As the music industry becomes increasingly conscious of — and vocal about — the challenges of the streaming model, fraudulent streams have become a source of growing frustration. “Every penny that goes to a fraudulent stream is a penny that doesn’t go to a legitimate stream,” says Richard Burgess, president and CEO of the American Association of Independent Music. “Fraudulently increased stream counts can affect recording budgets, licensing deals, catalog valuations and can result in the misallocation of marketing budgets.”

The French government, which recently published the results of a months-long, country-wide investigation into streaming fraud, portrayed understanding the impacts of this activity as an imperative. “The stakes are high in our country as well as in the rest of the world: the development of music services, which can be free and financed by advertising, or paid through subscriptions, as individual or family plans, constitutes a tremendous opportunity for the music sector, after years of a long crisis,” the report asserted. “…Such growth whets the appetites and stimulates the creativity of those who seek to abuse the system.”

“The multiplication of fake streams, that is to say the processes allowing [bad actors] to artificially boost play counts or views to generate an income, is nothing short of theft,” the report continued.

The French study, conducted without data from YouTube, Apple Music, or Amazon Music, found that 1% to 3% of plays were fraudulent, while also noting somberly that “the reality of fake streams goes beyond what is detected.” BeatDapp, a Vancouver-based company that creates fraud detection software for labels, publishers, distributors and streaming services, believes the global level of fraud is higher. “In 2020, estimates were 3 to 10% of all streaming activity was fraud,” the company wrote in 2022. “Today, we confidently say it’s at least 10%, and more in some regions. That equals ~$2B in potentially misallocated streaming revenues this year, and will be ~$7.5B by 2030 if left unchecked.”

So what forms does streaming fraud take? According to Burgess, the practice “covers a multitude of techniques used to increase stream counts or impressions by other than legitimate means.”

Here are three of the most common:

Bots

Discussion of streaming fraud often turns quickly to bots, which Burgess defines as “automated software that can be used to generate views, streams or interactions.” To detect bot activity and prevent it from affecting royalty payouts, companies build models that trawl streaming data and look for listening patterns that appear anomalous: BeatDapp likes to discuss an example of finding tens of thousands of accounts all streaming the same 63 songs.

“If I’m trying to push numbers up, I’m going to do it across streaming services in a subtle fashion this way,” BeatDapp co-CEO Andrew Batey says. “Spread it across a lot of accounts and multiple platforms, and you can drive a significant number of plays with no one looking.”

Click Farms

Streaming services are looking for suspicious play patterns that don’t reflect human behavior. Fraudsters are aware of this, so they try to camouflage their activity in ways that appear human. One method is to get actual humans to press Play through what are known as “click farms.”

Eric Drott, a professor at the University of Texas in Austin who has written about streaming fraud, describes these as “enterprises concentrating low-paid, precarious workers who are engaged to perform the sort of rote, repetitive tasks that keep the flows of digital capitalism moving: creating social media accounts, moderating content for platforms, clicking online ads, liking or rating items and, of course, generating plays on streaming services.” Accounts that stream music 24 hours a day or stem from a smartphone that never moves or dips below 100% power could be evidence of click-farm activity.

Imposters

A third prominent form of fraud identified by Burgess involves impersonating creators by uploading a version of their song to streaming services and illegally collecting creators’ legitimate royalties. This is a common problem faced by artists who are having a moment on TikTok, for example: Imposters post a version of the TikTok hit on streaming services under a slightly different name, aiming to divert some streams (and hopefully royalties) their way.

“It happens to every single viral artist,” says one manager who shepherded a viral act to a major-label deal last year. There are many distribution companies out there, and managers say that some of them have lax oversight of what’s being uploaded to the DSPs through their platforms. This means artists and their teams have to keep close watch on streaming platforms and issue takedowns when they find imposter versions.