Had Pandora died, it would have joined myriad music start-ups in the tech company graveyard, like SpiralFrog and the original Napster. Instead, with a successful iPhone app fueling interest, Pandora is attracting attention from investment bankers who think it could go public, the pinnacle of success for a start-up.
Pandora’s 48 million users tune in an average 11.6 hours a month. That could increase as Pandora strikes deals with the makers of cars, televisions and stereos that could one day, Pandora hopes, make it as ubiquitous as AM/FM radio.
“We were in a pretty deep dark hole for a long time,” said Mr. Westergren, who is now the company's chief strategy officer. “But now it’s a pretty out-of-body experience.”
At the end of 2009, Pandora reported its first profitable quarter and $50 million in annual revenue — mostly from ads and the rest from subscriptions and payments from iTunes and Amazon.com when people buy music. Revenue will probably be $100 million this year, said Ralph Schackart, a digital media analyst at William Blair.
Pandora’s success can be credited to old-fashioned perseverance, its ability to harness intense loyalty from users and a willingness to shift directions — from business to consumer, from subscription to free, from computer to mobile — when its fortunes flagged.
Its library now has 700,000 songs, each categorized by an employee based on 400 musical attributes, like whether the voice is breathy, like Charlotte Gainsbourg, or gravelly like Tom Waits. Listeners pick a song or musician they like, and Pandora serves up songs with similar qualities — Charlotte Gainsbourg to Feist to Viva Voce to Belle and Sebastian. Unlike other music services like MySpace Music or Spotify, now available in parts of Europe, listeners cannot request specific songs.
Though Pandora’s executives say it is focusing on growth, not a public offering, the company is taking steps to make it possible. Last month, it hired a chief financial officer, Steve Cakebread, who had that job at Salesforce.com when it went public.
It is all a long way from January 2000, when Mr. Westergren founded the company. Trained as a jazz pianist, he spent a decade playing in rock bands before taking a job as a film composer. While analyzing the construction of music to figure out what film directors would like, he came up with an idea to create a music genome.
This being 1999, he turned the idea into a Web start-up and raised $1.5 million from angel investors. It was originally called Savage Beast Technologies and sold music recommendation services to businesses like Best Buy.
By the end of 2001, he had 50 employees and no money. Every two weeks, he held all-hands meetings to beg people to work, unpaid, for another two weeks. That went on for two years.
Meanwhile, he appealed to venture capitalists, charged up 11 credit cards and considered a company trip to Reno to gamble for more money. The dot-com bubble had burst, and shell-shocked investors were not interested in a company that relied on people, who required salaries and health insurance, instead of computers.
In March 2004, he made his 348th pitch seeking backers. Larry Marcus, a venture capitalist at Walden Venture Capital and a musician, decided to lead a $9 million investment.
“The pitch that he gave wasn’t that interesting,” Mr. Marcus said. “But what was incredibly interesting was Tim himself. We could tell he was an entrepreneur who wasn’t going to fail.”
Mr. Westergren took $2 million of it and called another all-hands meeting to pay everyone back. The next order of business: focus the service on consumers instead of businesses, change the name and replace Mr. Westergren as chief executive with Joe Kennedy, who had experience building consumer products at E-Loan and Saturn. Pandora’s listenership climbed, and in December 2005, it sold its first ad.
But in 2007, Pandora got news that threatened most of its revenue. A federal royalty board had raised the fee that online radio stations had to pay to record labels for each song. “Overnight our business was broken,” Mr. Westergren said. “We contemplated pulling the plug.”
Instead, Pandora hired a lobbyist in Washington and recruited its listeners to write to their representatives. “A lot of these users think they’re customers of the cause rather than users per se,” said Willy C. Shih, a professor at Harvard Business School who has written a case study on Pandora. “It’s a different spin on marketing.” The board agreed to negotiations and after two years settled on a lower rate.
Some music lovers dislike Pandora’s approach to choosing music based on its characteristics rather than cultural associations. Slacker Radio, a competitor with three times as many songs but less than a third of Pandora’s listeners, takes a different approach. A ’90s alternative station should be informed by Seattle grunge, said Jonathan Sasse, senior vice president for marketing at Slacker. “It’s not just that this has an 80-beat-a-minute guitar riff,” he said. “It’s that this band toured with Eddie Vedder.”
Yet in 2008, Pandora built an iPhone app that let people stream music. Almost immediately, 35,000 new users a day joined Pandora from their cellphones, doubling the number of daily signups.
For Pandora and its listeners, it was a revelation. Internet radio was not just for the computer. People could listen to their phone on the treadmill or plug it into their car or living room speakers.
In January, Pandora announced a deal with Ford to include Pandora in its voice-activated Sync system, so drivers will be able to say, “Launch my Lady Gaga station” to play their personalized station based on the music of that performer. Consumer electronics companies like Samsung, Vizio and Sonos are also integrating Pandora into their Blu-ray players, TVs and music systems.
“Think about what made AM/FM radio so accessible,” said Mr. Kennedy, Pandora’s chief. “You get into the car or buy a clock for your nightstand and push a button and radio comes out,” he said. “That’s what we’re hoping to match.”