Maybe more "traction" is not that important, in the end?

BEAM has been around for decades and has proven its worth. Elixir has been around for quite some time too, is super stable and reliable, has a pretty decent user base, real-world applications, ecosystem, etc. In many cases, it's arguably the best tool for the job. And it's not like some critical piece of puzzle is missing.

So it may stay kinda niche forever because people who fully understand the problem and fully realize how to solve it will also always be niche.