One thing a microservice architecture does really well is enforce bounded contexts. Oh you want to access that data? Well you need to go through the public API because it exists in a separate process. In a monolith it's all too easy to just 'call this piece of code and grab what I need' (no one will ever know). Project isolation can help but it's not a silver bullet.

The author makes good points though, there are many places doing microservices because it's the hip thing to do and a monolith would easily suffice. But if you have independent software teams in your org that should be able to deploy code independently, then microservices makes a lot of sense.

As in all things engineering - it depends :)