... then proceeds to vague MBA word without a single number to support them.

> We debated making smaller reductions throughout 2024 and 2025

What was not debated was keeping people they hired.

> we still have too many people dedicated to supporting work and even doing work around the work rather than contributing to opportunities with real impact

We're left to guess what this means. What's work around the work? If that work is now unsavory, why can't they work on work rather than around it? Is this describing reducing the management layer? (support) or customer/partner support? Will they be replaced by automation? I get you don't want to go into specifics of who's let go, but then don't pretend you're providing a clear analysis, and don't give a washed out business lingo salad instead.

TBH I don't see what changed on Spotify for a customer perspective in the past few years. I still see bugs I reported years ago, the UI is largely the same. Not that I'm complaining, I just care about the music. But that leads me to think either the dev team is producing stuff that's on the fringe and optional, being quite inefficient, or mainly working on maintenance, and the bulk of the opex is going elsewhere.

> TBH I don't see what changed on Spotify for a customer perspective in the past few years. I still see bugs I reported years ago, the UI is largely the same.

He didn't get into it because you answered it yourself. And anyone who has used it for a while has most likely the same initial thought.

Which is particularly surprising since they have multiple teams of highly compensated product people who are supposed to be guiding the overall product direction and preventing just that kind of incoherence.

Do they?

I have ‘melancholic’ and ‘I want to hype myself up’ playlists. Both from the tone of the music, voice, and lyrics (even the titles, I’d say) it is extremely obvious what the overarching vibe of the playlist is. Yet the radio-mode of those playlist or the recommendations below the playlist are fairly obviously just “other people who played the songs in this playlist also liked these songs”.

As far as I can tell there are no deep analysis smarts at work, or even something like EveryNoiseAtOnce.

I don't understand this argument at all, which suggests that reviewing a monthly report is an O(n) task. Spotify, if they chose this route, trivially solves this with a simple open source app that analyzes your monthly report making it, basically, an O(1) task.

The complexity is not the player, it's the whole business... Winamp didn't have recommendations, tack on another 100-200 people working on data to surface those. There are close to a 100 markets being offered content with licencing deals, each with different regulations.

There are internal platforms, infrastructure, development, each platform requires team(s) to maintain, and develop them.

What's the size of the largest tech company (with a single product) that you've worked at? Just so I know how to translate how larger organisations work to a worldview that you have experience with.

I would presume it's many more products than just Spotify app.

Still sounds simpler than what Spotify does.

Just thinking about dealing with payment processing, regulatory environment of music licensing, customer support, and record labels around all countries on which they operate makes my head hurt a little.

But then again, your bad take is the reason I keep coming back to HN. I always find it amusing how people here always seem to have all the answers for things they never worked with. Most of the times I think people being butthurt for whatever reason.

That's a reductionist way of looking at it, but an alternative way would have been what Daniel said on that letter; They were "more productive and less efficient and they needed to be both".

From my ivory tower I do see some things that could have been done differently, such as adding only one new offering (maybe just podcasting), or rely more in contractors than employees and set the temporary expectation of the roles up front.

Me, being a naive speculator: Maybe it has something to do with the time. The original contracts between spotify + labels were probably written 15 years ago. Over time they might have changed numbers like how big spotify's cut is, but never revised the rest of the blueprint contracts.

So, my bet is laziness / not caring enough.

Okay, sure.

> [...the rest of the post...]

I don't see how this connects.

> your opinions and the opinions of the employees don't matter

But your opinion does? It must matter to you, since you spent a few paragraphs on it.

Obligatory “If I’d asked customers what they want, they would have told me a faster horse.”

People may not realize the value of this new feature because they have not even imagined it.

I care if they can produce better playlists that way. As a Spotify user for many years I can honestly say that after just a few hours with Beats I find the playlists are a lot better and a lot easier to find. They are recommend loads to me and most are ones I like.

My favourite thing is that if I visit an artist page and I haven't listened to them before it has an 'Intro to [artist name]" playlist. If I do know the artist it also has a 'Deep cuts' playlist. I've also seen '[Artist name]: influences' playlists which are an interesting idea. NB: I doubt they have these playlists for every artist but they've been there on most I've searched so far.

I don't get any of that on Spotify. Most of the playlists I get recommended on Spotify are "Hangover songs", "Workout mix", "Christmas tunes" etc. etc. They are nice but a lot less useful.

Can I ask you how many songs are in your Liked playlist? It's a lucky week if I find 3 songs that I like in my Discover Weekly. The rest are forgettable at best, often terrible. I think I am not using Spotify the way I'm supposed to, because i find it really hard to discover new music I like.

Philosophically, an answer is probably is above my pay grade. That said, time has always been a factor to how we understand and label music. Rock today is very different from Elvis, but that doesn't mean we need to throw out genres as a classification system.

Practically speaking, as paying a Spotify user, I really dislike their "browse new releases" tool. Browsing by genre, even some basic ones, would be a HUGE improvement. And, I'd love to follow a micro-genre or two.

Well, if the goal is to pick trending music before it trends, I imagine it's largely a matter of letting time pass to see what % of music popped with the desired demographic segment.

Does Spotify even know who was in the band? Is that metadata anyone attached to the tracks? Sure: if it wasn't they could have asked for it, but if they already have a ton of music without that metadata is the feature ever going to be useful as an afterthought? Does any other music platform have this feature? This example is so weird it is making it really difficult for me to concentrate on whatever point they are trying to make.

This was orignaly Pandora's big selling point.

Every track sent to them was analyzed by someone with a degree in music to help build a recommendation engine. It was a combination of algorithms and people.

I am not sure how well that scaled or if they still do that, for the longest time Pandora had a tiny selection of music and I presume the reason was their intake process.

You say that, but I attended a conference with Spotify- they are specifically working on that problem now.