Alright, suppose I'm passionate and an expert, and I want to write a blog going into a deep dive about logarithm approximations:

How do I frame it? Is this for a math audience or a programming audience? If for programmers, are we talking about tricks to make it faster on modern x86_64 platforms or when you need to do voodoo on a microcontroller? If for modern x86_64 platforms, is this a short, punchy article empowering people with a new trick? Is it a long, dry article empowering other experts with nitty, gritty details? Do I frame it in terms of some other problem (like randomized, weighted sort for example)? If so, which problem?

All of those are just questions about "what" I intend to write, after I've ostensibly already chosen what to write about using my expert knowledge and opinions. "What" questions aren't the only unknowns which are improved in the face of knowing your audience though. For the content to land successfully you need to know something about who's reading it. Should you use Python/F#/C/Zig/curl-braces-pseudocode/python-pseudocode/JS/... to deliver the examples? Is it worth incorporating the resulting assembly into the article? If targetting programmers, do they have enough of a math background that you can or should include a derivation or two? Is this the sort of audience who would appreciate seeing all the tricks and solutions which _don't_ work?

And so on. Analytics are a powerful tool for ensuring your article will actually be useful to somebody, even predicated on your assumption that the best content is written by passionate experts (a statement I largely agree with) -- and the hidden, implicit assumption that somebody informing their writing decisions with analytics is _not_ acting as a passionate expert for the resulting work (a statement I wholeheartedly disagree with).

Diving slightly into that point of disagreement, my niche isn't logarithm approximations; it's using a wealth of math, old-school ML, and low-level CPU knowledge to push computers well past their ordinary limits. I don't have near enough time to write. When I do find time, what should I write about? I'm a passionate expert about a lot of things if our yardstick is the length of a blog post, and I don't have time to write about all of them. If I were to write that logarithm article in today's world I might instead write its dual, fast approximate softmax, signalling some sort of expertise to the "AI" people. Or not. Unless I know who's reading my stuff I only have the vaguest of inklings about the specific topics people might find interesting.

In my mind at least, that choice isn't pandering in the same way as the least-common-denominator drivel being shovel-fed on LinkedIn. I still have my own unique ideas and unique takes on those ideas. You can imagine a Venn diagram of the things I'd like to write about and the things the hive mind wants to consume. I'm definitely biased in that I'll pander to the intersection of those two things, but I think your complaint is about people who exit their own bubble completely.