Is AT&T's New Tiered Data Plan Enough for Mobile Enterprises?
Why? Put data-guzzling mobile devices such as Apple's iPad together with an unlimited data plan for $30 a month and it's a recipe for disaster. What you'll inevitably get is a data network that is brought to its knees by congestion. Try to push an unlimited amount of data though a network with a finite capacity and you don't have to be a rocket scientist to realize that something's got to give.
It's precisely because AT&T's network -- like those of all other carriers -- have limited bandwidth that data capping is vital. Capping is a crude tool in that it doesn't guarantee that networks will run quickly at any given time, but it does help a bit. It certainly prevents individuals downloading 31GB over 48 hours using the cellular network, as Zak Epstein did last month to torture test his new iPad 3G. We can only imagine the annoyance of his neighbors, whose data throughput likely slowed to a crawl.
Now picture this: You've just finished a PowerPoint presentation, you need to send it from your laptop using your phone's data connection, and you need to send it quickly. That's going to get ever harder, as more and more Zak Epsteins connect data-guzzling mobile devices to your carrier's network to download movies, share music, and generally use the network in ways that it was never originally intended.
"Enterprise users are being impacted, but they are precisely the customers who need fast data access and who are providing the carriers with most of their revenue," says Ricky Watts, CTO of the network planning and optimization firm Aircom. "At the moment, all users, including enterprise users, are getting treated as equals. Going forward, I think this will change as carriers see the value of enterprise customers."
How will it change? In the world of fixed-line DSL connections, businesses pay a premium for a service with a lower contention ratio. That means that instead of sharing the available bandwidth with 50 other bit-torrenting, video-downloading, network-torturing consumers, businesses share with 5 other businesses. Because of that they are more likely to get all the bandwidth they need at any given time, and certainly never less than a fifth of the total that's available.
Unfortunately you can't do this type of contention ratio fixing on mobile networks. Still, if enterprises are prepared to pay more for a better service, then some multi-tiered pricing structure seems appropriate. Clearly, one way to achieve this would be using a dynamic pricing system. Like off-peak tariffs for electricity usage, dynamic pricing would aim to encourage consumers to do their downloading and other non-essential data intensive activities at times when business users don't need the network and there's excess capacity available. But because we're talking mobile users here, excess capacity can pop up in different places at different times, rather than between certain set times everywhere.
"The idea is that you would send out alerts to customers in particular areas to say that for the next hour, for example, data is half price," says Watts. This type of pricing has already been implemented for voice traffic in some parts of the world, and although doing it for data would be more complicated, it could certainly be achieved.
A more radical, multi-tiered pricing structure would involve prioritizing traffic for enterprise customers who, naturally, pay more than consumers for this privilege. When an enterprise customer tries to send that PowerPoint presentation, the network would recognize that the user is on a premium tariff and ensure that they receive a certain minimum network throughput.
To achieve this on congested cells it would be necessary to restrict the bandwidth available to non-premium users, with the result that consumers might notice their downloads slowing while the PowerPoint presentation went over the network. Of course the prioritizing could be even more sophisticated than that, giving enterprise users' email traffic high priority but giving their web traffic only medium priority, and so on.
Watts foresees that enterprise users may even be able to prioritize their network usage on the fly. "When you start to transfer that PowerPoint file, you could get a message on your phone asking whether you would like to send it in one minute for $5, or over three minutes for $2, or using the network's best efforts for free, with an estimated transfer time of twenty minutes," he says. It's a seductive idea.
Unfortunately, none of these sorts of sophisticated multi-tiered pricing systems is likely to come along any time very soon. So perhaps the best way to increase data speeds might be to significantly increase the capacity of the carrier networks? Well, that's what the carriers are doing. A bit, anyway.The problem is that they are reluctant to invest much more in their legacy 3G networks when these will be made redundant in a few years by the higher capacity 4G networks that they are also funding.
4G networks offer the promise of far more bandwidth than is available today, but will it be too little too late? With the explosion in mobile data usage that's currently under way, chances are that by the time they are ubiquitous the 4G networks will be just as congested -- and slow -- as the 3G ones they are meant to replace.
The only possible conclusion then is that until carriers start offering enterprise users premium services at premium prices, mobile data networks are likely to be slow for everyone for some time to come. And with iPads and other data-intensive devices becoming increasingly common, the inconsistent performance of wireless networks is likely to remain the status quo.
The message to carriers is clear: First, don't even think about unlimited data tariffs on your network anymore, and, second, bring in premium rate data services for enterprise customers who need better performance and are willing to pay for it. Let's hope they get that message soon.