Why does AT&T Fear OTT?

OTT stands for “over the top”. But in the telecom sense it does not mean outrageous, it means providing a service using the data network that competes with some service that the carrier offers and uses as a revenue stream. The most obvious of these is running Skype and so not making a voice call but perhaps the biggest threat is messaging services. After all, young people rarely use their phones for making calls anyway, and even old fogies like me send more and more text messages.

In fact there are over 6 trillion text messages sent per year, according to the ITU (2010 data). That’s 17 billion per day, or 2½ for every person on the planet (including babies and other people who don’t have phones). Retrevo in the UK reported that 10% of young people in Britain think it is OK to send a text message while having sex.

If you have an iPhone and you send a “text” to another iPhone then it will show up in blue not green. This means that in fact you didn’t send a text at all. Apple simply bypassed the carrier and used the data service to transfer whatever you typed. Except to the extent that this uses a miniscule amount of your data cap (which is measured in gigabytes per month) the carrier gets nothing. For a text they get several cents depending on what sort of plan you are on, more if it crosses an international frontier. In the same way, if you use Whatsapp or similar messaging services (Skype, AIM, Blackberry Messenger etc) the carrier is cut out of the loop. It is their worst nightmare, being reduced to a dumb pipe.

Texting is huge. And hugely profitable. How big? Texting is as big as handsets. That’s right, carriers make as much revenue selling text messaging services as they do selling handsets. They make about a quarter of their revenue on the hardware side. Of the remaining $1T in service revenue, 85% comes from voice and text messaging.

But texts require essentially no bandwidth and so have minimal cost. As a result they are very profitable. How profitable: 45% of carrier profits (another 50% comes from voice calls and just 5% from everything else). This is perhaps a bit distorted since handsets are sold at a loss in some countries such as the US, and it is made up on the service revenue, but the bottom line is that nearly half the profits of the network operators (AT&T, Verizon, Sprint, Vodaphone, Orange, China Mobile etc) comes from text messaging.

Here is an interviewfrom a couple of weeks ago with Randal Stephenson, the CEO of AT&T in which he says:“You lie awake at night worrying about what is that which will disrupt your business model. Apple iMessage is a classic example. If you’re using iMessage, you’re not using one of our messaging services, right? That’s disruptive to our messaging revenue stream.”

To make it worse, it turns out that the heaviest users (young people, people who travel a lot, people communicating internationally where text messaging has a premium price) are the most likely to use these services that bypass the carrier. A similar phenomenon happens with voice. Carriers make a lot of their voice profits on international calls, since the rates are outrageous and they don’t even pretend to compete on price. But those are exactly the calls most likely to be bypassed by users switching to Skype instead and creating no revenue at all (for anyone, not even Microsoft who owns Skype these days).

So if people are making fewer voice calls and texting more, and more and more of the texting is bypassing the carriers then how does the system hang together. With next generation LTE networks, there won’t really much of a technical distinction between the carrier’s voice and Skype, or the carrier’s text message and Apple’s iMessage. They are all carried over the same packet switched infrastructure. There will no longer be a separate voice channel and a separate channel for text messages (which were originally only intended for network administration). It is just the billing that is different. The carriers are desperate to find a way to make more money than just providing a data service that smart hardware uses to deliver premium services such as voice in which the carrier doesn’t get a cut.

Since the cost of all the network infrastructure (base stations, backhaul, billing etc) has to be paid for out of carrier revenue, expect data rates to go up a lot and voice/text rates to come down. At least that’s my prediction. And they had better or the carriers won’t be able to afford to (indirectly) buy all those chips that go into smartphones and the base stations to serve them. Or alternatively the carriers will successfully block certain services such as Skype (indeed one carrier in Sweden already excludes use of Skype from their fixed price data bundle).