Spotify is up against myriad challenges: widespread complaints about unfair artist payouts, accusations of higher payouts for major labels who cut special deals, two class action lawsuits around copyright infringement, plus the small detail of failing to turn a profit. Now it has another problem to deal with: musical spam.
Both legitimate artists gaming the system in hopes of a more fair payout, and bots created to harvest artificial listens, are costing the company enough money it’s starting to get aggressive about combating click fraud and spam. The problem is growing, in part as a backlash by listeners and artists to the infamously small royalties musicians earn from the service.
Spotify pays artists/rights-holders somewhere between $0.001 to $0.007 every time someone streams one of their tracks. While these figures struck many artists and listeners as shamefully low, some have used the business model to their advantage, to generate big bucks.
There are two ways to make real money on Spotify without being a popular artist: musical spam and click fraud. Musical spam encompasses a few different strategies for tricking people into playing spammy tracks. “White noise,” for instance, is the strategy of uploading uncomposed static titled and presented like popular tracks. “Sound-alikes” require more effort: these are poorly executed knockoffs of popular hits. Other musical spammers simply upload the same song thousands of times with different titles that are purposely similar to major hits. People click these spam songs out of curiosity or because they typed their desired title slightly incorrectly while searching, and over time the tracks accrue revenue.
Click fraud, somewhat differently, is the act of artificially raising the play count on a legitimate track or album in the hopes of boosting revenue for the artist or label who owns the track. The most famous case is Vulfpeck, a Michigan band whose Sleepify stunt publicly encouraged listeners to stream their completely silent made up of 30-second tracks intended to manipulate Spotify's pay-per-stream model for profit. (A song must be played for at least 30 seconds to count as a stream.) The scheme racked up massive press coverage and, in turn, huge play counts and nearly $20,000 in revenue. Spotify asked Vulfpeck to remove the album, and then when they would not, Spotify simply removed it itself.
Since then all kinds of artists and listeners began to quietly, and not so quietly, experiment with creating artificially high stream counts. And recently, Spotify seems to be intensifying its efforts to combat fraud with actions that could leave the door open to censorship.
Here’s the latest snapshot of the growing war on musical spam. An acquaintance of mine, a musician who wishes to remain anonymous, forwarded me the following email informing them their song had been removed from Spotify because it was streamed a “massive” number of times. It was sent from DistroKid, a company that serves as middleman between small artists and huge digital storefronts to help independent artists get their work onto services like Spotify and iTunes.
Redacted email from DistroKid
My anonymous source was testing how long they could get away with constantly streaming their own 45-second track on Spotify, to see whether the company had mechanisms in places to restrict artists from playing their own songs on repeat, and whether those micropennies per play would add up to some real cash. They racked up about 45,000 plays over a month before receiving this takedown email.
Sure, that's dirty play, and in a general sense, the action taken by Spotify in this case is justified and correct. But on closer inspection, the notice reveals a few troubling issues that could forecast a dark future for music fans.
These services are slowly writing the rules of what is and isn’t considered a song.
For one, Spotify took the extreme measure of removing the song entirely without any kind of probation or appeal process. A logical endgame of this type of defensive action is an environment where a corporation gets to decide how users listen to music. This is particularly unsettling considering Spotify's goal is to replace people's record collection and become their primary source of music—and there's research indicating that this goal is becoming a reality. Earlier this month, an article headlined “Streaming has replaced the personal record collection,” reported that a recent study found that “55 percent of all music listening is streaming—more than radio and the record collection combined.”
While it might not seem like a significant problem now that Spotify is removing tracks from its service, consider a possible future 10, 20, or 50 years from now where streaming services are the only way people listen to music, and then consider the frightening reality where these same corporations decide not only what songs get to remain on their platforms, but also, potentially, what kind of listening habits are permissible.
The line in the DistroKid email that stood out to me was that “Real people don't listen to the same exact song thousands of times in a row.” The problem with this sentence is that a real person did generate these plays, using only Spotify’s internal “repeat 1” button. The motivation was an experimentation with click fraud, yes, but for a company desperate to become profitable, it’s not hyperbolic to suggest that Spotify's definition of what's normal or acceptable may become increasingly stringent. It raises the question: Do you want corporations deciding which of your music listening habits are legitimate?
I contacted John Seay, an entertainment lawyer with an expertise in digital streaming, to ask him if Spotify’s removal of this song is a sign of things to come. “Your question of what is legitimate listening is interesting,” Seay told me via email. “I think Spotify's approach is that, like the Supreme Court on pornography, they know it when they see it.” That leaves some room for misinterpretation.
It reminds me of the time I became obsessed with Can’s 1972 song “Vitmain C” and listened to it on repeat (on a CD player) for three straight days. It drove me a little nuts, but it was also an incredible experience that I’m glad I had. There is nothing in Spotify's 24-page Terms of Service that prohibits shutting all the blinds in your house, disconnecting the phone, and listening to the same song endlessly on repeat. But the rules do prohibit “artificially increasing play count or otherwise manipulating the Services by using a script or other automated process.” I’m fairly certain that my experiment with Can could be interpreted by Spotify as violation of these rules.
If I replicated the “Vitamin C” experiment on Spotify, I would’ve simply used the “repeat 1” button and never turned it off, which is quite similar to what the listeners of Sleepify did that got them kicked off the site. Would my slightly insane listening habits cause “Vitamin C” to be removed from Spotify’s library?
Which leads to another question: Why remove the song instead of banning the user account? The action taken here seems strangely backwards. “Perhaps Spotify would rather keep the user, especially if it's a paid user,” suggested Seay. “It might want to preserve user accounts for marketing purposes, i.e., so that it can boost its user numbers.” In this way, Spotify users could be more valuable to the company than the songs on the platform itself.
Well, not all songs, only the ones that aren’t especially popular. If a user endlessly streamed the Beatles' shortest song, “Her Majesty,” for example, is there any way that Spotify would remove the last track of Abbey Road from its library? Or is this policy only going to apply to smaller, relatively unknown artists?
Spotify spokesman Graham James told me, “We use a mix of algorithms and humans to scan our catalog for potential fraud. Once we have identified fraudulent streaming then we take down the actual content as opposed to blocking the user. If the track is down then it is not generating any payouts, so that pretty much takes care of things.” When I asked what would happen if the track belonged to someone really famous, he replied, “that's probably too hypothetical for me to answer.”
Regardless, this policy could leave room for abuse. If I wanted to seek revenge on a relatively unknown artist that I had a personal vendetta against, couldn't I simply repeatedly stream one of their songs in an effort to get their work removed from the service? If Spotify is going to be in the business of censoring songs based on listening habits, shouldn't it publicly announce the criteria for such actions so that users can decide if they want to continue supporting the service?
Removing songs suspected of fraudulent plays is just one example of the power that digital storefronts like iTunes and Spotify have over how we listen to music, and the music itself. These services are slowly writing the rules of what is and isn’t considered a song, or an acceptable song title.
Take the case of Matt Farley, who released some 14,000 original songs on Spotify and iTunes, some with celebrity names in the title enticing people to play the tracks to earn some revenue for his efforts. He was slapped on the wrist by iTunes for breaking some rules he didn’t know existed, presumably intended to combat spam. For instance iTunes will not allow a single album to have more than 100 tracks, that he cannot have the name of a well-known album title or band in the title of his songs, and that his lyrics cannot contain the name of a record label. He was even told that he could not name a band “How To Ask a Girl to the Prom” on iTunes because it described the music so plainly.
Spotify and its ilk may be slowly squashing illegal downloads, but streaming fraud and musical spam are on the rise, and the people behind these new scams are sometimes the very artists creating the content in the first place. In an attempt to earn a tangible income in the streaming age, these musicians are inventing phantom plays of their own work and, potentially, jeopardizing their catalogue’s permanent place in everyone’s new, cloud-based, streaming record collection.