The Mobile Data Deluge
The storm surge in mobile data and the carriers’ race to handle it.
In April, Palo Alto start-up Pebble Technology posted a project on the Kickstarter funding site for a smartphone-connectible wristwatch. The developers wanted to raise $100,000 from potential customers. Within 28 hours, they had $1 million in pledged funding. Less than a month later – with 12 days still to go – they had nearly $9 million in support from nearly 6,000 backers.
People are so addicted to data that they’ll pony up $100+ for a watch yet to be built that will connect them to their data stream just a little faster than reaching into their pocket for their phone.
Whether "the last mile" finishes up "the last meter" via a smartwatch or not, the world is ravenous for mobile data. In the U.S., the number of wireless subscriber connections is now greater than the population, and the number of mobile data-capable devices – 295.1 million – is close to averaging one for every U.S. man, woman and child. 1 While all of those devices are not data-hungry smartphones or tablets (yet), consumers are gobbling up mountains of mobile data – streaming videos and music, shopping, social networking, cruising the Web. They can’t seem to get enough of it.
Wireless carriers are working every angle to push more data to users faster, from adding more cell towers to new antenna technology to Wi-Fi offloading. But the solution that holds the most promise for handling the burgeoning data load, and spiking data speed at the same time, is fourth-generation wireless technology: 4G, or more accurately, 4G LTE/Long Term Evolution.
With speeds that rival desktop/Wi-Fi connections, 4G LTE will be a boon for users lucky enough to have it. For website owners, though, LTE will be far from a mobile panacea; end-user adoption is going to take years, and a large swath of users will still be on slower connections. The challenges of balancing content and speed will continue for the foreseeable future.
For now, the spotlight is on the carriers and how they’ll make the transition to a new generation of wireless to serve a world that just won’t sit still.
Data doubling every year
AT&T’s CEO recently said he wished the company had never offered an unlimited data plan for the iPhone. 2 No wonder. In the five years since the iPhone effectively blew open the doors of the mobile data era, AT&T data traffic has grown 20,000 percent. 3
Of course, AT&T is not the only carrier feeling the pain; the surge in mobile data is a universal phenomenon. In 2011, U.S. wireless data traffic jumped 123 percent to 866.7 billion megabytes, on a 43 percent increase in mobile devices. 4Worldwide, mobile data grew even faster, by more than 130 percent to 546 petabytes, or eight times the traffic of the entire global Internet in 2000. Average smartphone data consumption jumped from 55 Mb to 150 Mb per month. 5
Mobile data traffic has basically doubled every year in the five years since the iPhone was first introduced.
The "datapocalypse" shows no sign of abating. Mobile data traffic is projected to grow at a compound annual rate of 78 percent through 2016 – an 18-fold increase, to 10.8 exabytes per month. (For visual reference, an exabyte is 10 followed by 18 zeroes.) 6
Last year saw another significant milestone: for the first time, video made up the majority of mobile data traffic. This marks an important trend that exacerbates the data problem: nothing uses more data faster than streaming video, and by 2016, two-thirds of mobile data traffic worldwide will be video. 7
Squeezing the data, stretching the pipes
The easy and obvious answer to an increasingly jammed highway is to add more lanes. Every wireless carrier is pursuing more spectrum to handle their data signals (see afterword). But acquiring and deploying spectrum is a long-term process. In the meantime, they’re taking a variety of incremental steps to boost data capacity and throughput.
One of the most basic strategies is simply to add more cell towers and antennas. In the U.S., the industry added more than 30,000 new cell sites last year– about a 10 percent increase – but growth of cell sites is dampened by availability of real estate and by local zoning limitations. 8 The federal government estimates it would take an additional 300,000 new cell sites, basically doubling the current network, to meet the data demand (in lieu of additional spectrum) over the next five years. 9 Nonetheless, increasing network density with more towers and better technology is a frontline strategy for handling more data.
Another strategy is "offloading," or moving traffic to Wi-Fi or fixed networks wherever and whenever possible. Estimates indicate that slightly more than 43 percent of data traffic worldwide was offloaded from cellular radio networks last year, and that by 2016 more than 68 percent of mobile data will be offloaded. 10 Offloading can range from simple home femtocell units (the ones consumers can buy to boost their indoor signal, which actually use Wi-Fi to create a "personal" cell) to large antenna systems that cover an entire arena or large building interior.
Another hardware strategy is MIMO, or multiple input, multiple output, which, by employing multiple antennas at both ends, increases data throughput.
Besides new and more closely located transmission hardware, carriers are employing a variety of software solutions that control how data is transferred and how efficiently cell towers handle traffic, among other things.
And then there’s the elephant in the room no one wants to talk about: data throttling for heavy users of legacy unlimited plans, and pricing-based demand management. Users are going to have to resign themselves to watching their data usage, just as they have before with voice minutes.
But all of these efforts are so many fingers in the dike trying to keep the data surge at bay. By far the most promising solution is the transition to 4G networks – though the term "4G" itself means different things to different carriers.
Will the true 4G please stand up?
Sometimes it seems it would take an act of Congress to decide what’s really 4G and what’s not. In fact, a bill was introduced in Washington last year that would have required carriers to specifically spell out their 4G performance claims. 11 But it only got as far as most U.S. legislation gets these days, that is to say, nowhere.
The industry itself has settled on a two-tiered definition of the next-generation networks that succeed 3G. There’s generic 4G, which includes a variety of enhanced 3G networks such as EVDO (evolution-data optimized) and HSPA/HSPA+ (high-speed packet access) that deliver far faster data speeds than plain-vanilla 3G.
The real gold ring, though, is 4G Long Term Evolution, or 4G LTE. Regarded as the true next-generation wireless network technology, 4G LTE requires a significantly greater capital investment for network upgrades than the largely software-driven process of enhancing an existing 3G network.
Theoretically capable of download speeds up to 299.6 Mbps and upload rates up to 75.4 Mbps (purely theoretical) 12, in practical terms LTE can deliver average download speeds of 5 to 20 Mbps, a large multiple of current 3G speeds that hover around the sub-2 Mbps range. LTE handles more data, more efficiently – much more efficiently. And the faster the data moves, the more room there is for more data to go through. User experience also is improved by dramatically reduced latency – itself solving one of mobile performance’s biggest challenges – and seamless transitions as users move throughout the network.
While every carrier in the universe is marketing their "4G" networks – though most of some and all of a few are still based on 3G technology – every single one is also scrambling to get true LTE coverage in place. It will simply be impossible to be competitive with anything less.
In a recent "speed showdown," PC World compared 3G and 4G performance across the U.S. and found the slowest 4G LTE download time (Verizon) was nearly twice as fast as the fastest 3G time (T-Mobile). Within carriers, the difference was even more dramatic. AT&T’s 4G LTE was nearly 3.5X faster than its 3G, and Verizon’s 4G LTE was fully 7X faster than its 3G download.13
The growing LTE footprint
Figuring out the coverage of a carrier’s LTE network is almost as hard as figuring out what’s 4G and what’s not.
In the U.S., Verizon, AT&T and T-Mobile all claim to have the largest 4G network. At this writing, though, Verizon is the clear LTE leader, with its network reaching over 200 million “pops,” or population capable of accessing the service. While AT&T’s 4G network has similar coverage, the LTE portion is significantly lower, at 74 million pops. T-Mobile doesn’t have any LTE coverage yet, but its HSPA-flavored network matches the 200 million pops coverage.
Actual LTE subscribers in the U.S. were at about 10.5 million as of May. That’s not a huge number, but not bad considering how new the networks are and how few LTE-capable devices are yet available.
Worldwide, LTE is available on 57 networks in 34 countries; that’s expected to expand to 100 by year-end. A total of 22 million LTE connections are forecast for 2012, and the U.S. will account for about half the total. 14 By 2016, 428 million LTE subscribers are projected 15 and expected to generate $265 billion in revenue. 16
An expensive and complex transition for carriers
The investment to build out LTE networks is huge. AT&T spent $20 billion overall on its networks in 2011, and expects to spend that much again in 2012; 17 this is not an atypical investment for the major carriers. Juniper Research estimates that carriers worldwide will have to spend $840 billion over the next five years to upgrade their backhaul networks – the connection from their wireless networks back to the Internet – to handle the increased data. 18
“Some of these operators don’t have a good backhaul – the pipes are small," says Mohammad Zaman, North American sales director for Keynote SIGOS. "You have to build a bigger pipe in the backhaul, on the network side. As use increases, there’s a limit to how much you can handle."
The transition to LTE is complex on many levels. The LTE hardware has to be put in place and tested, and carriers have to make sure that as users move into and out of LTE coverage areas – which will be inevitable for some time to come – their experience is seamless and uninterrupted. There’s also the issue of voice. Currently, LTE is a data-only technology. Voice calls will have to be handled via a maneuver called circuit switch fallback.
"As you build these 4G networks, you’ve got to be able to do this seamless transition, seamless handoff – the circuit switch fallback between these 4G and 3G networks," Zaman says. "You need tools to do that. You want to make sure that you can monitor the customer experience, the quality of service of the network, so that the user doesn’t know what’s going on – so that it’s transparent to them."
Roaming issues also have to be sorted out, as different carriers will operate LTE on different bands across the spectrum. Maintaining quality of service across all these variables will be a challenge during the transition to LTE and beyond.
“All of my customers in the U.S. are looking for solutions for testing LTE deployment, as well as for testing inter-carrier roaming," Zaman says. "They are going to set up these roaming agreements where they need a solution like Keynote SIGOS GlobalRoamer to test the quality of service between the carriers."
Still a balancing act for site owners
No matter what new things happen in mobile, site owners can’t seem to catch a break. LTE adds still another wrinkle into the complex tangle of OS/device/carrier variables that complicate mobile strategy formulation.
Theoretically, higher-speed connections are good news, enabling sites to push out more content and functionality to create a richer experience. On tablets in particular, sites could deliver desktop-type deep content enhanced for the greater interactivity tablets allow (see the Benchmark article, "Tablets in the Fast Lane"). Smartphones, too, could handle more complex, deeper content.
But while LTE coverage is growing rapidly, actual LTE adoption by users is and will lag significantly behind. The 428 million subscribers projected by 2016 are just six percent of the total subscriber population. 19 That leaves 94 percent of the mobile universe connecting at 3G or non-LTE 4G speeds, that is to say, at some speed significantly slower than Wi-Fi-like LTE.
So while Long Term Evolution may promise a revolution, at this point the emphasis has to be on "long-term." For site owners, the same mobile rules and best practices apply today as they did yesterday: Create or optimize sites for the two major mobile form factors – smartphones and tablets. Tailor the experience to the device. Make sure the site(s) are lean, efficient, and fast-loading over a typical 3G network. Test, monitor, tweak, repeat. Some of your lucky users will enjoy blazing LTE performance. And the rest will have an experience that meets their expectations and creates satisfaction for your brand.
Afterword: What about spectrum?
All the talk about the "datapocalypse" and the race to build out LTE begs the question: What about spectrum? Wouldn’t more spectrum solve the problem? The answer is – yes, but it’s complicated.
The federal government says we have a "spectrum crunch." The carriers refer to it as more of a crisis. Everyone agrees that getting more of the right spectrum into the hands of the wireless carriers would go a long way to easing the current and looming data crunch, as well as enhancing the potential of wireless data to transform how our country lives and works, boosting our economy and enhancing our competitiveness.
But reallocating spectrum is easier said than done. There’s a lot of it still in the hands of broadcasters who no longer really need it. The government controls a lot of it, though no one seems to know quite how much or who has the authority to reallocate it. Satellite and cable companies are sitting on significant swaths that they no longer seem interested in using to enter the wireless market.
The big wireless carriers are desperate to get more spectrum any way they can. The reason AT&T tried so aggressively to acquire T-Mobile was really to acquire its spectrum; ironically, as part of its deal, it now has to turn over some of its own spectrum to T-Mobile as a consolation prize for the failed acquisition. Now, AT&T is looking to make other deals with smaller carriers and/or satellite companies to score more spectrum.
Verizon is aggressively pursuing a deal to acquire spectrum held by SpectrumCo, a consortium of three cable companies, which it would use to build out its LTE network.
For carriers, acquiring spectrum is a golden ticket; once they own it, no one else can use it. It’s a competitive advantage well worth even the highest price.
The problem is, it takes a long time to put together and execute these deals, and they are all subject to FCC scrutiny. Congress has authorized the auction of commercially held spectrum (amazingly, as part of the Middle Class Tax Relief and Job Creation Act of 2012). The White House supports it. But still, it is typically a matter of years, not months, before spectrum is successfully transferred and comes online.
In the meantime, the carriers will have to look to LTE and other network-enhancing technologies to solve their data-capacity challenges and boost their data revenues. Short of additional spectrum, there’s still much they can do. There are even some who believe that technology can solve the entire problem. Martin Cooper, the inventor of the cell phone, was quoted in The New York Times as saying, "Somehow in the last 100 years, every time there is a problem of getting more spectrum, there is a technology that comes along that solves that problem." 20
For now, the hope is that 4G LTE will be a big part of the solution.