Why the Charts Are Full of Old Songs, and Why Your Playlists Are Too

Every day, more than 100,000 new songs are submitted to streaming services. More music is being made than at any point in history, and yet what people actually play is increasingly old. Luminate, the entertainment data company that began tracking music streaming in 2008, found that in the first half of 2026 about 60% of the music played in the UK and US was released five or more years ago. Current music, defined as anything out less than 18 months, was dwarfed by the back catalogue.

The UK singles chart shows it plainly. Eighteen of the year's top 20 bestselling songs were released 12 months ago or earlier. Zara Larsson's Lush Life, first out in June 2015, is the year's eighth biggest hit. Michael Jackson's Billie Jean, from 1983, sits in 22nd place, and Fleetwood Mac's Dreams, from 1977, is 30th. The industry calls this "deep catalogue," and with nearly every recorded song a button press away, listeners are reaching for it. In the US, about 60% of streams in R&B and hip-hop, the two most-played genres, are older catalogue hits.

New songs that do break through tend to stay. Sam Fender and Olivia Dean's duet Rein Me In set a UK record with 23 non-consecutive weeks at number one. In America, Ella Langley's Choosin' Texas is now the longest-running number one in US chart history. Hits need millions of streams a day to reach the top, but the concentration runs deeper: the UK's Competition and Markets Authority says nearly two thirds of streams go to music by just the top 0.4% of artists.

For artists trying to climb, going viral is not the same as building a career. Some musicians say it is hard to promote follow-up releases after a hit, because a platform like TikTok is built to create viral moments around single songs, not to boost artists. J Erving, founder of Raye's distribution company Human Re Sources, told Billboard it is easier today for people to be passive fans. To count as broken as a mainstream artist, he said, people must care about you beyond the song.

The industry says it wants to change that. Oliver Schusser, head of Apple Music, told BBC News that his service is trying to champion "tomorrow's artists" in its curation. With the Super Bowl halftime show to fill, he said, there must still be performers in 10, 20 and 30 years, and the recent headliners all started somewhere. Sofar Sounds, which works only with grassroots musicians and stages intimate pop-up gigs worldwide, held its Shoreditch Sofar Fringe festival this October, featuring East London rapper J Casa. He argues live shows let audiences form a lasting connection with a new artist. Sofar is also launching a Human Discovery Movement: people can text a mystery number and get voice-note recommendations from its tastemakers.

AI may be compounding the problem. Spotify, Apple Music and Amazon Music let premium subscribers generate playlists from text prompts describing a mood or occasion. Musicologist Anna-Kaisa Kaila says there is real potential for bias, since a system short of training data in a less commercially dominant style will struggle to surface it. London DJ and curator Tim Garcia notes that recommendations rest on everyone's listening data, so you are likely to be handed something close to what you already play. Dan Mackta, managing director of Qobuz, which curates by hand, says the technology is never going to want to throw a curveball.

Independent London artist Chiminyo goes further. Scrolling a relaxing jazz piano playlist with more than 370,000 saves on Spotify, he found top artists with hundreds of thousands of monthly listeners but unverified accounts and almost no online trace. To his ear the tracks share the same piano feel, microphone placement and tape machine, which he takes as a sign of synthetic mass production.

The scale of AI music is contested. Spotify says it has removed 75 million spammy tracks, yet Deezer says AI accounts for only 0.5% of streams on its platform. Spotify told the BBC it has rolled out platform-wide policies against impersonation and is improving its filtering of AI-related problems, while acknowledging that AI use in production sits on a spectrum. New music is not dying, but it faces a hard year, and the test is whether curation efforts can keep emerging artists visible beyond a single hit.