Blog Post

Enough about data caps: They’re a terrible idea

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

In a recently published piece, Prof. Daniel Lyons of the Boston College Law School argued that broadband data caps are a reasonable form of price discrimination. Lyons believes that data caps allow ISPs to more equitably distribute network costs among users based on how much they value internet access. He then goes on to suggest the best model of price discrimination comes from the airline industry, and that ISPs would be wise to learn from them.

Okay, wait a minute. The airlines? I had to read that twice to make sure Lyons was actually recommending that companies like Comcast and Time Warner –  you know, two of the lowest-ranked U.S. companies in terms of customer satisfaction – ought to be taking marketing tips from the industry that rivals them for most-hated status. (Interestingly, according to the American Customer Satisfaction Index, the airlines are third from the bottom, followed by… the cable industry!)

Opaque pricing models are opaque for a reason

This seems to me to be just awful advice (Disclosure: see below). One of the primary reasons consumers hate the airline industry as a whole is precisely because of standards of pricing that make no sense, are unnecessarily opaque and completely unpredictable. Certainly, there are other issues consumers gripe about, like on-time arrival (something they also share with cablecos, who sometimes, maybe show up between 8 and never), but at best, their pricing model simply makes no sense to the consumer, and at worst seems suspect and predatory.

Consider the confounding and inconsistent factors consumers have to wrestle with when trying to figure out how they can “use” a gigabyte (or, for many consumers, trying to figure out what one even is).  How do you know when you’ve used one?  Or are close to using one?  If some things are “under the cap” and other things “count,” how can you tell?  And why is that so?  Will the number of gigabytes of, say, a streamed movie, be listed along with movie ratings and reviews when searching for one – and if not, how will you know if you’re “allowed” to watch it?  Do security updates count?  Skype or Facetime?  The home Wi-Fi network that your neighbor’s kid set up for you?  And what happens when you go over the cap?  Will my TV suddenly turn off?

Make up your mind: Is it costs or capacity?

In January 2013, National Cable and Telecommunications Association (NCTA) president Michael Powell clarified in a speech that cable’s interest in data caps was no longer (or never was) about network congestion but instead about pricing fairness.

I had to read that one twice, too. So what of the angst over bandwidth hogs and bytes and bits and network management and capacity constraints? That’s not actually true? Well, okay, if the new argument is about how companies recover their investments and fairly allocate those costs then we can all agree that is quite reasonable. But if that is the case, then changes the debate to one about pricing (costs) and not about capacity (caps).

There are plenty of ways to address pricing that fairly charges customers without requiring them to pursue an engineering degree or a private investigator to figure it out. Let’s be clear: Broadband is not like electricity, where utilities must first generate the power they deliver to customers, requiring them to charge heavy users more because it costs the utilities more to serve them. Even the ISPs themselves allow that marginal costs for additional bandwidth are negligible between light and heavy broadband users.

A pricing model that works: the current one

Indeed, ISPs already have a way to offer consumers different price options for internet access – it’s called speed. If you are a comparatively light internet user who goes online primarily to send email and surf the web, you can buy a lower-speed tier and save yourself some cash. If you don’t see daylight much, and use your connection to watch a ton of online video, you’ll probably need to upgrade to (and yes, pay for) something faster.

Virtually all ISPs use this pricing model already which, it turns out, works pretty well. Most consumers don’t know a gigabyte from a hole in the ground, but they do know when their internet connection is slow. Pricing by speed offers consumers predictability on their monthly bills and an understanding of what they’re paying for. With data cap-based “penalty” fees there’s a big chance they’ll instead get a nasty bill shock at the end of the month and then wonder what on God’s green earth they did to deserve it.

Disclosure: The author’s company, Glen Echo Group, has a number clients involved in the broadband field representing a spectrum of interests: from the Alliance for Broadband Competition, to Gig U,. to Google, to Sprint, among others. See a full list here:

Maura Corbett is the president and founder of the Glen Echo Group, in Washington, D.C.

Have an idea for a post you’d like to contribute to GigaOm? Click here for our guidelines and contact info.

Photo courtesy  Suzanne Tucker/


10 Responses to “Enough about data caps: They’re a terrible idea”

  1. Ben Martin

    I’m sick and tired of this data cap nonsense. I can’t upload any of my beats much less start a business with these darn caps. It’s not fair to those that are in areas where ISP’s don’t exist and 3g is the only option. I meant 10$ for one GB!! I think I could get more out of a drug addiction then that!

    The government needs to do something to stop this and something right now! Or I’m going to go on a mad rampage and destroy half the city’s property in exchange for their ignorance that cost me time and money!

  2. Richard Bennett

    Maura argues that consumers are too stupid to understand any rational pricing plan for broadband, so broadband providers should use a status quo, irrational scheme that discriminates along a dimension that has nothing to do with the costs of providing service.

    This kind of thinking is unfortunately all too common in Washington, DC, where tech policy is written by people who generally fail to grasp technology in a meaningful way. While many of tech policy’s non-technologists bring useful knowledge to the discussion, generally law or economics, Maura’s public relations-based arguments are simply cynical and counter-productive.

    The most rational scheme for price discrimination is one that would incentivize providers to be early adopters of each new generation of broadband technology while offering low prices to low users and to non-users. Such a scheme would make advanced technology available to those who know how to use it and would also encourage the third of Americans who don’t subscribe to any form of broadband to get on the bandwagon. Such a scheme would offer the highest speeds to everyone, but would mediate congestion and pay investment-related fees by rationing priority. Mobile broadband works this way, and some wired plans are a close approximation, such as the Comcast Fair Share system that de-prioritizes heavy users on congested links.

    Designing pricing plans for easy communication by public relations professionals is less important than designing them to produce maximum public benefit. Let’s leave the manipulators of public opinion out of the dialog and focus on the things that make technology work.

  3. Cody Horner

    I disagree with data-caps for the most part, but in some cases as long as the caps are used for legitimate purposes they can be beneficial -in a sense-.

    For instance, stopping people from running home servers -or heavy P2P abusers- and tying up bandwidth for the rest of the node in their neighbourhood. I don’t care if someone pays more to do this; on a consumer connection the bandwidth should be available for everyone at all times (as much as realistically possible). If you’re a business, no matter the location, I would agree on being able to pay more for higher upload speeds and caps removed.

    The prices of data plans for all your different devices these days is getting a bit ridiculous though either way. These companies are just raking in the cash like nobodies business.

  4. The data overages problem will get even worse when the providers adopt LTE Advanced. You think now on AT&T owning an LTE Android phone with no way to turn off LTE eating away at you’re already small monthly data cap is bad. Wait until this new tech kicks in. I don’t understand these business plans offered by AT&T either. $500 per month for 50 GB of data? Insanity

  5. If you’re an ISP It is expensive to cap the customers’ speed other than simply config it’s standard port: 10Mbps/100Mbps/1Gbps. You could count his traffic without much hassle or allow him unlimited in his standard port. But caping a customer in someething like 27Mbps it’s really painful

  6. elfonblog

    Data caps are about imposing rations on a resource which is not scarce in order to sabotage it’s use for competing purposes (Internet video).

    High bandwidth prices are about wrenching every possible dime out of the market and keeping competitors from making a profit. The market (end customers) often think current prices are fair because they don’t know that the ISPs operate on a 95% profit margin. Competitors, specifically the small ISPs which don’t own backbone fiber, must buy their bandwidth from the big guys at whatever rate is available. They can’t peer data to earn credit. Their profit margin is razor-thin.

    The big monopoly telco/ISPs who own fiber peer their traffic so that they earn as much credit as they get billed. What these guys pay to keep multi-gigabit nationwide networks running is the relatively miniscule cost of keeping equipment powered on and cared for. It’s several orders of magnitude less than what they charge for bandwidth. Still, they cry about the “real” high cost of bandwidth, when it’s a level they’ve chosen to impose, and is a cost that is negated from their balance sheets.

    Monopoly providers have been reaping spectacular profits, and NOT investing in expanding their network’s capacities proportionately. To be blunt, they are conducting an embargo. They are creating a ‘shortage’ to justify high prices. But they have been spending money influencing our government and serving propaganda to the customers.

    They tried the “bandwidth hogs” angle and it didn’t work. Heaven forbid that someone out there was getting close to utilizing the bandwidth that they were promised. Now they’re returning to the “fair pricing” pretense. Like, it’s not fair for grandma to have to “subsidize” the other customers who download movies. Except that there’s damn little difference in cost to the ISP to deliver bandwidth to either of them. Most of the cost is in just keeping the lines functioning reliably. But the bandwidth represents activity that the ISPs find convenient to arbitrarily attach their billing structure onto. And customers go along with that, because it resembles the way other utilities bill them. But it’s entirely inappropriate since bandwidth costs nothing at all to provide. All of the cost is in the maintenance of the infrastructure, and new generations of equipment have far higher capacity at lower power consumption. For all practicality, it costs more to provide less bandwidth.

    Google is demonstrating that they can offer service an order of magnitude faster at current market price points, and nearly break even in the short term. We could have been at this point a decade ago if the entrenched telco/ISPs had been held to their obligation to build out the infrastructure as they promised. They were given billions (in tax breaks and construction waivers) but instead they built fiber to the cell towers and now want us all to pay more for wireless data plans.

    AT&T has gotten undeserved praise for scuttling up to the mic after Google’s gigabit fiber announcement in Austin, TX. Folks are saying that AT&T plans a competing service similar to Google’s. This is utterly false. AT&T said they were “prepared” (not promising) to build a gigabit “infrastructure” (not fiber to homes) and that they expected to be given the same sweet deal the city gave to Google. What AT&T is really after is the freedom to provide service only where the profit is highest, like Google. They want out of their obligations to provide service to almost every citizen. Any way, they’ve already got a fiber infrastructure expansion under way, and they want to skim some costs off it before it’s done.

  7. RickyRocky

    Datacaps are a horrible idea.

    Any company that uses them will see a flood of customers to other companies…so the marketplace will take care of this problem (in that the data capping companies will cease to exist).

    Only way datacap pricing could thrive in a broadband market open to competition is that the government mandates such pricing.

  8. Maybe the caps have nothing to do with capacity or costs of service, but just allow carriers a plausible way to offer different prices to customers with different abilities/willingness to pay.

    Maybe it’s simple price discrimination.

  9. Mcbeese

    Think of another time in history when network capacity needed managing – long distance telephone.

    The solution then was usage-based pricing and peak vs non-peak period usage.

    I think that speed is a good price variable for today’s Internet, but I also think time-of-day would be good. Services like iTunes and bit torrent could be configured to save the user money by only running in non-peak periods.