In the autumn of 2007, five musicians from Oxfordshire invited millions of people to pay whatever they thought an album was worth—and discovered that nobody really knew what music was worth.
The album that would blow up the familiar machinery of distribution was born slowly and painfully—the way things are born in a band that had long since stopped rushing to meet somebody else’s deadlines. Recording began in February 2005, a year and a half after the release of Hail to the Thief, and stretched through June 2007. Over those twenty-eight months, Radiohead moved between several spaces: the country-house Halswell House, the grand Tottenham House in Wiltshire, London’s The Hospital Club, and their own studio in Oxfordshire. Each of these places was more a residence than a studio in the usual sense: the band moved in, lived there, worked there, and left when the walls stopped helping.
The production story did not run in a straight line either. First behind the desk was Spike Stent—an experienced mix engineer who had worked with Massive Attack and Björk. But the sessions with Stent produced nothing the band was happy with; the collaboration was called off. Back came Nigel Godrich, Radiohead’s constant sonic collaborator since OK Computer, and the album finally took shape. Godrich is a figure without whom it is impossible to understand why this band’s records sound the way they do: at once engineer and director, a man who knows how to turn hours of jamming into architecture. By the summer of 2007, the master recording was ready, and the management—Bryce Edge, Chris Hufford, and Brian Message—faced a question that had really been ripening since 2003: what do you do with an album when you no longer have a label?
Radiohead arrived at EMI in the early nineties—a young guitar band signing the kind of standard multi-album deal common at the time. By 2003, when Hail to the Thief came out, their contractual obligations had been fulfilled. The band was free—and crucially, free at the very moment EMI itself was coming apart at the seams. In 2007, the label passed under the control of the private equity firm Terra Firma, headed by Guy Hands, a financier with no connection to the music business. Hands tried to restructure the company, and negotiations with Radiohead over a new contract and rights to their back catalogue took place, but collapsed. The terms offered by the new EMI suited neither the band nor its management.
This context often gets lost in retellings. Radiohead did not walk away from their label in a burst of idealism. The contract expired on its own, the negotiations went nowhere, and the album had already been recorded. The band found itself in an extraordinarily rare position: a finished product, global name recognition, complete freedom of control, and not a single obligation to a corporate partner. This was not a manifesto but a set of circumstances that management turned into an opportunity. Revenue from the digital release flowed through Xurbia Xendless Ltd—a legal entity created to manage the band’s finances outside the label infrastructure.
On 1 October 2007, guitarist Jonny Greenwood posted a message on Radiohead’s website whose meaning fit into a single sentence: the album was ready, it was called In Rainbows, and it would be out in ten days. No teaser six months in advance, no radio singles, no magazine ad campaign. The ordinary promotional cycle for a major release—those months-long negotiations with retailers, cover approval with the marketing department, press embargoes—was replaced by one blog post.
On 10 October 2007, the digital version appeared on the band’s website. Technically, the process looked like this: the buyer went to the page, entered the amount they wanted to pay (including £0), and received a link to a ZIP archive containing 160 kbps DRM-free MP3 files. Delivery was handled by PacketExchange, a network specializing in the rapid distribution of heavy content. Distribution was coordinated by W.A.S.T.E.—the company that had managed Radiohead’s merchandise and direct sales since the late nineties, effectively functioning as a mini-label answerable to the band itself. Digital sales through the website continued until 10 December 2007—exactly two months.
What still fascinates me is that ten-day gap between the announcement and the release. It worked like a vacuum: no critic had time to write a review, no pirate could leak the album (it did not yet exist online), and anticipation grew exponentially—through word of mouth and news headlines alone. The band had, in effect, replaced a marketing budget with a single gesture.
The “pay what you want” principle was not invented by Radiohead. Street musicians had been working that way for centuries; in the digital world, indie bands and individual artists had conducted similar experiments long before October 2007. The difference was scale. It is one thing when an obscure electronic musician offers an EP for any amount and is delighted by two thousand downloads. It is another when a band on Radiohead’s level, with an audience measured in tens of millions, puts a finished album on its own website and says: you decide. It was a stress test, and what it tested was human behavior—technology was the easiest part.
The model was simpler than it appeared from the outside. The band’s website took the payment—or no payment—W.A.S.T.E. processed the orders, PacketExchange provided download speed, and the money flowed into Xurbia Xendless Ltd. No intermediary like iTunes, Amazon, or a retail chain. No label cut. Every pound paid by a listener, minus processing and infrastructure costs, went directly to the band. For an artist who would normally receive label royalties in the range of 10–20% of the wholesale price, this was arithmetic from another universe.
But the model had a downside too: 160 kbps is not studio quality. The files sounded perfectly decent for 2007, but that bitrate did not satisfy audiophiles or collectors. And the band understood this: the digital release was only the first act.
Alongside the digital giveaway, the band’s website opened pre-orders for a limited edition—the discbox, priced at £40. Inside: a CD of the album, two 12-inch 45 rpm vinyl records, the bonus disc Disk 2 with eight additional tracks, an art book featuring illustrations by Stanley Donwood, and lyrics. It was a physical object made with a degree of tactile attention impossible to reproduce in a ZIP archive. Discboxes began reaching buyers in December 2007, and demand was such that deliveries stretched on.
Then, on 31 December 2007 / 1 January 2008, came the third act: a standard retail edition on CD and vinyl was released through independent label XL Recordings (in the States, through TBD Records). The band did sign a licensing deal after all, but on its own terms: XL received the right to physical distribution, not the master recording. The album debuted at number one on the UK Albums Chart and topped the US Billboard 200. It later won a Grammy for Best Alternative Music Album.
The three-tier structure—a free/cheap digital file, an expensive collector’s object, a mass-market retail edition—was a carefully designed funnel. Each level served its own audience: casual curiosity-seekers, devoted fans, and those who still bought music in stores. By October 2008, total sales of In Rainbows had surpassed 3 million copies worldwide—a figure including all formats and channels.
This is where the murkiest part of the story begins, and it is the part that irritates me most. Almost immediately after the release, analytics company comScore published a study claiming that 62% of downloaders had paid £0 for the album and that, averaged across all downloads including free ones, the average payment came to around $2.26. These figures instantly became canon: the press seized on them as proof that people were not willing to pay for music if given a choice.
The problem was that comScore’s methodology raised serious questions. The company surveyed a panel of users rather than having access to Radiohead’s actual server data. Essentially, it was an extrapolated poll, not a transaction audit. The band’s management publicly challenged the accuracy of those estimates but—and this is the crucial part—never disclosed its own data. Not the exact number of downloads, not the average price, not total revenue from the digital release. Silence, allowing each side to see in the experiment confirmation of its own beliefs.
At the same time, BigChampagne, a company that monitored file-sharing networks, together with the British society MCPS-PRS, calculated that 2.3 million people had downloaded the album from torrents in the first two months of its free availability, with the peak coming on the first day of tracking—around 400,000 downloads. These figures need context too: people were downloading the album through torrents even though it was available free on the official website. Partly out of habit, partly because the band’s servers could not handle the load in the first hours, partly because torrent versions sometimes offered a higher bitrate (re-encoded, naturally, from the same 160 kbps, which made it pointless, but who was going to stop them).
Trying to isolate “revenue from the digital file” and judge the experiment by that metric alone is an accounting trap. The digital giveaway worked as a giant sample, generating demand for the discbox, for the retail CD, for concert tickets, for the band’s catalogue as a whole. Thom Yorke later said the band was happy with the financial outcome, but he never named specific figures—and probably never will.
After In Rainbows, the media split into two camps. Some wrote that Radiohead had “killed the music industry” and proved labels pointless. Others said the experiment had failed because most people downloaded the album for free. Both versions are caricatures.
Radiohead did not destroy labels. They themselves used the services of XL Recordings for physical distribution two months after the digital release. They did not “invent free music”—Napster had made music free seven years earlier, only without asking. And pay-what-you-want as a principle already existed in dozens of variations. What the band did was use a specific window of opportunity: an expired contract, a finished master, worldwide recognition, and a W.A.S.T.E. infrastructure capable of handling hundreds of thousands of transactions. Try repeating that if you are a debut artist with no audience, no recorded album, and no money for servers.
Thom Yorke himself said this honestly: the experiment was possible precisely because Radiohead were Radiohead. An emerging artist offering their debut for any price will get mostly zeros and silence, because they have no accumulated trust, no fanbase willing to pay out of loyalty, no touring machine that monetizes attention. In Rainbows is a demonstration of what a band with a twenty-year history and complete control over its recording can afford.
By 2008, the dust had settled, and it became clear that In Rainbows had not rewritten the industry’s rules but had shifted the conversation. Labels began discussing direct sales, flexible pricing, and the importance of direct artist-audience contact more actively. Bandcamp, launched in 2008, built pay-what-you-want into its core model and became a platform for thousands of independent musicians—it is hard not to see the shadow of In Rainbows in that architecture. Streaming services, which would swallow the market a few years later, solved the problem differently: instead of “name your price,” they offered “pay a fixed subscription and listen to everything,” effectively setting the price of an individual album at zero, while making up for it with volume.
Radiohead, for their part, did not repeat the experiment. Their next album, The King of Limbs (2011), was sold through their website, but at a fixed price. The band found an answer to its own question—and apparently it proved more complicated than “just let people decide.”
The central legacy of that October is not the sales model or the figures that nobody really knows in full anyway. It is the question that hung in the air and has still not received a final answer: what exactly is a listener buying when they pay for music? A physical medium? A file? The right of access? A relationship with the artist? The feeling of a fair deal? In Rainbows forced millions of people to confront that question personally—at the moment when the cursor blinked in an empty price field, with no hint beside it, no suggested amount, no one else’s opinion.