— DYLAN

Pandora led the internet radio revolution for years by delivering highly-relevant playlists based on their Music Genome Project algorithm. But somewhere along the way, Pandora fell by the wayside. Their technology is still impressive but besides rolling out mobile support and updating their desktop interface every once in awhile, they haven’t really innovated

If they don’t make some radical decisions in the next few years, I believe that Spotify (and a consortium of smaller competitors like Soundcloud, Hype Machine, etc.) will slowly eat away at their user base, listening hours, and ad revenues. Here’s why:

Obsolete Vernacular: The mere fact that Pandora has the word radio in its title is the first sign of its eminent doom. Why model your service of a medium that had its heyday in the 1930’s? For me, radio connotes bothersome advertising and constantly switching stations in search of something palatable. Sure, if you just want to put some noise on in the background then Pandora is a great option. But in the era of active listening, iTunes and Spotify users want more control over what they listen to. They want to save what they hear for later and easily share songs with friends. To the next generation of music listeners, being forced to listen to whatever song comes up next will seem as archaic as vinyl (maybe neo-Hipsters will listen to Pandora ironically).

Data: Pandora has always operated under the somewhat faulty premise that science is the best approach to music curation. In 2000, cross-referencing songs on hundreds of objective parameters may have been the best method. But in the age of social media, are they still looking at the right data set?

Let’s suppose that Spotify has the capability to analyze the complete libraries and listening habits of all its users. Shouldn’t they be able to figure out just as well as Pandora can that The Doors, The Stones and The Beatles make for a good playlist? In the long run, this newer curation methodology could result in more serendipitous curation algorithms based on what people actually listen to rather than playlists based on pure similarities.

Globalization: From a growth standpoint, Pandora and Spotify have both been limited by licensing issues. Pandora has been siloed in the US and Pandora was only able to enter the market here in July. But by making it work here, Spotify has shown that they’re willing to fight to take their product global. Pandora may still have the advantage when it comes to active monthly users (51M vs. 18M) but I could easily see Spotify tipping the scales in the next two to three years if they can sustain their momentum and gain footholds around the world.

Social: Spotify is a good product without Facebook integration but add in easy sharing, collaborative playlists, and the news ticker and all of a sudden you’re challenging not just Pandora but iTunes as well. As an added bonus, social integration will theoretically allow Spotify to be much more targeted with their advertising. Pandora can serve you ads based location and listening habits but Spotify could provide the borderline-creepy level of exactitude that Facebook ads are known for.

The Ecosystem: By opening up their platform to HTLM5 app development, Spotify forges connections with existing media brands and outsources some of their innovation for free. My gut tells me that more apps will result in more listening time per user, a higher perceived value of the service, and the potential for a supplemental revenue stream down. Though it wouldn’t make sense strategically, a Pandora app in the Spotify ecosystem would be a beautful thing.

Freemium Model: Pandora’s paid offering is what I call a de-handcuffed product: you pay to make their service marginally less shitty. Spotify Premium, on the other hand, is a whole new experience entirely. In addition to no advertising, you get access on your mobile devices which allows for a synced library between desktop, phone and tablet. I wouldn’t be surprised if mobile devices get access to Spotify apps as well within the next year. They also have roughly 3 million paid subscribers and counting which generates between $15M and $30M in monthly revenue. Pandora is extremely reliant on their advertising whereas Spotify gives its users a reason to pay.

Pandora simply won’t last if they don’t start to address the evolving music landscape. Solid social integration would be a good start. Perhaps they need to target an older demographic and leave the young listeners to Spotify. Stagnation, however, is clearly a losing strategy and when your market cap is approaching what Facebook spends on a zero revenue acquisitions.

UPDATE (4/30/12): In my original post I neglected to mention that the CEO’s of Spotify and Pandora were recently asked what single band or musician represented their company. Spotify said Radiohead because “they’ve always pushed the boundaries and done things a bit differently”. Pandora’s response: “There is no one artist or band that defines Pandora, because by definition we’re personal. The definition of Pandora is, ‘Who is my Pandora?’ … It’s kind of like that Time Magazine Person of the Year, it’s you.”

Enough said.