In Search of the Breakthrough Mobile Network

The past year may have been a wireless dream come true if you're a fan of the iPhone and more open handsets and networks, but it was a nightmare for some major new alternatives to traditional mobile services.

Municipal wireless networks, WiMax and MVNOs (mobile virtual network operators) began the year buoyed by high hopes before each faced a reality check. In each case, business woes at one company were just part of the reason.

The recent growth of wireless in many forms helped to generate excitement and investment for new technologies in 2006 and early 2007. Municipal wireless networks seemed like a new channel to deliver ubiquitous high-speed Internet access without waiting for traditional carriers to step up. WiMax, possibly an alternative to 3G and Wi-Fi, finally had a marquee customer. And MVNOs promised to deliver complete cellular services, including phones, targeted to specific kinds of consumers.

But as it turns out, it's not so easy to flip the wireless world upside down.

As the year began, San Francisco reached a deal with EarthLink for one of the most hotly anticipated municipal wireless networks in the world, one in which Google would provide a free citywide service in addition to EarthLink's paid offering. As in EarthLink's deal for a Philadelphia network, the city would pay nothing. Meanwhile, leaders in Silicon Valley were planning an even bigger network, spanning 1,500 square miles and multiple technologies.

Politics kept the San Francisco Board of Supervisors from approving the EarthLink plan for months, until the service provider delivered some bad news of its own. In April, EarthLink said it would focus on both the municipal networks it was already committed to and to driving up usage in large cities rather than seeking out new deals. The San Francisco project's future got blurry. Then, in August, the company said it wouldn't invest any more money in its free-to-cities business model. A shrinking dial-up business and other problems were forcing big cuts at EarthLink.

"We will not devote any new capital to the old muni Wi-Fi model that has us taking all of the risk by fronting all of the capital, then paying to buy our customers one by one," President and CEO Rolla Huff said. By November, EarthLink was considering "strategic alternatives" for the Wi-Fi business, an indication that it may be sold off. The Philadelphia network is going forward but has drawn fewer than expected subscribers. Meanwhile, backers of the Silicon Valley network were left searching for a new builder and operator after their first choice couldn't attract enough financing, even in the heart of the IT industry.

Municipal wireless turned out to have one big technology problem -- it required more access points than expected -- which worsened its business problem: Where would the money come from? Analysts say the key now is to find areas that lack other broadband alternatives or get cities to sign up for services themselves. Telscape Communications, a mobile operator focused on the U.S. Hispanic market, is negotiating to buy a municipal Wi-Fi network in Tempe, Arizona, that has 1,000 access points and only 500 subscribers. The company hopes to use better marketing and some combination of cellular and Wi-Fi voice services to make it a success.

WiMax has been promoted as Wi-Fi with a wider reach for several years, and its big opportunity came in August 2006, when Sprint Nextel chose the technology as what it called its "fourth-generation" network. Vendors including Intel, Motorola, Samsung and Nokia piled on, and it looked like the high manufacturing volume that would drive down WiMax prices was finally on its way. The network would reach 100 million U.S. residents by the end of 2008, Sprint said.

In July this year, the carrier announced it would team up with wireless data provider Clearwire to jointly fund and market the WiMax service, called Xohm, and allow roaming between the two networks. Sprint would build 65 percent of the network, Clearwire the other 35 percent, each covering some cities the other couldn't reach. However, the companies gave the same target: 100 million people by the end of 2008.

Then Sprint CEO Gary Forsee, a backer of the estimated US$5 billion WiMax plan, was forced out of the struggling carrier. A few weeks later, the Clearwire deal was off, with the companies saying they couldn't reach final agreement on its details. And Sprint said it was reviewing its WiMax plans and would say more in early 2008. As the year's end approached and new Sprint CEO Dan Hesse took over, it wasn't clear what his take on the project might be.

Sprint's soft launch of WiMax in Chicago and the Baltimore-Washington, D.C., area went ahead before the end of the year as planned, and the carrier said it still plans commercial service in the first half of 2008. But WiMax is starting to generate more interest for emerging markets than for advanced economies, including from Cisco Systems, which acquired Navini Networks in January. The company said its main focus with WiMax is to get broadband widely deployed in countries that don't have enough wired infrastructure.

The uncertainty surrounding Sprint and its plan is affecting the image more than the reality of WiMax, said IDC analyst Godfrey Chua.

"It's not problems with the technology, it's problems with the company," Chua said. "The momentum we're seeing behind WiMax in the rest of the world continues."

Mobile virtual network operators (MVNOs), which resell mobile capacity from established carriers, pulled in a lot of investment money in 2006 but suffered some high-profile failures in 2007.

At the end of July, youth-focused Amp'd Mobile shut down after it attracted nearly 200,000 customers in less than two years but failed to collect enough of the revenue it was due. In September, Disney said it would shut down its specialized service for families by the end of the year and look to offer its features through an established carrier. Helio, a money-losing joint venture of SK Telecom and EarthLink, stayed afloat thanks to $270 million of added investment from SK after EarthLink decided the MVNO game was too rich for its blood.

MVNOs and their supporters often didn't realize what an uphill battle they faced, according to analysts. They needed to build up in months the kinds of billing systems and distribution networks that traditional carriers had taken years to refine, and frequently they fell short. Meanwhile, trying to make a profit by buying minutes and bytes from another carrier and reselling them is hard even for a well-run company, analysts said. The foggy future at Sprint, which provides the underlying network for most MVNOs in the U.S., cast yet another cloud over the business.

GreatCall, a startup that launched the Jitterbug service for older cell-phone users last year, learned these lessons the hard way. The company changed billing companies twice before finding a provider that worked, said Arlene Harris, Jitterbug's founder, chairman and chief strategy officer. It also struggled to get funding, especially after other MVNOs started folding.

The service now has "tens of thousands" of subscribers and is growing, Harris said.

"Everything is working well," Harris said. "But I have to tell you, it's been very hard."