Consultant Martin Geddes has for some time argued that telcos are in the trading platform business, not the “pipe” business.

Basically, Geddes argues for a future business model based on customized, on demand services and features that are not based on “speed” and “retail price,” but on how use of bandwidth supports the applications that now are what people will be buying.

And that is a dramatically different way of looking at a telco’s product, from the standpoint of what the customer buys. “You don't want to lease a circuit for a year,” says Geddes. “Instead you create an association of local and remote resources for a minute.”

That is sort of “bandwidth on demand” on steroids, closer to an “capabilities on demand” approach than anything else.

The advent of cloud-based computing will drive the change, whether service providers like it or not. “You don't want to spend $10 million setting up an MVNO, but rather one cent to establish a transient local computation resource from wholesale assets,” Geddes says.

Customers still will pay for bandwidth, but they increasingly will pay for bandwidth that supports the applications they want to use, when and where they want to use those apps, on whatever devices they prefer to use, with whatever experience quality they deem preferable at that moment.

Granted, Geddes himself says he tends to be about a decade “too early” with his predictions, but the general direction is rather clear. The purpose of a network used to be to support the business model of the network owner.

Cable TV and satellite TV networks were optimized for delivery of multiple channels of entertainment video, while broadcast TV networks and radio networks were optimized for single channels of content delivery.

Telephone networks were optimized for symmetrical bandwidth voice calls, as were mobile networks. Each network tended to be optimized for one particular media type.

These days, all networks are, in principle, capable of carrying any type of media, though network architecture now is paramount. Broadcast networks still are inefficient for two-way traffic of any magnitude.

The more important strategic change, though, is the shift to Internet Protocol, which separated that historic bundling of “owned network and owned apps.” That creates, for the first time, a disaggregated or separated application and access business. That is what “over the top” means.

That is both a problem and an opportunity. There are problems, to be sure.

“My educated guess is that we will see traditional telcos have a reduced role,” says Geddes. Access providers are trucking companies, he argues, and application providers are digital logistic platforms. That implies less power for access providers and less direct relationships with customers.

The opportunities, if less well assured, also exist, though. “Rational network pricing can be re-established” if use of the network is directly related to the demands application data imposes on the network.

“Operators should then consider tiering their offering by quality, and not just quantity, at different price points,” says Geddes. “Users are not homogenous in their demands or ability to pay, but we treat them as if they were.”

That’s what Geddes means by the business becoming one of trading platforms, not pipes. The use of the network has differential value for different users and providers, depending on value of the apps, time of day, amount of congestion and so forth. 

Network operators will make money providing the type of network support each app requires, or each app provider or end user wants to pay for.