Data and recent trends suggest that pop songs are getting shorter – we asked the experts how streaming and social media are making new rules
What’s a good length for a song? It’s the music industry’s version of the proverbial “how long is a piece of string?” For decades, the average song length has been a little over three minutes. The root of this is assumed to be technology: in the 40s, to get played on the radio, your song had to be recorded on a 45rpm (revolutions per minute) disc, which only allowed for three minutes. Since then, pop songs have, on average, gotten slightly longer, but not by much; radio stuck to the status quo, and it seems we unanimously agreed, somewhere along the line, that between three and four minutes was best.
Then came Tierra Whack. The visionary Philadelphia artist achieved critical acclaim in May 2018 when she dropped Whack World, a 15-minute audiovisual project made up of 15 one-minute songs. As well as showing how satisfying a one-minute song could be – her ideas are somehow richly developed, yet always leave you thirsty for more – Whack clearly had Instagram in mind, as each of her new videos fits perfectly within the grid’s 60 second limit. With Kanye West’s extremely economic Wyoming releases, Lil Baby and Drake’s brief “Yes Indeed”, Bebe Rexha’s under-three minute “Meant To Be”, and even Lil Pump’s irrepressible two-minute 2017 chart hit “Gucci Gang”, we’re seeing other artists experiment with super-short length. Instagram, too, is becoming a common part of artists’ release strategies: Jaden Smith recently debuted a whole album there.
These are artists who are playing with extremes. They’re testing out novel ways to debut music that makes it stick in a new, fast-paced world of ephemeral content. But there’s some evidence to show that they’re also part of a wider trend towards brevity in pop songwriting – at a glance, it seems short, Instagrammable songs could be the new norm.
Last year, data scientist Mark Bannister put together a report after analysing the past six decades of number one hits on the Billboard Hot 100. He found that, on average, US number one singles had increased from around two-and-a-half minutes in the early 60s, to four-and-a-half minutes in the late 80s. Then, things took a turn. “You could argue that US number one singles started getting shorter from 1991,” he writes in an email to Dazed. “At the time I wrote the report, duration was below the long term average of 3.8 minutes.” So hit songs, on the whole, are getting shorter, after decades of the trend moving in the other direction. But why?
A common theory is that music is getting shorter because our attention spans are shrinking. In his report, that’s what Bannister suggests, linking to a Telegraph article that claims humans now have shorter attention spans than goldfish. Last year, producer Mark Ralph also told the BBC that he believed song intros were getting shorter because “attention spans have now decreased.”
The idea of an “average attention span”, though, is one that’s in dispute. Dr Gemma Briggs, a senior lecturer in Psychology at the Open University, researches attention in relation to dual tasks like talking on the phone while driving. Speaking in an email to Dazed, she explains that to her knowledge, the idea that our attention spans are generally getting shorter isn’t supported by “any credible evidence”.
“Research into attention tends to look at how well people can focus their attention on specific tasks, rather than how long they might pay attention to something,” she continues. The concept of an “average attention span”, she says, is too vague to be a useful term to researchers. “How we apply our attention very much depends on what we’re doing: we may have to make more effort to pay attention to a visual task, for example, than an auditory one, so discussing an average attention span wouldn’t really tell us anything meaningful.”
What’s true, however, is that there’s more content and tasks tugging at our attention than ever. “It could be argued that the growth in social media has shown how distractible we can be, and how important it is to control our attention to a main task,” says Briggs. “But there’s nothing to suggest that engaging with social media affects how long we can pay attention to something.”
So perhaps songs are getting shorter not because we can’t pay attention to longer songs, but so that they can quickly distinguish themselves from the billions of other songs competing for our ears on streaming platforms. Laura Snapes, deputy music editor of the Guardian, puts it like this: “I don’t think we’ve lost the ability to listen to a long song, but Spotify induces a very non-committal listening mentality. I suppose the other streaming-related thing is that a shorter song can be played more times in an hour.” One hundred and fifty streams of a song now counts as one song purchase, in Billboard chart terms (in the UK, you need 100 paid streams or 600 free streams). The charts literally reward songs that both catch your attention fast, and leave you wanting more, hitting the replay button again and again.
“I definitely think streaming has allowed many people to explore and let the music talk for themselves rather than be dictated by the labels. There’s no rules right now” – Carla Marie Williams
As well as Spotify, the insurgence of artists experimenting with songs and teasers that are a minute or less is evidence of the impact of social media. Instagram, Musical.ly, and other social media apps are quickly becoming one of the main ways young people share and discover new music – could that be making an impact on music itself? Speaking over email to Dazed, Tierra Whack doesn’t say that Instagram was the reason she wrote her “nursery rhyme” inspired one-minute songs, writing: “I wanted my project to be 15 minutes exactly, and in order for that to happen every song had to be a minute.” She acknowledges, though, that the platform was on her mind: “Instagram just made sense – 60 second videos.”
However, it could be easy to over-state the role of technology in all this. In an analysis for the Financial Times earlier this year, journalist Ludovic Hunter-Tilney cast doubt on the “Cassandra-like” claims that Spotify and its competitors are having a significant effect on songwriting. He notes that the only area of songwriting that shows a notable change since streaming rose to prominence in 2010 is the amount of swearing in lyrics – “Freed by streaming from relying on radio airplay,” he wrote, “the likes of Migos’ ‘Bad and Boujee’ appear to have ushered in the age of the profane chart-topper.”
While hit songs are leaning towards the short and sharp, songs that don’t make it high in the charts vary in length massively. If anything, artists today seem liberated to experiment with extra-long songs, as well as the extra-short. Snapes notes that recent releases from Chvrches, All Saints, and Years & Years have all contained sprawling songs that run almost too long. The cultural moment is perhaps best summarised by the experimental pop album I’m All Ears from Norwich duo Let’s Eat Grandma. Some of their songs simmer and unfurl over 10 minutes; elsewhere, they clock in at little over a minute.
Carla Marie Williams has been writing hit pop songs for over a decade. She worked on “The Promise” and “Can’t Speak French” for Girls Aloud in the 00s, and more recently on “Runnin’ (Lose It All)” and “Freedom” for Beyoncé. In her view, writing pop songs is much freer now than it used to be, when radio play was king. “With Girls Aloud, because it was straight-up pop, (with) Brian (Higgins) who ran the production house, we used to be tight on those melodies, and tight on the length of the song. Because the audience was very much about the instantaneousness of it. It needed to be sharp.” By comparison, she says, “when I was doing ‘Runnin’ and ‘Freedom’, it was more about the message. I was listening to ‘Runnin’’ the other day, and I was like, ‘wow, the build is really slow.’
Radio has long dictated that songs needed to be under three and a half minutes to be played – this is perhaps a lingering effect of the days when technology meant songs had to be that long to be put on a record that the station could play, but really, there’s no reason for this other than it’s the way that things have always been. As streaming has become more dominant, though, radio stipulations matter less. “Once the song is done, we look at length,” says Williams. “We do sometimes have to chop them down.” But, she adds, “I definitely think streaming has allowed many people to explore and let the music talk for themselves rather than be dictated by the labels. There’s no rules right now.”
There’s two stories you could tell here: one about how technology is shrinking the length of the average pop song, and one about how tech is opening up a more experimental playing field for pop music at large. It doesn’t seem like there’s quite enough evidence to support the first one. Over on Wired in 2014, Rhett Allain noted that while songs were originally three minutes long because of the technology they were recorded with, artists have always pushed the boundaries of what’s possible. Between 1970 and 1980, songs got longer even though artists weren’t yet using CDs; and yet, when CDs became widely available, and technology allowed for songs of any length, artists still on average stuck to making tracks of about four minutes. The conclusion that Allain drew was that there are bigger factors at play here than technology.
It’s pretty reductive to assume there’s a neatly linear relationship between the rise of streaming and social media and the length of songs. It has echoes of when adults say kids don’t have as good attention spans any more, or can’t form relationships, because of their phones – there’s actually not really an empirical way to measure that, and it’s a sweeping, vague conclusion to make, largely rooted in technophobia. Perhaps what is happening with song length right now is actually more of an artistic reaction to a status quo that hasn’t been shaken up in decades. Perhaps it’s a reflection of an industry that’s in flux, as artists adapt to the fact that there’s so many different ways in which their music can be encountered by fans now.
“Trends in music are usually a reaction to the previous trend,” says Mark Bannister, the data scientist who analysed the history of the Billboard Hot 100. “For example, the way punk (short, simple songs) was a reaction to prog rock (long, complicated songs). So in theory, I think we could see songs get longer again at some point.” His own research revealed that trends in song duration aren’t precisely mappable against changes in technology – the introduction of CDs, by his analysis, seems to have had less of an impact than say, the shift from rock to pop domination.
Snapes concurs that “every generation reacts against the one that came before”, and so it’s most likely that the average pop song will keep on expanding and contracting in length – but, she adds, “I do think the 3.5 minute pop song is immortal.” Right now, we’re seeing both artistic innovations and commercial cash-grabs in response to the growth of Instagram and streaming – as with any generation, ours is adapting to new technologies and cultural shifts. This doesn’t necessarily mean that the art of songwriting is dying, nor that pop songs are turning into ad jingles; a great pop song, long or short, will live forever. Or, as Snapes aptly puts it: “‘Wannabe’ is 2:53 – what more do you need?”