18 Comments

Summary:

Sending a bit over a wireless network is 200 times more expensive than sending a bit over wireline, which explains some of the high costs and limits of wireless data plans. How can operators drive down these prices so wireless doesn’t lose its luster?

teenstexting

News about wireless bandwidth hogs, new session-based pricing from Leap Wireless and the appearance of a new web site aimed at helping consumers understand their data caps and the limits those impose, all point to a growing problem in the wireless industry. And that problem isn’t congestion. Rather, unless the industry figures out how to give people connectivity at a reasonable costs, wireless will always be luxury access technology and ubiquitous connectivity will be a pipe dream.

The problem isn’t congestion, it’s a stagnation.

A study Friday noted that the top one percent of wireless users consume half of all the data. Meanwhile, Public Knowledge on Friday launched a web site designed to help consumers understand their data caps. On Thursday Leap Wireless’ CEO said the company would begin offering data sessions in addition to its regular tiered data plans. Under that scenario a user might buy a data plan just like he would buy pre-paid minutes on an as-needed basis after reaching his cap. All these bits of news are linked by one key problem: wireless data is in high demand, but it’s also expensive to deliver.

And the tension between what consumers want from their wireless networks and what operators want to give them is leading to stories that harp on congestion, new pricing models and consumer advocacy around high-priced plans. But it’s time to stop trying to address that tension solely with new types of rate plans, and customer education. If we want wireless data to become ubiquitous and deliver on the promise of connectivity, the industry needs to address its costs and educate consumers on those costs in a transparent way.

Part of the problem is just a matter of physics — airwaves can only carry so many bits per hertz — but other aspects of the high cost are related to policy and the reluctance of the industry to embrace, or even talk about, technologies that will help them deliver wireless at a lower price per bit. Right now, sending a bit over the cellular airwaves costs a lot more money than it does to send that same bit over fiber or even DSL. How much more depends on if you are in a populated city or out in rural America (it also depends on if you are in America) as well as the type of network the bit is sent over (i.e., LTE, CDMA or HSPA). But broadly speaking it’s at least 200x more expensive to use cell networks according to analyst Chetan Sharma. He estimates that number will drop over time to 100x, but clearly that’s still a huge disparity.

Not all bits are equal (or as expensive). So let’s rethink the network.

Fortunately, not all data has to travel over the gilded cellular pipes. Smart consumers already use Wi-Fi networks for streaming video and movies, but ideally this will become more automated. This means operators must include Wi-Fi in their networks, and actively shunt certain types of traffic to those networks when available. In short, we need application-aware wireless networks that send traffic to the cheapest, but most appropriate network the application can use and the consumer will accept.

This means when I stream YouTube videos, my carrier routes me over to Wi-Fi if it’s available but my email and voice calls stay on 3G if the Wi-Fi is weak. As a consumer I would advocate Wi-Fi as the default network with carriers switching me over to a cellular plan only when absolutely necessary, much like upstart Republic Wireless tries to do. Buying cell phone plans becomes a little more complicated, perhaps involving a short questionnaire that a consumer fills out ranking what types of traffic he needs to get instantaneously versus the traffic he is willing to compromise on.

This new type of plan also means that consumers may have to accept lower quality service for streaming video, might end up paying for access to a carrier Wi-Fi network and will need to accept their operators monitoring the applications they use. There’s a role for developers here in building tools that help consumers see exactly what their operators are doing, and the FCC should stay active in enforcing the spirit of the network neutrality rules. I have a hard time believing that carriers could behave well enough for me to trust them with something like this — just look at their historical stances on Wi-Fi, or the recent questions around Google’s Wallet service on Verizon’s network, but something has to give and I don’t think it will be the operators.

We want what we want. Until we have to pay for it.

CTIA says ladies like their mobile data.

Despite the cost of wireless plans, we want and will use wireless data. On Friday, the CTIA put out a study noting how women use the wireless network for an increasing amount of stuff. And articles offering a word of caution about viewing the Superbowl on your mobile phone get that while it may make you bust through your data cap, people will watch bits of a big game on the go. That very idea was unthinkable a few years ago, but mobile has changed our surfing, shopping and even our TV watching habits.

Carriers have moved forward in delivering faster networks that can deliver between 5 and 12 Mbps down — enough for video, voice and even the most demanding web services — but their current cost models don’t match up with the usage expected and advertised on the networks they’re building. Consumers look at carriers’ pricing, their marketing (which shows customers streaming video on their phones) and their comments in the press about high costs for mobile data and congestion, and recognize that carriers are not telling the whole truth. If network resources are such a precious commodity, then why not price data so it costs more at peak times? Or why even encourage video on the LTE network?

But when will that disconnect between the ease of using a service and the high cost of that service start to change or curtail consumer behavior? In short, when will a user suddenly think, “Maybe I shouldn’t use my phone for this, right now?” In a mobile-first world, will wireless become a second-class access technology, or will carriers adapt their networks and their cost structures in time?

You’re subscribed! If you like, you can update your settings

  1. rob the computer guy Sunday, January 8, 2012

    Who the whatsa are you writing this for? Did you get a kickback from a wireless operator? The notion that providing the data is expensive is absurd, and thus your article isn’t even readable. It’s farcical on its surface!

    The cost for adding to the infrastructure is always going down, and that means it’s cheaper to add to older technologies and it’s cheaper to add new ones. This is always on a DOWNWARD TRAJECTORY.

    For you to write a ridiculous article upon any other premise than that these companies charge what they want, unrelated to how much it costs to provide, is just obnoxious and abhorrent to basic American values of capitalism.

    And if you think capitalism is about charging whatever you want, it’s not.

    The prices of data should have at least stayed the same during the time that the cost of adding capacity to their networks was steadily going down. Instead, they have skyrocketed.

    Unfreakingbelievable POS article.

  2. This article assumes the status quo in wireless communications. Mesh networking (802.11s and related standards) allows communication without a pre-established infrastructure, without mobile towers, without satellites, without wifi hotspots, and without carriers. If the carriers can’t figure out how to deliver a cost effective product, someone will always eat their lunch. Unless of course they buy legislation to make that illegal, as is so popular these days.

    1. Stacey Higginbotham Mace Moneta Monday, January 9, 2012

      Mace, I am intrigued by mesh and other tech to help bring data from the air where it’s expensive back to the wireline network which is cheaper. I do think carriers need to deliver cost effective products and am curious how much control they will cede at the edge to do it.

  3. Anthony Williams Sunday, January 8, 2012

    Total load of rubbish. I pay for my access to wireless internet. Therefore it is the company’s job to provide that service. Lets face it..right now the hot potato in technology has been in the mobile industry. Super-phones that have the capability to connect, download, upload large amounts of data and information. How is it the consumers fault that these companies tout a service that they can’t either provide or have not prepared for? You can’t put the genie back in the bottle. I believe that share holders should stop looking for the massive payday and understand that the mobile industry is not going anywhere. Profits will be there not however all at once. It’s a safe bet. So stop squeezing the consumer by offering what you cant provide, ante up, build out the infrastructure, and provide the service that your all bragging on.

    1. Anthony, you are right that operators promise things like HD video in their marketing and then whine about congestion when it comes to policy and pricing, but it’s not rubbish that wireless bits cost more to deliver. So operators do have to think of new ways to to deliver their services and do it in a manner that helps educates users as opposed to gouges them on price.

  4. hmmm much as I like gigaom, I am not sure this article is addressing the right problem .. as some of the early comments are saying. I said this yesterday – what if you called a bandwidth crunch and no one came? aka Do you ever hear of the spectrum/bandwidth crunch in Boise Idaho? – without giving up the FUD – its hard for anyone to take the carriers seriously .. http://www.opengardensblog.futuretext.com/archives/2012/01/do-you-ever-hear-of-the-spectrumbandwidth-crunch-in-boise-idaho.html

    1. Stacey Higginbotham Ajit Jaokar Monday, January 9, 2012

      I agree. But there are engineering issues associated with delivering HD video for example over wireless ( see here http://gigaom.com/broadband/video-haunts-and-shapes-the-spectrum-debate/ ) that make cellular an expensive and impractical medium for delivering it. Like any industry watcher I have my doubts about the carriers, but it’s not like they are going to disappear.

  5. The proposed solutions have been written up here in GigaOM showing how absurd they are:
    – AT&T charging your mobile plan for data that travels over your DSL/WiFi package
    – Verizon wanting to count data while roaming abroad with an iPhone and a local SIM card

    These all are bad business models based on excuses that make the mobile industry sound like the motion picture industry – “keep my old profits and model as untouchable regardless of technology available to use it.”

  6. The key point in this ground-breaking article is not $$$ but QoE. To qoute “operators must include Wi-Fi in their networks, and actively shunt certain types of traffic to those networks when available”. In todays cellular networks $$$ can not buy you QoE. Any attacker equipped with 10 smartphones and tailing you closelly will make your BW drop to snail-rate. Unless there is alternative network (such as WiFi) you have no freedom to escape, and operator has no means to provide you with alternative.
    It is a well known legend that operators discoraged cell-phone vendors from including WiFi. It was one of Apple’s bald moves to have WiFi in iPhone and to think in terms “WiFi 1st, 3G – 2nd”. See http://www.wired.com/gadgetlab/2011/11/steve-jobs-iphone-no-carrier/
    I’m sure cellular operators will adapt, in the meantime let’s hope that MVNO will exploit this venue and leave incumbents in dust WRT QoE that intellegent WiFi and Mobile cooperation can provide.

    1. Who is going to be equipped with 10 smartphones and tail me closely?

  7. Emanuel Fleishman Monday, January 9, 2012

    This is great and timely article. Key take-away is “operators must include Wi-Fi in their networks, and actively shunt certain types of traffic to those networks when available”. This unfortunate reality for operators who historically feared WiFi – check story of Apple’s rebelt in http://www.wired.com/gadgetlab/2011/11/steve-jobs-iphone-no-carrier/ .I expect MVNOs to differentiate themselves by proactive (and transparent to user) sharing of WiVi bearers to provide the best possible QoE. I believe that Operators will follow, and we will live happily ever after without worrying to much about which application uses what network.

    1. Emanuel
      Shunting traffic to wifi is old news. what is new (and which I dont agree with) is the below.

      “This new type of plan also means that consumers may have to accept lower quality service for streaming video, might end up paying for access to a carrier Wi-Fi network and will need to accept their operators monitoring the applications they use”

      In that sense, I agree with the other commentators – absurd, ‘load of rubbish’ etc. Nothing great and timely about shunting traffic to wifi

      rgds Ajit

  8. Thomas Sachson Monday, January 9, 2012

    I agree with Stacey — we need solutions to the mobile bandwidth problems that have emerged and will only become more pronounced with greater data use and resulting data caps. But the solutions (BOTH economic and technological) need to make sense not just for the end-user but also for the carrier.

    For the end user the solution must:
    A) allow the end-user 100% choice and control over how they download data;
    B) is easy and obvious for the end user to use everyday; and
    C) lowers their monthly bandwidth and per byte cost.

    And for the carrier the solution must:
    A) be easy for the carrier to deploy and maintain (no BSS/OSS issues);
    B) result in better traffic management;
    C) makes economic sense for the carrier to deploy (preserve margins / ARPU).

    One technology and business model solution that satisfies all of these criteria (and is also very good for content creators since they get a direct relationship with the end-user — no more expensive content delivery aggregator commissions) is for the carriers to enable a data platform where third parties can voluntarily provide”free/subsidized” delivery to the end-user by paying the carrier for the cost of “byte freight” (just like the 1-800 phone call for voice, which was extremely successful for all involved). The technical trick, however, is to not attempt to do the byte auditing and billing inside the network (DPI — very tough to scale/implement for every billing situation) but to disintermediate the process — pushing the monitoring of bytes to the edge and onto individual apps (e.g. Android) on the phones / tablets. We call this process of pre-bundling free bandwidth into the apps themselves “FreeBand” (see http://www.boxtop.tv // info@boxtop.tv) and think it is a very good solution — and one that can be rolled out quickly should the carriers choose to give their customers the ability to (i) either pay for the bytes themselves (out of pocket cost to end-users), (ii) to have someone else pay for the bytes (free delivery of bytes wrapped up with the content), or (iii) some combination of both.

    If anyone (mobile carriers, OEMs, content providers) is interested to discuss in greater detail, I am at CES Las Vegas this week and happy to get together. And again, many thanks Stacey for another timely article.

  9. Demand your providers upgrade the infrastructure with INFINERA gear now!!!

  10. LTE has provisiones for SFN broadcasting, of course telcos don’t want it because it can ruin their current traffic-based business models and they are in bed with content providers. There us also IP multicasting, not as simple but doesn’t requiere new hardware on the user side.

    1. AT&T and V are trying to offload as much traffic onto wi-fi as they can. My iPhone asks to join wi-fi networks at every available opportunity. And if I have logged in to a wi-fi site once, it automatically does it after that.
      I think this article is proposing that phones automatically use any available unencrypted available wi-fi. But even unencrypted wi-fi usually requires the user to access a screen and click to accept terms of use. Whenever I am in Starbucks, my iPhone switches to wi-fi, but I have to click to accept the AT&T terms of use.
      The only way to seamlessly switch from wireless to wi-fi is for a carrier to use its landline facilities to offload its wireless data users. So lots of carriers are adding femtocels to their networks for this purpose. Some are also using their customers’ modems to accept wireless traffic. This last option raises legal issues. Most large carriers have asked for customers to opt into allowing their modems to do this. But, like all opt-in programs, very few customers have proactively taken the steps to accept the program, even when the carriers have provided incentives. That is why it is mainly start-up companies that are making this work. They can incorporate the agreement in their terms of service, because they are dealing with new customers. And regulatory bodies, AGs, public interest groups, etc. are less likely to go after start-ups for consumer protection issues.

Comments have been disabled for this post