Can playlist sites be a boon for record labels? RIAA action against Project Playlist, a site that allows users to search for songs, stream the results and create playlists, has the media acting as if this particular company has stumbled upon something incredible. If you believe an op-ed in today's Los Angeles Times, you'd think ad-supported, on-demand music is a proven business model that is the key to turning around revenue declines.

From the op-ed:

Playlist services could be an enormous boon to the labels. The vast quantity of music online creates a need for aggregators that can introduce new songs and artists, assemble individual tracks into hours of programming, and organize the chaotic mass of music into something coherent. That's the niche that companies such as Project Playlist fill. The problem for labels and artists, though, is that their business has long relied on selling music rather than generating money from what people do while they're listening to it. And services like Project Playlist are designed to sell advertisements, not songs -- which would be a fool's errand, given how easy it is for consumers to find free music online.

How are playlist sites supposed to be a boon for record labels? There are two theories. The Times seems to imply there is a wealth of advertising revenue waiting to be tapped and that energetic-yet-constrainted startups are poised to push their diamond-coated drill bits into multi-billion-dollar reservoirs. Gushers unsue, everyone is happy. The other theory is that streaming services will spur purchases of downloads -- although it's clear the Times believes selling music is a 20th century way to generate revenue.

There are a couple problems with those theories. First, ad revenue is a poor way to monetize demand for music. Selling music -- which the Times considers an archaic way to run a business -- results in more money and more money sooner. Consider a song sold at iTunes. The label collects about 70 cents (ignoring publishing and distribution). A typical on-demand streaming payout would be, realistically, around half a cent per stream. That means the song must be streamed 140 times to generate the revenue of a single track purchase. A sale results in an immediate 70 cents while streaming spreads out that revenue over time (the time value of money dictates that money now is better than money later). The sale of an entire album results in revenue equivalent to 1,400 streams. The buyer would have to spend about 5,600 minutes -- over 93 hours -- listening to music on a playlist site to generate the same revenue generated from the purchase of a digital album. It's no wonder labels would rather attempt to spur sales rather than collect fractions of pennies. Nor is it any wonder that when labels do license music to the likes of Project Playlist, they strenuously push to maintain a certain minimum value for their product.

Second, there isn't enough advertising revenue to make ad-supported, on-demand streaming anything close to a "boon" for any of the involved parties. Make no doubt about it, this is a business model in search of an identity, in search of advertisers, in search of customers, in search of believers. We are years -- decades is more like it -- away from advertising being able to substitute for the tens of billions currently generated by recorded music sales. Today, CPMs are miserable and most services provide poor user experiences. You can't blame labels for these services' failure to find a workable business model and the lack of advertiser interest. (You can blame labels for some things. More on that later.)

Third, streaming services do not appear to be spurring a net increase in download sales. Growth in digital revenue is best explained solely by consumers' adoption of digital music (hardware, applications, etc.). And growth is slowing as the amount of music experienced online explodes. Between P2P, YouTube, blogs and the many sanctioned online services, there is an incredible amount of music being heard online. If there was a net positive effect on sales, it would have been felt by now. Sales would be through the roof. Instead, revenues were far better when consumer choice was weaker, Clear Channel controlled most radio airplay and songs were trapped on CDs. Don't concentrate just on what sales come from experiencing music on the Internet. Consider what sales are being lost as a result.

The best endorsement for on-demand streaming sites is their ability to capture revenue from a generation that has grown up with countless options for free music. That argument, though, doesn't completely hold water. Even teenagers buy music -- a lot of them still buy CDs -- which means some may buy less music because there are ad-supported services to take their place. And some older customers, who are used to paid models, may buy less as they migrate to free options.

Criticism is more valid elsewhere in ad-supported business models. The requirements placed on startups by labels will squash innovation. By demanding large advances and costly legal wrangling, labels may end up placing their stamps of approval on the wrong start ups. The best ideas and best services are not necessarily the products of the best funded companies or those executives with the most willingness to jump through the industry's hoops. Isn't it ironic that labels seek this avenue in a decade filled with "rags to riches" stories such as "American Idol" and "Britain's Got Talent"? If A&R departments followed the strategies of their colleagues in business affairs, labels would sign only artists who have had for years a superstar manager and a publisher. A&R scouts understand that success can come from the least likely of places. If only the rest of the companies acted this way.