9 Comments

Summary:

In 2012, a “turbo” button will appear on mobile phones, making it the first of many new options that allow customers to customize their data plans by quality of connection, rather than megabytes consumed. We may even see the resurgence of the unlimited plan, with a catch.

Speedometer

Updated. In 2012 we’re going to see the emergence of a “turbo” button on our mobile phones. Verizon Wireless is developing a network API that would allow customers to buy temporary bursts of bandwidth on their phones, giving priority to their video streams or downloads even during the most congested network conditions. On Thursday, Leap Wireless CEO Doug Hutcheson revealed at a Citigroup conference that its mobile carrier Cricket would offer a similar bandwidth boost option.

Cricket Both operators plans to implement their its turbo buttons this year, while Verizon has not set a launch date, but they plan to position them differently. Verizon’s will be a premium service for customers looking for speeds beyond the norm, while Cricket’s will be a means for customers to reclaim their regular data speeds after customers use up their monthly gigabyte quotas and are throttled down to the equivalent of dial-up modem access. (Update: Verizon clarified to us on Monday that, contrary to earlier reports, its turbo technology is still deep in the labs, and it hasn’t settled on a timeframe for commercial release or a specific business model.) But they both accomplish the same thing: giving customers a bandwidth oomph beyond what they would get from their normal data plans.

A turbo button is an intriguing concept. Most of the data services I use on my phone – Twitter, e-mail, web browsing, Pandora — don’t require a stellar mobile broadband connection, but for some apps the network’s best effort isn’t good enough. If there is a video I’m aching to see or a big download I want to speed up, having the option to pay a few cents to a dollar for instant gratification might be worth it.

What about the other guy?

There’s a trade off. The wireless network is a shared resources, so every packet of mine that gets prioritized means someone else’s packet is getting de-prioritized. If you’re in a congested cell where a bunch of other people have temporarily turbo-charged their phones, your network connection has to make up the difference, sticking you with even slower speeds than normal.

These kind of data management policies are starting to make their way in to mobile networks and they could result in a new order of data plan tiers, based on quality of connection rather than gigabyte tonnage consumed.  Tekelec, which develops the policy management technology behind Verizon’s turbo button, believes we’re going to see a lot more ways to slice and dice data in the near future. In a recent interview with GigaOM, Tekelec CTO and VP of engineering Doug Suriano said we can expect to see operators start offering “upside down” data plans, which will allow customers to customize their plans – and the prices they pay – based on the types of applications they use and the times they typically use them.

The next type of pricing policy we’re likely to see, Suirano said, will be the opposite of a turbo button: the equivalent of a network brake. Rather than pay more money to  get temporarily faster speeds, customers may be willing to pay less money to temporarily suffer sluggish bandwidth. Customers often would enjoy the same robust speeds as their full-fare counterparts, but when the network gets congested their data traffic would be the first be deprioritized, clearing the capacity fast lanes for premium turbo subscribers, Suriano said. He used Verizon as a hypothetical example for such a service:

Verizon Wireless has a reputation for providing a quality service to all customers, so for any premium customer there will be mechanisms in the phone to ensure that you always have a ‘Verizon-quality’ experience. But if you want to pay $10 to $20 less a month for data, Verizon could offer you a service that’s good for most of the things you want to do with your phone, but doesn’t always guarantee you that Verizon level of quality.

Could prioritization be the answer to data caps?

What’s particularly interesting about such a scenario is it might revive what a practically extinct concept in mobile: the unlimited plan. While there is a set cost for to delivering any megabyte of data to a customer, the really expensive thing for an operator is to deliver a megabyte each to hundreds of customers in the same cell. It’s a matter of network scale and load balancing. If you’re wandering around the streets of your city’s financial district at 4 AM, the network is wide open, its capacity lying dormant. But at Noon that same network is overloaded.

If an operator had a way to ensure that customers’ speeds were downgraded only when other customers were vying for that same capacity, then it could open up the data spigots, offering unlimited, though restricted, access — or at least a whole bunch of gigabytes for cheap. But such a policy might produce its own problem of scale. If everyone is accessing the network all of the time, then the network is always congested and everyone’s traffic — except for that of a few premium customers — gets de-prioritized to oblivion. To make it work, operators would need to strike a delicate balance between the network capacity they can feasibly build and the way their customers access it.

Of course, any time you start talking about prioritizing certain types of traffic over others, controversy ensues. Wireless carriers may not be bound by the same net neutrality restrictions their wireline counterparts face, but that doesn’t mean they aren’t vulnerable to the ire of their customers. Preferential treatment of one person’s packets over another’s could create a stratified mobile Internet, with an upper class that’s willing to pay to access what they want, when they want and lower class left with the network’s scraps. Depending on how operators implement these new pricing policies, they could spark a revolution in the pricing and availability of mobile data or they could create data apartheid.

Speedometer image courtesy of Flickr user Nathan E Photography
Yield image courtesy of Flickr user Seth J
Buffet image courtesy of Flickr user Wesley Fryer

  1. There is another service that will soon be showing up to improve service — except it will be free to users. It is called the toll-free app (FreeBand) and works like a 1-800 voice call where a third party pays for the bandwidth consumed. So the carrier gets a new revenue stream and the user free bytes. And no one gets slowed down in the process (it’s just a subsidy — no throttling up or down). For more info visit Box Top Solutions at CES in Las Vegas this week. Thanks!

    Share
  2. So, I’ve been annoyed in the past with mobile network performance, so I wrote an app to measure it, in terms of real world speed, over time, in different locations: free – enjoy.

    https://market.android.com/details?id=com.spdreality.SpeedReality

    Share
  3. A lot of this is very fine in principle, but unworkable wishful thinking in reality. There are hundreds of ‘gotchas’ for all of this. What happens when lots of people all hit Turbo during congestion? It can’t magically generate more capacity, so the speed uplift will be minimal and the carrier will need to process refunds & complaints. More generally, how does the customer know they got what they paid for?

    The per-app charging is even worse. Unless there is specific collaboration with the app/content co the network will make too many mistakes trying to guess what’s what.

    It’s telling that VZW says this is still ‘deep in the labs’ – they’ll have to go through multiple cycles of testing & problem-fixing, and then deal with the fact the mobile data goalposts keep moving at the same time.

    Also none of this fixes the problem of signalling load

    Dean Bubley
    @disruptivedean

    Share
    1. Kevin Fitchard Monday, January 9, 2012

      Great points, Dean. And herein lies the problem of policy management (or at least trying to service plans around policies). With a shared network you can’t maintain a consistent baseline from which to base a premium service.

      I’m curious, do you think operators will ever try to break out signaling from data consumption, charge for it separately or even charge the application developer?

      Share
    2. “The per-app charging is even worse. Unless there is specific collaboration with the app/content co the network will make too many mistakes trying to guess what’s what.” — We totally agree that doing any byte tracking in the network (DPI) is very tough to do well for any large number of data sources (herding billions of cats). But if you move the tracking into the app itself (which reports to a billing server periodically — not real time) it is secure, accurate, and easy to implement for the carrier. Even better, the apps are not made by the carrier, but by the developers using a carrier supplied toolkit (e.g. modified Android ADK) who want to push content, so even less deployment costs to the carrier and in turn lower costs / greater choice to their customers.

      Share
      1. Hey Thomas, we’re already starting to see this overseas. Operators aren’t charging per app, but excepting certain apps from the cap. Orange is selling a Facebook phone in some countries which gives you unlimited Facebook but a bucket of MBs for all other Internet use. Would be interesting to see if they try that in the U.S.

        Share
  4. In the Philippines, most ISPs and providers do this – all unlimited, but plans are based on speed. You pay this, your maximum speed is 1mbps, you pay this and it gets bumped to 2, and so on and so forth. Yeah, the speeds are pathetic, but we’re not really that well off…

    Share
  5. Unless you can schedule traffic (which you can’t), this is just robbing from one user to give to the next. It will probably make money for the carriers, but result in lower quality for users.

    Share
    1. Hi Roman, yep I agree. The tricky thing is establishing a baseline of what is a “standard” service. All mobile data — with the exception of a few carrier apps — is all best effort today, which can be good (such as when you get a 20 Mbps connection on LTE) or bad (when you get 100 kbps on a congested network). By suddenly creating a “premium” category, you aren’t boosting someone’s speed. Rather you’re downgrading someone else’s speed to ensure that the premium customer always gets speeds in the upper range of their typical service. I just don’t the see that concept gelling with consumers.

      Share

Comments have been disabled for this post