> I'd expect that to be a non-trivial, not-automatable business critical task that requires lots of bodies.
From my brief but traumatising stints working with companies dealing with music, yes, it is a clusterfuck of nonsense that requires industry knowledge, contacts, etc. and only parts of it are amenable to automation (not the important bits, either.)
Spotify (surprisingly) doesn’t deal with many artists or labels directly, and music is delivered against an industry-code schema that contains all the metadata (ok, most of the metadata) needed to attribute usage and route payments.
That happens via a bunch of intermediary aggregators/distributors and a decent part of this automated/automatable.
As I understand it a lot of the staffing overhead at Spotify is in product and marketing - and R&D efforts for retention/acquisition. It also seems like these were a lot of the folk laid off which then impacts growth/retention.
They don't have to deal with the legal aspects of licencing content, nor need to do marketing, nor sales, nor have specialised teams working for features in specific markets; They don't have to ingest content, match content with potential existing ones to pay royalties properly (if not you get legally fucked), they don't have to do any kind of recommendations, so on and so forth...
I think you have a very simplistic view of how businesses work since your comparison is not even wrong.
What exactly? They didn't lay off 1500 engineers, I would imagine the bulk of Spotify employees handle all sort of licensing deals, talent management, ad sales etc. Telegram is a chat application that does none of this.
Me too. I’m not aware of a single change they’ve made in ten years. Offline still doesn’t really work. I’m sure there’s been tons of scaling work, but why does that require thousands of employees?
User facing changes are usually the tip of the iceberg for most mature companies. Feels very naive to judge a large company that deals with complicated licensing, sales, marketing by some pet peeve feature you are missing in the app.
Not really. The engineering roadmap of large public companies optimize one thing: stock price. Gone are the days of working on features that benefit the user. That’s called enshittification
Ten years ago was 2014, there's been a lot of changes to Spotify in 10 years, not even just in types of content (podcasts, audiobooks, music videos) but the whole recommendation system, daily playlists, discovery playlists, lyrics, even banal things like playlists folders didn't exist 10 years ago.
There's also a lot of scaling into new markets they didn't have presence in 2014 (which requires scaling in billing, legal, accounting, etc.).
I think when you hire them they make themselves useful which is the problem. If you stay small you solve problems using solutions in ways that small teams can manage.
I'd expect that to be a non-trivial, not-automatable business critical task that requires lots of bodies.