Access, Not WAN, is Bandwidth Issue

As always, the bandwidth choke point remains the access network. Local area network bandwidth is easy to upgrade. Long haul bandwidth also is not much of an issue, on most routes. The issue, for any service provider except a cable company or satellite company, is that replacments for linear TV will have to be delivered over the IP pipe, not a special purpose video delivery network. Multicast or broadcast networks are the best way to deliver linear TV. IP networks are a better way to deliver any sort of symmetrical service, but grossly inefficient for streaming content on a unicast basis. Access remains the choke point.

iPhone, at&t Deal Still is Significant

Big changes take time in the global communications industry, for good and not so good reasons. Still, it's important to recognize that Apple has succeeded where others have failed. It has wrested some significant degree of control away from normally powerful wireless carriers. But not in the way many disrupters prefer. As Apple tends to do, it demanded a great degree of creative control because that's the only way to ensure the sort of user experience it wants to provide. And the iPhone is no exception.

at&t agreed to leave its brand off the body of the phone. at&t also abandoned its usual insistence that phone makers carry its software for Web surfing, ringtones and other services. The deal also calls for Cingular to share with Apple a portion of the monthly revenues from subscribers, says the Wall Street Journal.

In another break with standard practice, the iPhone will have an exclusive retail network: The partners are making it available only through Cingular and Apple stores, as well as both companies' Web sites. The deal points the way to how service providers can balance control and freedom, vertical integration and horizontal innovation.

Cingular executives were willing to cede control to Mr. Jobs for the privilege of being the exclusive U.S. provider of one of the most highly anticipated consumer electronics devices in years, and to deny rivals a chance to do the same. Considering the number of music-capable mobile phones now being sold, a wise move.

Not many companies will have the bargaining power to do the same, but it will be better for mobile providers if they can learn to do this on a wider scale. Some things have to be controlled, for technical reasons. Some things have to be controlled for user experience reasons. But not everything has to--or should be--controlled by the carrier.

There's Always an Elephant in the Room

That's not a problem if it is your elephant. Vodafone CEO Arun Sarin warns his mobile compatriots that “as an industry it takes us a long time to get things done; we need to move faster otherwise others will eat our lunch." Forgetting for the moment about Microsoft, Google or the likes of Skype, what about WiMAX, which sometimes is seen as a comeptitive business platform, and sometimes just a platform?

"WiMAX is now a serious contender for mobile broadband," he thinks, positioning WiMAX as a platform any mobile provider can adopt, as Sprint Nextel already has done. Others might see WiMAX as a competing business platform that will be used by attackers to assault the legacy mobile providers, whatever technology they may use for fourth generation networks.

Nobody seems to disagree about the need, though. “If we don’t build our broadband networks we will have this opportunity taken away from us,” Sarin says. For some time Vodafone has distanced itself from the "3G or WiMAX" debate, in part because it seems to be thinking it will have to adopt WiMAX as its fourth generation network standard.

Vodafone subsidiary SFR in France may already have begun WiMAX testing. Vodafone network partner MTC-Vodafone recently won a license for WiMAX spectrum in Bahrain. The company also is expected to bid on WiMAX licenses to be auctioned in Egypt and Saudi Arabia in the coming months.

Vodafone acquired a WiMAX license in Greece last year, and is deploying a network in Malta. The company also has been testing WiMAX in New Zealand, and in South Africa.

Sarin notes that less than 10 percent of Vodafone revenues are derived from 3G services. The implication, it seems to us, is that 3G has proven to be not much more than a "wideband" niche between narrowband and broadband. Sort of like the analogy of narrowband voice (64 kbps); basic rate ISDN (up to 128 kbps) and "true" broadband in the megabit per second range.

So if legacy mobile carriers adopt WiMAX, how much sense does it make to track WiMAX deployment separately from wireless provider 4G? Really, the issue is not WiMAX. The issue is whether WiMAX can provide the platform for a competitive challenge to dominant wireless providers, because the early notion had been that WiMAX would open up another access alternative. But it won't if the dominant service providers simply offer it themselves.

There's an elephant in the room. The question is, whose elephant is it?

Global Capex Up, Video Drives It

Overall, North American telecom companies are projected to spend $70 billion on new infrastructure this year, down from the $110 billion they spent in 2000, up 67 percent from their 2003, says Infonetics Research. Global spending on new telecom gear is expected to rise to $240 billion in 2008, up 19 percent from 2005. Video is part of the reason.

While sending 100,000 emails costs a telecom company around 20 cents, transmitting 100,000 low-resolution videos costs around $15, and sending 100,000 high-definition movies costs around $10,800, says Infonetics Research.

Compared to Vonage...

Rightly or wrongly, Vonage's key marketing metrics get compared to those of cable or telco companies. Vonage's cost of acquiring a customer, its churn rate and average revenue per customer each month always are pointed to as issues. Well, compared to the best run telcos, that's true. Consider Telus, which operates largely in Western Canada. It has average revenue per unit of $64.50. Vonage is in the twenties. It has churn of 1.33 percent a month. Vonage's churn is double that. Acquisition cost is $436 a customr, which in about what Vonage spends. The issue is that Vonage's ARPU is less than half of what Telus gets, and Vonage keeps a customer half as long as Telus.

On the other hand, Vonage's costs of doing business are lower as well. Also, keep in mind that Telus is, by some measures, among the best-performing service providers in the world, and its fourth quarter 2006 results confirm that. On average, Telus keeps a customer more than six years. It is aided mightly in that regard by its wireless performance. So far, Telus and other leading wireless providers are turning in the best churn performance for virtually any consumer application one can think of.

Directionally Positive

North American fiber to home numbers still are small, but the direction is positive, says the Fiber to Home Council. The markets don't like the magnitude of investment, but telcos don't have much choice but to bet the farm.

Managerial prowess will prove important, but this bet will pay off. If IP-delivered video really does take off, in both the enterprise and consumer spaces, private networks have to be part of the solution.

And private networks will result in more revenue for access providers. To the extend that public networks are involved, it isn't clear that "all you can eat" plans can survive. Users will have to pay more for consumed bandwidth. That doesn't mean current pricing metrics necessarily are affected. But applications that chew up 1,000 times more bandwidth than voice will have to be provided at retail prices that reflect those levels of consumption. New networks will be built, no doubt.

Serious Moolah

The typical Fortune 500 company spends 3.6 percent of revenues on network services, gear and assets. Last year, more than $327 billion. The largest recurring cost is wide area network transport costs. Any wonder why enterprises are looking to switch that function to Ethernet?