microsoft product licensing.

It’s time to reframe it as something beyond SKUs and commercial contracts. Ultimately, licensing is architecture with a price tag. You should not separate it from the other design choices you make around app patterns, automation, identity, capacity, monitoring, or AI features.

That is the core idea behind this first Deep Dive. Instead of trying to memorize product names and price lists, the session looks at the workload patterns that actually determine the answer. Who is getting value from the automation? Where does Dataverse capacity start becoming the real issue? When does pay-as-you-go help, and when does it just hide the problem for later? Oh, and there are demos of my “licensing as code” concept, too!