[kbolino]

How much of this is actually new to the streaming era? Radio stations may have paid better per play/listener, but they had finite airtime and so fewer songs got played and fewer artists got paid. Physical media may have paid better compared to streaming the same songs once, but the media could be copied and replayed infinitely.

There was a lot of unrealistic hype that software magic could make everything between the consumer and the producer practically free, and that just hasn't happened and probably never will. Engineers need to get paid and infrastructure needs to get maintained too.

[ToucanLoucan]

Respectfully you're looking at the wrong wing of the (old?) industry. Radio plays have never been a substantial revenue source, they have a lot more in common with streaming in that way, and in fact, many labels would use paid radio placement basically as a form of advertising. The tradeoff being, for the consumer, that the radio didn't play whatever they wanted; it played what was scheduled, with an occasional request from a lucky listener.

What streaming has cannibalized is album sales, which is the problematic aspect. The fixed price-per-month to access a vast library of music is a killer value prospect for the consumer, which is why Spotify et al have succeeded as they have. However, again, it's bad for the artists; they don't make shit. And I mean, think about the economics there and it'll become evident why: previously to access between 10 and 20 songs depending on the album costed you about $10-15, one time purchase, but it was yours for the life of the media. Now you're paying (if you're paying) about $15-20 per month to access all of the music ever. And yeah there's less cost, no printed CDs, no shipping, no retail markup, but come on, if you have 4 CDs in your spotify library and are on the most expensive family plan, which IIRC is about $30/month, you're already ahead by 50%. That doesn't bode well for the artist's revenue split.

[kbolino]

We really need some sort of vaguely apples-to-apples comparison.

Looking only at CDs ca. 2005, which sold at about $15 each, an album that went platinum (1 million sales) would gross $15 million in retail, probably $10 million in wholesale, leaving about $1-3 million for the artist depending on contract terms. Estimates seem to put Spotify compensation at $3000 paid to the artist per million plays. So, you'd need somewhere between 333 million and 1 billion plays to get comparable revenue to a platinum album.

"Blinding Lights" by The Weeknd is currently the most-streamed song on Spotify at 5 billion plays [1]. That's roughly equivalent to a 5-15x platinum album from a single song and streaming platform alone; the record for most platinum album seems to be 34x platinum for Michael Jackson's Thriller [2], which had 7 songs that all were hits to some degree or another. Michael Jackson did a lot besides just record the songs; he also put out elaborate music videos and went on showy tours. All told, there are more than 100 songs with more than 1 billion plays on Spotify (in fact, the 100th is still above 2.4 billion plays).

So, I think, the situation has not really gotten worse.

[1]: https://en.wikipedia.org/wiki/List_of_Spotify_streaming_reco...

[2]: https://en.wikipedia.org/wiki/Thriller_(album)

[ToucanLoucan]

> Looking only at CDs ca. 2005, which sold at about $15 each, an album that went platinum (1 million sales) would gross $15 million in retail, probably $10 million in wholesale, leaving about $1-3 million for the artist depending on contract terms. Estimates seem to put Spotify compensation at $3000 paid to the artist per million plays. So, you'd need somewhere between 333 million and 1 billion plays to get comparable revenue to a platinum album.

Solid comparison, and illustrates the issue pretty well I feel.

> "Blinding Lights" by The Weeknd is currently the most-streamed song on Spotify at 5 billion plays [1]. That's roughly equivalent to a 5-15x platinum album from a single song and streaming platform alone; the record for most platinum album seems to be 34x platinum for Michael Jackson's Thriller [2], which had 7 songs that all were hits to some degree or another. Michael Jackson did a lot besides just record the songs; he also put out elaborate music videos and went on showy tours. All told, there are more than 100 songs with more than 1 billion plays on Spotify (in fact, the 100th is still above 2.4 billion plays).

So then let's bear that out. I don't have the data on the Thriller album for for the sake of argument, we'll rely on a quick google which says it was about $8 in 1982. I'll use the same split you did (33% off for wholesale, about 30% of which went to the artist) so if we run all that down, with US sales being 35 million units. So $55.44 million went to Michael, if we assume all of that is true and correct, which adjusted for inflation would be about 184 million dollars.

We'll compare that to the Weeknd's streams for Blinding Lights. We'll put him on the high end of receiving a half a penny per play, that works out to about 30 million dollars, for as you point out, the highest streamed song on the platform.

It's still not quite a 1:1, unfortunately, but: Michael Jackson with 7 hit songs of varying popularity was able to shift 35 million albums for a personal profit of 184 million in 2025 dollars, versus the Weeknd, who with the most popular song on Spotify, made $30 million.

And it's worth pointing out here, these are both superstar heavy hitters. The story gets a lot worse for any artist not in that category.