Economics of Spotify is making songs shorter | Hacker News Reader

Economics of Spotify is making songs shorter

qz.com

66 pointslaurex36 comments

One infuriating thing about economic incentives and Spotify:

Very often, when a song comes up, I will want to delete it from my current playlist because it no longer interests me. In many years it has been practically impossible in the UI. The only way to do it is to hunt through thousands of songs, then click on a context menu to delete it, despite this being something that would be incredibly easy to add to the current song interface and accomplish in one second (Adding the same song to a playlist is an option there).

So why is this the case?

I recently realized that preventing me from cleaning up my playlist means MORE SKIPS for Spotify. They can shove more ads in my face by forcing me to listen to more stuff I don't care about over time. It sounds silly but I would bet money that this is the reason considering how technically simple this feature is and how large the value would be.

I'm not sure about the mobile app, but command-f to search for songs inside a playlist works on the desktop app. It's really not hard to delete songs.

Thanks for the tip! I usually guess this command to be available in a text based UI and represent a search, but it's presence in a menu list UI like Spotify and it's function as a filter is a complete surprise.

On mobile, searching inside playlists recently turned into an option in the three-dot menu on the display page. I miss the old interface where you just drag down to expose a search field, focus it and start typing. :(

I still have that on iOS? Never noticed it going away and it’s still there for me. That feature really did take way too long for me to realize although.

I really should have said Android. I guess I just assumed they used a cross-platform framework on mobile, with no justification at all. Oops.

Jesus, you are attributing something like this to malice, where it is simply:

It is a power user feature.... and most likely nobody bothered to prioritize it as it wont be changing any metrics at all.

Spotify (the desktop client), had all kinds of features (drag and drop, etc..), that were nice but were eventually dropped in the sub-sequential redesigns. The realization was that less of 2% of the users were using any of those features, and many of those features had costs (extra code, more maintenance, etc)....

Of course, many employees are power users themselves, and some were not happy, but over and over it seems that dropping those features didn't hurt metrics, actually improved them (usually because the UI got more simplified, less cluttered).

That little feature is missing, (and annoying you so much) simply because Spotify is a large company now, and nobody (usually either a PM, or senior engineer) has thought that including it, because it doesn't have a large impact.

When the company was smaller, engineers could just jump in and implement little things that 'annoyed' them. Once the company is large, is gets harder to do that.

That just how it is in most tech companies when they reach a certain size.

I have to wholeheartedly agree with your parent here.

The whole purpose of playlists is to organise songs, and almost the entire usecase for removing a song from a playlist is to remove the currently playing song. I almost never want to remove a song under any other circumstances.

This shouldn’t be a power feature, but Spotify makes it unnecessarily difficult to do on the fly.

You're making a few big leaps here. First that self-built playlists are an important feature for users, that people are actively managing their playlists, and that people curate while listening.

Those might all be true for the majority of users, or it might be that you're an outlier. The only way to know is to actually sample real world users.

Malice? No. Prioritization? Sure. Monetization never affects prioritization discussion?

I understand not everything that can make it in and I understand not making features for power users, especially ones that clutter up the UI.

But this is a basic use case. Maybe people use Spotify differently than I do, but I would assume they add their favorite songs to a playlist, put it on shuffle, and that's basically it. Having this functionality be effectively append only doesn't make sense to me.

https://i.imgur.com/a4PWICd.png

I don't see adding one extra option, which would logically fit next to "Add to Playlist", to this context menu with 8 options to be crazy.

Another user stepped in to let me know that playlists can be filtered with Ctrl-F. Now that's a feature for power users if I ever heard of one.

While playing a song in a playlist, clicking on the album art next to the song name will bring you to the song in the playlist, if you're using the desktop client. Doesn't work on mobile. Clicking the song name will send you to the song in the album.

It's similar (well opposite) to Github commits view, where clicking a user avatar will send you to his profile, while clicking his nickname will show you all his commits in the current repo.

It's not a very good affordance.

OK, fair enough. I could have sworn I tried this many, many times and never got it to work. Possibly because you have to click on the image and not on the expand button and not on the song name or album name for this to work...

And it's still too many click IMO.

>I recently realized that preventing me from cleaning up my playlist means MORE SKIPS for Spotify. They can shove more ads in my face by forcing me to listen to more stuff I don't care about over time.

I have a paid account and have the same difficulty removing tracks from a playlist via a mobile device. There are no ads on my account, yet the UX is the same, so I'm not sure that that would be their reasoning. Your idea sounds silly because it is, IMO.

I didn't say the UI would be different for paid accounts. That's a bizarre idea. The difference is with a paid account you can skip as many songs you don't like and face no problems. When I do the same, I quickly run into 3+ consecutive ads, without having barely heard any actual music. You don't have to deal with obnoxious ads. Of course you think it's silly.

I never get ads, at all. I stopped using the desktop app and just connect via the browser and u-block origin does the rest. Continuous music all the time. Is it lower bitrate? Maybe, I can't tell.

It was a beautiful song But it ran too long If you're gonna have a hit You gotta make it fit So they cut it down to 3:05 -- Billy Joel, The Entertainer, 1974

To be fair, the article acknowledges songs were already getting shorter in the 90s, after getting longer with LPs and CDs.

    It was a beautiful song
    but it ran too long
    If you're gonna have a hit
    you gotta make it fit
    So they cut it down to 3:05.

Billy Joel in "The Entertainer" referencing the radio cut of "Piano Man."

I think it's more likely that the decrease in song length is due to the fact that "music production" is becoming a highly specialized and advanced field of work- In the old days, artists just created songs with off-the-cuff lengths... nowadays, a team of experts are going to mull over every single measure in a mainstream music track and ask the question: "Is this measure providing enough value to deserve a spot in this track? If not, these engineers will "refactor" the song to remove bits that might bore a listener and cause them to play another track... the fact that spotify now exists is just incidental to this separate, orthogonal trend in music production.

It's crazy how technical and deep the music production space is becoming.

> DAMN. won the Pulitzer Prize for music, going to show that this trend isn’t necessarily lowering the quality of music.

A bit of a fallacy. Awards aren’t judged against music ever recorded, but against peers released in the same year. Quality could very well be going down or going up, but that would be a trend that can’t be measured across a single award season.

Economics of radio made songs shorter too. Listen to Sun Records tracks and there are plenty of 3 minute songs. I know a few recording artists with several million streams and “radio edits” have been a thing for a long time and not a single one mentions Spotify economics as a motivator for particular edits. I am not sure that recording artists are actually doing Spotify math when it comes to their mixes — they’re often just using the same “radio math” they’ve always done. Pop songs have always been radio friendly. The only artists that can get away with a Stairway to Heaven length song are Led Zepplin-level acts, and even then it has to be an extraordinarily massive tune to get a 5 minute+ song on the radio.

Long popular songs often had (and have) “radio edits”. This is just that again, isn’t it?

33 rpm and 78 rpm singles had much firmer time limits, and much wider impact on song length.

Yet my preferred song "service" is BGM on youtube - 10 hours of someone playing jazz. I think we shall see multiple markets developing as song type, duration and other variables are played with.

Honestly I'm not bothered, the kind of artist that would care enough about optimizing profit on Spotify to actually change their music isn't the kind of artist I give a shit about in the first place.

Maybe there's pressure from the publisher. Many artists have terrible contracts

I find it hard to believe Kendrick Lamar falls in that bucket and with that, I can't take this article very seriously.

Yeah I'd be much more inclined to believe it's just a desired stylistic change from him. Or perhaps he knows about shorter songs being more profitable and decided to do it without sacrificing integrity. (Constraints do make creativity more fun/easier sometimes).

The cognitive dissonance is real

Many artists were told to shorten their songs or at least produce radio versions, and that was part of the contract. At least according to the film Bohemian Raphsody, Queen was faced with such a predicament - now imagine if they had been weaker willed (though still the kind of artist you might give a shit about)...

Money (and security required for the lack of it) is unfortunately a big incentive in society's producton of art, and one look at the "culture industry" tells you more than you need to know.

It doesn't have to be the artists optimizing. It could be that everyone makes whatever they want but those who happen to make shorter songs are pushed up and those that make longer songs are pushed out.

Interesting hypothesis but it would certainly be strengthened by some input from musicians targeting spotify listeners.

Who is to say shorter attention spans aren't part of the cause?

Another important missing bit of information - what percentage of the song needs to be listened to for it to count as a full play? Is it 100%? That seems to be what is implied by the article.

This is one reason I get my music from youtube, not spotify. The videos just as long as the musicians want them to be. For instance https://www.youtube.com/watch?v=AzjvL6bRRKg

The tiny data points presented in the article do not justify the strongly worded title

Misleading title. The article doesn't substantiate the claim and just points out a coincidence. It even says so:

> Still, it’s hard to pinpoint exactly how much streaming has contributed to the recent shortening of songs. The length of pop songs had already been falling through the 1990s, before accelerating in recent years. Some music industry observers blame shortening attention spans—but there isn’t much rigorous evidence that our ability to focus has changed (paywall). Others believe that shorter songs may be a result of more consumer choice—songs need to be more compact and catchy to stand out in the crowd.

Interesting that the last part of the url is "is-spotify-making-songs-shorter". Maybe that was the original title and the editor wanted more clickbait?

Nothing new, really. Do people generally believe most mainstream songs being around 3 minutes in length is due to artistic choice?

Earl Sweatshirt’s new LP clocks in at 25 minutes; 15 songs.

He’s always had short albums

Reply on news.ycombinator.com