Emerging Technology: Skype and Hype

If its Internet phone service is free, how does it make money?

By Steven Johnson|Monday, May 29, 2006
RELATED TAGS: COMPUTERS

Earlier this year, at the annual World Economic Forum meeting in Davos, Switzerland, several titans of industry gathered for a panel discussion on the future of digital technology. The group included Microsoft cofounder Bill Gates, whose estimated personal wealth is greater than $46.5 billion, and Niklas Zennström, the upstart CEO of the Internet-based phone service Skype Technologies, who recently sold his three-year-old company to eBay for $2.6 billion.

Moderator Geoffrey Moore asked how many customers Skype had attracted in its short lifetime, and Zennström prompted a murmur of amazement in the audience when he replied, "Seventy-five million." Moore then turned to Gates and said, "Seventy-five million in three years—that's faster than even you managed, Bill."

Gates chuckled and said, "How many are paying customers?"

Zennström hemmed and hawed before offering the answer: "A single-digit percentage." From the way he said it, one couldn't help but suspect that the digit in question was a 1.

Gates's remark brought to mind a host of high-flying Internet businesses from a few years back whose would-be tech-titan executives made overnight fortunes by selling stock in companies that gave away services online with little regard to balancing the books short term. Most of those fortunes vanished, along with the companies themselves, when the Internet bubble burst in 2000. But Gates and Microsoft thrived by following a more traditional maxim of profitability: If you have to give a product or service away free, something must be fundamentally wrong with the business.

Skype's dearth of paying customers is easy to scoff at, but the foundation of the company is far more solid—and interesting—than the irrational exuberance that fueled Internet enterprises in the 1990s. Skype is built on the same sort of peer-to-peer networking technology that powered many notorious—and, depending on whom you talk to, illegal—file-sharing services, starting with Napster. In fact, Zennström's previous software effort was a wildly popular peer-to-peer application called KaZaA, which has been used to share millions of files over the past few years. (It has also triggered more than a few lawsuits.)

Zennström's vision with Skype was to take the power of peer-to-peer networking and apply it to a field without significant intellectual property issues. Instead of using his software to download free songs, people use it to place a voice telephone call to someone on the other side of the planet—free. This is the kind of thing that keeps phone company execs awake at night, but there is no question of intellectual property being stolen.

From the users' perspective, Skype is refreshingly simple. You download software from the Skype servers, and assuming you have a headset or a microphone and speakers attached to your computer, you can immediately start calling people. If you're calling other Skype users—no matter where they happen to be—the calls are free. It's exactly the same rate to dial your next-door neighbor that it is to dial someone 10,000 miles away: zero cents per minute.

Cost enters the picture when a user wants to call someone with a traditional phone number, for either a landline or a cell phone. The calls are billed at rates that tend to be cheaper than the rates for traditional service, although they vary depending on the country being called. Skype users can also pay to establish a "SkypeIn" number that can be dialed directly from traditional phones, which means that if you switch to Skype as your permanent number, you don't have to teach your grandparents how to install the Skype software on their computer—assuming they have a computer. Right now, Skype charges $50 a year for a SkypeIn number.

The only significant downside of Skype is the audio quality. Because Skype sounds travel over the Internet, they generally involve a small but perceptible time lag, which can have a disconcerting effect on a conversation if you're not used to it—as though the person you're talking with is emotionally distant or confused. In terms of pure audio quality, Skype-to-Skype direct calls can sometimes sound better than landline calls and far better than the noisy compressions of many mobile conversations. But a low-bandwidth connection or Net congestion can reduce the sound to the fringes of audibility.

Still, Skype is an appealing package, and you can understand why 75 million people have signed up for the service in such a short amount of time. So how does the company pull it off? How can it afford to connect 75 million users when only a single-digit percentage of those users are actually paying?

The answer points to the fundamental difference between the peer-to-peer paradigm and the pay-to-play paradigm. Bill Gates got rich selling people large, complicated software programs that cost a great deal to code and support. All of those costs were borne by Microsoft itself, which meant that Gates had to charge his customers for each copy of Windows or Office. Gates had an edge over the industrial barons who preceded him in that once a program had been written, it was a relatively trivial and inexpensive matter to reproduce the code millions of times. But those manufacturing and distribution costs—however minor—still had to be carried by Microsoft.

In the peer-to-peer paradigm, much of the labor is performed by the customers—who also create a huge amount of the infrastructure required to make the product work. Skype's voice network piggybacks on both the existing infrastructure of the Internet itself and the computation power of each customer's computer. When you sign in to your Skype account, your computer is not just making calls for you; by storing directories of Skype users and routing calls from one user to another, it's also making the overall Skype network more efficient. Instead of dialing into a massive bank of servers at a central location and having your call directed from there, Skype takes your request and disperses it throughout the overall system of connecting users, relying on the combined bandwidth and processing power of all those networked machines to make the system work. Every time Skype gets a new customer, even a nonpaying one, the overall network becomes more efficient.

All of which means that Skype's overhead is so low that the company seems positioned to make a healthy profit, despite the fact that it's giving away the vast majority of its services. According to eBay's statements at the time of the acquisition, Skype is expected to turn a profit by the end of 2006. The "we'll be profitable two quarters from now" refrain was practically a national anthem during the Internet bubble years, so caution is advisable. But I suspect the $2.6 billion that eBay paid for Skype will turn out to be a bargain.

Although Skype is based on a radically different business model than Microsoft, it has an important philosophical connection to the Microsoft of the 21st century: Google. At Davos, Zennström remarked that Google probably has a comparably low percentage of paying customers because the company makes most of its money from users clicking directly on ads. Ask yourself: How many of your Google searches result in your clicking on an ad? It might well be a fraction of a single-digit percentage. And yet Google is a financial powerhouse. How can this be?

Google now offers dozens of different tools and services, but the primary benefit it gives consumers is a search engine with an uncanny ability to organize the World Wide Web so that it's actually useful to explore. How did the company solve this problem? By outsourcing it to the entire universe of people online. Instead of hiring a thousand librarians to read Web pages and categorize them, Google's software looks for patterns in the links that ordinary people encode into Web pages. If a given page receives an unusual number of links from other sites, Google interprets those links as endorsements and promotes the page in its overall rankings. That system—which Google's founders call PageRank—is the secret sauce that made the company an order of magnitude more effective than its competitors. But Google doesn't have to pay people to create those links, just as Skype doesn't have to pay people to build out its telephone network. In both cases, they do it free.

Comment on this article
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

ADVERTISEMENT
ADVERTISEMENT
Collapse bottom bar
DSC-JanFeb15
+

Log in to your account

X
Email address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it emailed to you.

Not registered yet?

Register now for FREE. It takes only a few seconds to complete. Register now »