Shares of digital music service Pandora Media (NYSE: P) jumped as much as 19% on Thursday, thanks to a surprisingly strong first-quarter earnings report. If you bought shares at rock-bottom prices in the low $8 range, you're sitting on a 40% return in just four or five weeks now. Congratulations.

But does this bounce have any staying power? Let's dig into the numbers and the story to find out.

Fool by numbers
Pandora posted a $0.09 non-GAAP net loss per share on $80.8 million in revenue. That's 58% year-over-year revenue growth, but the losses actually tripled. Moreover, the year-ago period's $690,000 of free cash flow turned into an $11.8 million pile of burning cash.

The average analyst expected a $0.08 loss per share on revenue of just $74 million, so these results absolutely crushed the Street view. But you still have to wonder if that's good enough. Losing money forever is not a sustainable business model, even if analysts estimated that you would lose more.

Mo' users, mo' problems
Pandora's biggest problem is very simple: The more users the service gets, the more money it loses.

If you don't believe me, this is how CFO Steve Cakebread explains the situation: "We expect negative cash flow from operations resulting primarily from increased listener hours and resulting content spend." In other words, Pandora burns more cash when its customers use the service more.

That's diametrically opposite to the Netflix (Nasdaq: NFLX) situation. The digital video expert loves to see millions of fresh customers coming in, because its content licenses are paid in advance and the overhead costs are microscopic. So once Netflix hits the breakeven point where marketing, management, and other operating costs have been paid, every extra set of eyeballs translates into about $8 of pure profit per month. More users is always a good thing under that model.

Pandora's equation is very different because the company pays license fees based on actual usage -- and they're exorbitant and rising. With the free service so readily available, the paid subscription plan doesn't have much traction. So how does Pandora fix that problem? The only workable solution is to score better advertising contracts, because simply increasing the volume of ads is likely to drive annoyed users away to Spotify, or Rdio, or iHeartRadio, or... well, you get the drift.

What needs to work?
CEO Joe Kennedy says that ad sales are doing well but it'll take another year or so before ad revenues start rising faster than content costs. So that would mark the absolute bottom for negative cash flows and bottom-line losses. After that, it'll take several quarters to get back to breakeven again -- assuming that the ad sales stay on track and nobody disrupts Pandora's service in the meantime.

Given all these factors, I wouldn't touch Pandora today. The company has about a year available to prove that this quarter's surprise success has legs, and I'd be shocked if the stock didn't plunge a couple of times while we wait. If the story turns positive, we should see a couple of juicy buy-in windows opening up before the stock really takes off.

And then again, perhaps it never will. Pandora is a lot like Facebook (Nasdaq: FB) and Groupon (Nasdaq: GRPN). This is the new breed of unproven business models that could turn ugly in a hurry. Are we watching a miniature dot-com bubble in action? Hard to say until it either pops or matures, but at least investors are being mindful of that possibility this time around. Facebook and Pandora rely on growing ad sales and Groupon is a massive advertising racket if you ask me. Show me the money, guys.

Until they do, I've already rated Facebook and Groupon as underperformers in our CAPS system. Pandora joins the bearish CAPScall crew today because I'm not at all convinced that this song will play out the way management has planned it. David Gardner is excited about a handful of potential multibaggers right now, but Pandora ain't among them.