Your browser is not supported. Please update it.

4 March 2008

Covad gets into metrozones, trying to succeed where Earthlink failed

US-based ISPs have been struggling to adapt to broadband and fixed/mobile trends while avoiding having to pay for access using the major carriers’ lines. Earthlink was in the vanguard of the attempt to use alternative, largely wireless networks to balance the loss of its traditional dial-up business without relying on Verizon and AT&T, but its ambitious strategy based on Wi-Fi metrozones, WiMAX and an MVNO crashed and burned. Now business-focused ISP Covad Communications is testing whether a similar approach will work better in the enterprise market than it did for Earthlink among consumers. It has been experimenting with WiMAX and other wireless technologies for several years, directly or through partnerships, and is now potentially adding Wi-Fi metrozones to the mix too. It is testing a network in San Carlos, California, part of the ailing Silicon Valley Network plan to blanket the 1,500-square mile hi-tech region in Wi-Fi.

Its move will give hope to the backers of the network, which had looked in serious danger of going the way of Earthlink’s municipal WLan activities and many other failing or cancelled metrozone projects. The free, ad-supported access model, heavily supported by Google, has been largely discredited, but there are still some Wi-Fi zones that look more viable, being based on guaranteed revenue, usually from supporting services for a local authority or public safety agency; on traditional subscription revenues, where quality of service can be promised; or potentially on providing services where there is very little broadband alternative, mainly in highly rural regions. Unlike underserved areas of cities, which tend to be low income and so unattractive to a paid-for or ad-supported profit model, these rural districts often house affluent people who will pay a premium for broadband. If there are small and medium enterprises in these districts, the risk is even lower, and for an operator like Covad, they fall into its existing business target base.

Of course, Silicon Valley hardly falls into this category, but in hinting that it might rescue the plan, Covad would gain a good testbed with an established equipment partner, Cisco, and a demanding user base. It could then apply this experience to more fruitful customer bases, supporting a wholesale model for local ISPs plus business services. If the results of the 700MHz auction prove less favorable to rural operators than was originally hoped, there may be a real gap for Wi-Fi mesh in some districts, and this could complement Covad’s other wireless moves.

These have mainly focused on broadband wireless. In 2006 the operator partnered with LMDS spectrum holder NextLink Wireless, part of XO Communications, in order to target large enterprises, adding to its earlier acquisition of pre-WiMAX operator NextWeb, for a push into the SMB space. Like Earthlink on the consumer front, Covad aimed to build a patchwork of services using different wireless technologies, to balance pressure on its main business, wholesaling DSL connections to companies like Earthlink itself and to enterprises. This model has been under pressure because of changes in the US law, removing the requirement on incumbent telcos to share their DSL networks with third parties at regulated rates. To compensate, Covad launched VoIP services in 2005 via GoBeam, which it acquired earlier in 2004, and then engaged in WiMAX trials and the NextWeb purchase.

Covad is itself changing ownership, with its shareholders this week agreeing to a $417m buyout offer from private equity firm Platinum Equity.