By Kevin Kelleher, contributor
In 1995, Netscape went public. Underwriters believed the stock was worth $14 a share but demand was so strong they doubled the offering price overnight. It rocketed to $75 a share on its first day. But Netscape was a browser maker, with no clear plan to make money. In time, it was bought by AOL, which in turn was bought by Time-Warner (twx) (parent company of Time Inc. and Fortune). And it all ended in tears.
Two years later, Amazon.com (amzn) went public. The online bookseller had a loss of $6 million and went public valued at $300 million. People thought it was another crazy IPO: One analyst likened it to a hallucinogen, quipping, "Some people smoke Internet inhalant and their judgment gets bizarre." But time proved their judgment sound. Amazon's stock has increased more than 100 times over its IPO price.
Both IPOs are credited with launching what became the dot-com boom. But amid warnings that companies like Facebook, Twitter and Zynga are about to launch another crazy boom, it's worth recalling a often-overlooked lesson from the 90s: For every bomb like Netscape, there is an Amazon, a true giant in the making.<!-- more -->
So it is with a new generation of companies. For every OpenTable (open), which is up 344% from its IPO debut last year, there is a Mitel Networks (mitl), the IP-telephony company whose stock has dropped 61% since its initial public offering last year. And the class of 2011 isn't likely to be much different. There will be winners, there will be losers.
But unlike a year ago, many are willing to write off most web startups as actors in a drama re-enacting the dot-com bubble. Facebook is valued at 25 times its revenue, and Twitter at 100 times. (Amazon, by contrast, is trading at two times revenue.)
And now Pandora is going public without having made a cent in net profit. All we need is the Pets.com sock puppet to rise from the dead and the present will have come full circle to a past we'd all rather not remember.
But Pandora is neither Netscape (a web company with no plan for turning a profit) nor Amazon (a startup masterfully dominating the online business it pioneered). It's a typical IPO candidate – that is, a company that has yet to post a profit but nevertheless offers a reasonable promise of profit growth in the future.
Pandora is actually a good example of how the tech IPO market should be working in a time when people are worried about bubbles. It's raising money because it needs it, to pay for licensing fees and make acquisitions, and while it hasn't posted a net profit, it's operating costs are coming down fast enough to make it a reasonable expectation that it will soon do so.
Pandoras operating loss shrank to $15.5 million in the fiscal year through January 2010, down from a $27.4 million loss a year earlier. In the first nine months of 2010, Pandora posted an operating profit of $819,000 as revenue tripled to $78 million.
So while Pandora posted a net loss in the first nine months of last year – due to interest expenses and the effect of recognizing stock warrants at fair value – it's profitable at the operating level, having shown a steady pattern of increasing revenue while holding operating costs down.
Content acquisition, or the licensing fees Pandora pays, fell to 50% of revenue in the nine-month period from 72% a year earlier. Marketing and administrative costs fell to 35% of revenue from 53%. Cost of revenue, largely the costs of adding and maintaining networking infrastructure, fell to 9% from 19%.
There are certainly risks. Pandora's prospectus lists 25 pages worth of risks facing investors, some of it more than the usual legal boilerplate. Streaming music online is a nascent business, with stiff competition and the presence of greedy, inflexible music labels who are afraid of new technologies. Licensing costs will always eat up a good part of Pandora's profits, regardless of whether ad revenue is rising or falling.
But those risks aren't unusual for a company going public. In fact, Pandora in 2011 looks a lot like Netflix (nflx) nearly a decade ago. Netflix went up against established video-rental chains with a DVD-by-mail model that was easy for others to copy. Since then, Netflix has pioneered the nascent industry of streaming video.
Netflix has succeeded because of smart management and a service that is easy to use and builds user loyalty. Pandora has been following that formula, with users listening to 13 hours a week to its music. The company claims to control more than half the market for Internet radio stations.
Of course, Netflix has always had its share of doubters, just as Pandora's IPO is being greeted by skepticism. But Netflix ended up being more Amazon than Netscape. Now that we can expect to see more web companies go public in coming months, it's important to know how to tell the companies with more buzz than business plan from the ones that have a decent shot at being good investments.
More from Fortune: