Billboard Fights the Algorithm
Hello, I hope y’all are doing well. I wanted to start with a question: Do y’all like the links at the end of the newsletter? Just wondered if anyone actually reads them, or if that space could be better used, drop a line at pennyfractions@gmail.com, if you got some pressing thoughts on the matter. If you enjoy the newsletter every week the best way to support is to recommend to a friend, tell coworker, or drop it in an illuminati Facebook group. Whatever feels best for you, anyway let’s chat on some charts.
Let’s start this week in the rumor mills of Hits Daily Double, the American music industry’s best collection of delightfully snarky news and gossip. The nugget of a story I wanted to highlight was one that centered around the Billboard Top 200 chart: "Label sources say Spotify was unhappy with the initially proposed 5000:1 stream-to-album ratio. The bible then countered with a 3750:1 proposal that was also met with disdain. It's been rumored that Spotify has threatened to withdraw from reporting to SoundScan should they remain unhappy with any final decision."
I think at this point I know y’all are fairly avid followers of the Billboard charts, so this rumor over a lack of change with the Charts formula shouldn’t be surprising. Drake’s latest single “God’s Plan” debuted at no. 1 on the strength of streams, Migos scored 14 songs on the Hot 100 chart tying a Beatles’ record, and while the Atlanta trio only sold 38,000 physical copies of their latest album Culture II. Billboard reported it achieved 225,000,000 individual song streams, or about 150,000 equivalent album sales. If anything streaming at the moment is actually taking up a larger slice of Billboard’s charts formula.
Speculation over the role of streaming on the charts caught broader attention last fall thanks to everyone’s favorite rapper, who doesn’t want to be a rapper, Post Malone. Republic Records released a YouTube video for his single “Rockstar” that only featured a loop chorus. Matthew Trammell, a writer for the New Yorker, initially tweeted about it, then Fader picked it up, as it appeared to be a genuine loophole in Billboard’s system for measuring song popularity. A few other labels tried similar chicanery, but YouTube told Pitchfork in December they were putting a quash to the practice, which I found evidence of in a removed Big Sean video that said: “This video has been removed for violating YouTube’s policy on spam, deceptive practices, and scams.”
A few days after this initial Post Malone backlash, Billboard announced a change to their charts that targeted YouTube and Spotify’s free ad-supported tier. The updated Hot 100 and Top 200 charts would lower the weight of ad-supported streams compared to streams coming from subscription based services. The basic idea being that it doesn’t make sense to equally weight a stream from a YouTube video and Tidal, when according some data that Tidal stream is worth significantly more than a YouTube one. When I wrote about the initial change on TrackRecord I said: "Billboard’s new changes suggest that the company does not want measure blanket awareness of a song. Instead, Billboard, a self-proclaimed “unbiased” force of measurement, reveals that they favor those who pay and participate in the music ecosystem over passive consumers—at least with the Hot 100 chart...
The charts aim to measure popularity, but it’s becoming clear that Billboard is constantly recreating their definition of “popularity,” which has more to do with the financial gain derived from one’s work than how well-known it is."
What is Billboard’s goal with this change? The Billboard songs charts until the introduction of the digital downloads was purely radio airplay and record sales for the second half of the 20th century. Radio was effectively judging promotion and physical sales was capturing the monetary transaction derived from said marketing. That relationship is always rocky, but the basic idea was if you heard a song on the radio, you’d buy it. The record industry boom of the 90s help break that system, because when labels started pushing songs to radio, but never properly releasing them as singles, fans were stuck footing a higher bill of a full album just to get a single song.
Legal digital sales from iTunes help return power to the consumer, because it brought the single format back in a way it was hadn’t been in the previous decade. That similarity of digital downloads, might explain partly why Billboard was slow to incorporate streaming number in their formula, because it is fully breaks away from a near century old model of music consumption. No longer transaction based the function of streaming is almost purely promotional; that’s an issues that’s hung up many in the industry, not just the charts. The arrival of streaming on the coattails of digital downloads wrongfully conflated two modes of consumption that couldn’t be more divergent. Abstractly streaming is far closer to radio in what it’s measuring than digital sales, but the conversation this decade centered on comparing streams to downloads, when they’re in fact two entirely separate ways of understanding music consumption.
The distinction is important, because if streaming were contextualized like radio, where the numbers are reflected as audience impressions and reach, rather than direct 1:1 listens perhaps there might be a slightly healthier conversation around streaming information. The iTunes store digitizing the physical realm (Vinyl, CDs, Cassettes, etc.), but streaming effectively put the music industry on a path towards a radio like model and finally leaving behind the conceiving of music through a checkout counter lens.
This isn’t an easy spot to be for Billboard. Whether it is Spotify or another company holding up the chart adjustments, this would be a good moment to imagine new ways the charts could potentially work. The social media and streaming analytics site Chartmetric recently introduced a chart devoted to the number of saved playlist a song can get on. Billboard could try and get picker about what is a “stream” putting a fine comb over 1 second streams versus 30 second streams ones. It’s be easy to see how Spotify might not like hearing those kinds of changes, when they along with YouTube, encourage passive usage. A recent Guardian story that centered on YouTube’s algorithm, even got a little data showing just how big passive plays are for the platform: "Daniel Alexander Cannon, a conspiracy theorist from South Carolina, tells me: “Every video I put out about the Clintons, YouTube would push it through the roof.” His best-performing clip was a video titled “Hillary and Bill Clinton ‘The 10 Photos You Must See’”, essentially a slideshow of appalling (and seemingly doctored) images of the Clintons with voice-over in which annon speculates on their health. It has been seen 3.7m times on YouTube, and 2.9m of those views, Cannon said, came from “up next” recommendations."
For this creator that would equate to 78% of a video’s traffic coming from the “Up Next” slot, which on YouTube is automatically set to default. I even thought of this in regards to genres like Reggaeton and Latin Trap, where YouTube mixes offer a fairly small pool of artist videos, so if you start with Ozuna, you’ll constantly see Bad Bunny, J Balvin, Romeo Santos, etc. An endloop feedback loop of view ends up being created and while I don’t doubt their popularity. I do question if their insanely high YouTube numbers properly reflect engaged listening. This is another trend that Billboard’s shift would potentially dampen if those streams weren’t so highly weighted.
I don’t agree with every shift on the Billboard charts, but the Hot 100 and Top 200 albums chart historically were not measuring awareness, but engaged consumption. Increased data around music listening could offer new and exciting ways to contextualize how people are actually listening to music. Yet that really isn’t happening. Instead, the physical album model holds strong, because Billboard and the broader industry is stuck with that paradigm. Once the charts start to better reflect all of the data and information that is out there, hopefully it’ll lead to more accurate charts, but also towards an industry that better understands it’s relationships to its fans.
Links 2 Read
Apple Music on Track to Overtake Spotify in U.S. Subscribers - Wall Street Journal
I’ll be glib: Big If True.
'Fiction is Outperforming Reality': How YouTube's Algorithm Distorts Truth - The Guardian
Similar to the New York Times piece about Twitter bots, this piece about the YouTube algorithm was endlessly fascinating to me. I quoted it earlier in the newsletter, but the piece centers no only on the political implications of the YouTube algorithm, but also just on the basic ways that it’s designed in a way to exploit human behavior in ways it appear that creators and YouTube don't fully understand. Truly I love our advertising first, humanity seventh hellscape.
Do loot boxes have a place in the music industry? - Cherie Hu
First off, please subscribe to Cherie Hu’s newsletter. Last week, she put her focus on the topic of video game loot boxes, effectively unregulated digital slot machines, and wondered if this might be a business model the music industry might want to consider and how it potentially clashes with other music industry trends. There is a predatory nature of the business model I certainly don’t trust, but I do think music fandom and how fans support artists should try weirder ideas and this is undoubtedly a weird idea.