Over the weekend, blogger John Cozen posted a recent e-mail exchange with AT&T about why his smartphone data connection was slowed down after he breached 2.1 GB in his last billing cycle. His argument was his usage couldn’t be subject to throttling since his data use must be well under the top 5 percent cut-off AT&T stipulates in its terms. AT&T’s response was very interesting:

“To give you a baseline – the average data use across the country by the top 5% of AT&T smartphone customers was 2GB per month, effective August 2011. The amount of data usage of our top 5% of heaviest users varies from month-to-month and by market, based on the usage of others and the ever-increasing demand for mobile broadband services. To rank among the top 5%, you must use an extraordinary amount of data in a single billing period.”

When AT&T introduced throttling in October, its highest-tier data plan was 2 GB for $25, compared to the $30 charge for unlimited plans that Cozen and millions of older AT&T customers still hold on to. That seems reasonable enough. If the 2 GB is what the top 5 percent of smartphone users consume and is at the level AT&T considers abusive, then 2 GB is a good place to set its cap, charging customers more if they exceed it.

But AT&T just overhauled its plan pricing. It’s newest mid-tier plan charges customers $30 for 3 GB. Why is AT&T inviting new customers to consume a full gigabyte more of data while telling older customers – who pay the exact same monthly fee – that 2 GB of data is excessive? My bet is that the former is really a false invitation.

When AT&T first announced these new plans, I wrote there was good and bad in them for consumers. The bad is that new subscribers will have to pay $5 more a month than their predecessors for any of AT&T’s plans. The good is that AT&T is actually lowering the per-MB charges on data, which is ultimately necessary if average mobile broadband consumption continues to grow. Now I’m not so sure.

If 2 GBs is the average use for AT&T’s 5 percent hungriest users, that means 95 percent of AT&T’s customers are well under 2 GB each month. So the vast majority of A&T customers don’t get any real benefit out of the new 3 GB plans. To them it just amounts to a $5 a month price increase.

To be fair, AT&T can’t just price for what its customers are consuming today. It has to price for where they’re going, otherwise it would just be adjusting its rates every few months. It’s not unreasonable to assume that its customers average monthly data consumption will grow beyond 2 GB in the next few years, especially as mobile video services take off. AT&T is also a business that wants to make money off of mobile data, though this may be a rather sneaky way of doing it. Ma Bell’s data rates still far undercut those of its main competitor Verizon Wireless, however.

But AT&T also claims to be facing a capacity crunch. Since its planned acquisition of T-Mobile has failed, the carrier has used every podium it can find to proclaim that its networks are reaching critical mass in an effort to justify its spectrum acquisition aims. On its fourth quarter financial call, AT&T Randall Stephenson even blamed the government’s failure to let it have T-Mobile as the reason why its forced to raise data prices.

If AT&T is so network constrained, if 2 GB of monthly usage is too much for its networks, and if it doesn’t have the spectrum to meet future demands, then why is it opening up the data spigot, actually encouraging its customers to consume more for an extra $5 a month?

iPhone image courtesy of Flickr user mark sebastian

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• 2012: Data, spectrum and the race to LTE
• The future of mobile: a segment analysis by GigaOM Pro
• Mobile Q4: The scramble for spectrum continues

Indeed, in the past 11 months, Verizon moved quickly to shore up a competitive advantage. The LTE network launched in December of last year, showed extremely fast speeds, has a number of LTE smartphones available and is on pace to cover more than 185 million Americans by year end. What has that speed to market done for Verizon? Mike Roberts, principal analyst at Informa Telecoms & Media said in an emailed press release:

Verizon has added around 2 to 2.3 million iPhone subscriptions every quarter in 2011, but new LTE subscriptions nearly tripled from 500,000 in 1Q11 to 1.4 million in 3Q11. Looking at a different metric, LTE accounted for 53% of Verizon’s total postpaid net subscriber additions in 3Q11, up from 30% in the second quarter. The bottom line is Verizon is increasing market share, subscriptions and revenues with LTE.

As a result, Verizon may be in a good position to combat AT&T’s comparable LTE network, which recently launched in five cities and is soon adding more markets. But Verizon’s broader LTE network still faces one challenge that AT&T’s doesn’t: It’s not backwards compatible with the carrier’s 3G technology. In contrast, AT&T doesn’t have the same problem since LTE networks are based upon GSM standards that AT&T has long used.

Perhaps that’s why Verizon is running a desirable promotion right now. With the purchase of any new LTE smartphone, Verizon is doubling the monthly 4G data limit at no extra cost. The $30 plan that gets 2 GB of data now allows for 4 GB; the same doubling applies to Verizon’s 5 GB and 10 GB plans, which are currently $50 and $80 per month respectively.

Not only does this take some of the focus away from AT&T’s new LTE network; it helps address another problem for LTE handset owners. The faster speeds are fantastic, but that means it’s easy to blow through a monthly data cap in a shorter time. Thanks to the aggressive new promotion, the risk of hitting that limit is cut in half.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Sprint’s tightrope walk: finding a balance for its network modernization plan
• 4G: State of the Union
• U.S. Wireless Data Market: Q4 and Year-End 2008

To help its customers adjust from unlimited plans, Verizon has created an online calculator with a number of data-intensive activities, which complements but differs from AT&T’s own data usage calculator. And that’s the crux of the problem here, because different smartphone uses eat up different amounts of bandwidth in a given time. Think of the system like electricity, although in this case, you have to measure by the megabyte per hour instead of a kilowatt per hour. You can light a room with 100 watt incandescent bulb and use 100 watts in a given hour or you can illuminate the same room with a more modern LED light bulb using (and paying for) only 13 watts of power in a hour. The same concept applies to emails, web browsing, VoIP calls, and streaming video: All use the same network, but require different amounts of broadband.

Verizon’s calculator does a fine job at providing a monthly estimate of broadband data needs based on the wide range of mobile activities, but here are some guidelines on just how much data different activities use on a smartphone. Bear in mind that these are general guidelines as the broadband needs of some similar activities will vary, which makes this all the more complicated. Streaming the same video in a high quality or high-resolution setting will use more bandwidth than streaming the same video streamed at a lower setting, for example.

Again, these are guidelines to get you started in understanding how much mobile broadband smartphone activity requires. Because some activities use small bits of data, such as social networking status updates, tweets and text-only instant messaging, I’ve bypassed them in this table. But data-intensive activities, like those in the chart, can add up. A 2 GB monthly plan, could be used up in as little as a few hours if you stream high-quality video to a smartphone, for example.

There are a number of varying factors when considering mobile broadband needs, but one mantra should hold true: The higher the quality or the more immersive the experience of any media or file, the more bandwidth it will require. That’s just like the electricity example. If you want brighter lights in a room, you’re going to need more electricity.

It’s also worth noting that the more mobile broadband you purchase under either the AT&T or Verizon plan options, the cheaper the service is on a per-gigabyte basis. At the low end, a 200 MB plan, or one-fifth of a gigabyte is $15 or $30, depending on the carrier. That works out to a per-gigabyte cost of $125 to $150. But a 2 GB plan from AT&T is $25, making the price of a gigabyte a much more palpable $12.50. And a whopping 10 GB from Verizon is $80, bringing the per-gigabyte unit price down to $8. Overages for both carriers in most cases are $10 per gigabyte, which happens to be very close to what it costs a carrier to deliver a gigabyte of mobile broadband, according to Chetan Sharma, a wireless analyst.

Note: the table was built upon certain assumptions (which will vary by user) and created with help from an online bandwidth calculator as well as information from the carrier sites.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Carrier IQ and the continued erosion of operator trust
• Report: How Mobile Cloud Computing Will Change Tech
• The fourth quarter of 2012 in mobile

The new tiered plans have been rumored for months — Verizon signaled its intent last year and confirmed it this past March — but have just been confirmed today. Fierce Wireless received confirmation from Brenda Raney, a Verizon media spokesperson. The new monthly plan costs and limits break down as follows, with a $10 overage for an additional GB of data:

• 75 MB for $10 (Update: per a correction to the original Fierce Wireless report, this plan is solely for non-smartphones, i.e.: 3G-capable feature phones)
• 2 GB for $30
• 5 GB for $50
• 10GB for $80

Depending on the service level, some plans are more expensive than the similar offerings from AT&T, while others are more attractive. A 2 GB data plan from AT&T, likely more than enough data for most casual smartphone owners, is $25 per month, or $5 less than what the same amount of mobile broadband will cost new Verizon customers. But heavy data users will do better with Verizon: It would cost $55 for an AT&T smartphone owner to have 5 GB of available data. And although 10 GB is overkill for all but a small percentage of current smartphone customers, AT&T would charge $105 for such capacity as compared to Verizon’s $80. AT&T wins at the very low end though: Instead of a paltry 75 MB from Verizon for $10 a month, $5 more nets 200 MB from AT&T. (Note: see update above, as the 75 MB plan from Verizon is for feature phones.)

Related to the new smartphone plans are details about Verizon’s LTE and mobile hotspot services. The Droid Life blog reported last week that customers with an LTE-capable device, such as the Samsung Charge or HTC Thunderbolt, will have to start paying for the 4G mobile hotspot feature; a service that has been free up to now. On July 7, in tandem with the new tiered data plans, customers will reportedly pay $20 per month to gain both an additional 2 GB of data and the ability to use the mobile hotspot functionality of the 4G smartphones.

Given the rising appetite for mobile broadband in the face of increased smartphone adoption, Verizon’s move was expected by many and considered inevitable by some. Coverage and network reliability are still key decision points for any mobile broadband service, but now customers on the two largest networks will be factoring in their varying data usage needs, and the monthly costs required to meet those needs. Sprint still offers unlimited data, and while T-Mobile offers “unlimited” plans, the carrier does slow network speeds after a tiered amount of data is reached in a month. With T-Mobile very likely to be consumed by AT&T, that may be a moot point, however.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The future of mobile: a segment analysis by GigaOM Pro
• Report: How Mobile Cloud Computing Will Change Tech
• Consumer privacy in the mobile advertising era

The least expensive contract plan, which includes 500 voice minutes, unlimited messages and just 200 MB of fast data access is now priced at $59.99. Heavy data users can bump up their monthly mobile broadband limit to 2-, 5- or 10 GB of data for additional costs; the 10 GB plan in that case is 109.99 per month. The carrier continues to call such data plans unlimited, which is technically correct, since data access is not shut off. However, I suspect consumers will be upset if they choose the wrong data plan and hit their limit earlier than expected, forcing them to see lower data speeds.

Data tiers may be the sticking point for many consumers as we’re using our phones more for apps and web browsing than for voice in most cases. And T-Mobile recently modified its Wi-Fi calling terms, which can help reduce the number of voice minutes a consumer needs. Instead of Wi-Fi calls counting against monthly minutes, they’re now free on qualified plans.

To a degree, I think T-Mobile is playing loose with the term unlimited, because it’s unlimited with an asterisk. Even the carrier’s own blog makes note of that: Just last month, a new plan was introduced and T-Mobile put together an infographic that focused on the company’s “truly unlimited” plans. Note the asterisk and caveat about the slower speeds:

Again, T-Mobile isn’t limiting the amount of data it will provide a paying customer. Instead, this approach is a slight twist on a traditional tiered data plan and has been in place since last April for some plans. However, I wonder how many customers fail to see the asterisk once they see the word unlimited. Ultimately, I’d prefer to have full speeds and simply pay for the data I use, but the industry, at least in the U.S., is moving towards tiered data buckets. Some of those buckets have asterisks, and some don’t.

Image courtesy of Flickr user, san_drino

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Confused about the wireless markets? Here’s a breakdown
• Mobile 2012 and beyond
• LTE-Advanced: what it is and isn’t

This is really about operators being upset over the Googles and Apples of the world profiting off the pipes the operators have built. Instead of realizing they are about providing access or building awesome content of their own, they sit at the table begging for a chunk of profits from companies which have built something on top of the operator’s networks. When begging doesn’t work, they threaten to stifle access and innovation. The carriers are like old dogs that can’t learn a new trick. Unfortunately, right now, the only trick they have is peeing on your rug.

Ed Whitacre, AT&T’s former CEO, tried to play this card a number of years ago on the wireline side, pushing for content providers to pay a quality of service tariff. The plan didn’t go through, but it showed some of the vacuous thinking of operators. That wireless carriers in Europe are looking at flogging the same idea shows they’re either dumb or really bold. Either way, it smacks of desperation.

What the carriers don’t understand is that they provide access. Yes, mobile data usage is soaring but that’s the business the carriers are in. They rely on Google and others to get people to buy their data plans and their phones. Without great services, why would people want to pony up for a smartphone or a data plan? It was great just a couple of years ago when data revenue helped offset declining margins on voice. But now that data is exploding, it’s put the carriers in a tough position.

However, trying to get money from content providers is not the way to go. The carriers are already charging users for data, and are in the midst of changing their pricing plans to reflect the very real flood of traffic caused by user demand. Right now, their wireless businesses are still very profitable; for example, Verizon reported operating income on its wireless business of 29.9 percent. It’s preferable that the operators look at tweaking that model to manage traffic and revenue. We’ve advocated innovative pricing for data plans and dynamic pricing, which makes more sense in a world with limited bandwidth and spectrum. They shouldn’t try to build a double-sided market by dipping into the revenue of companies, who have legitimately created valuable services on top of their networks. A flour mill owner can’t approach the baker for a cut of his bread revenue. This would stifle innovation and potentially undermine the dynamism in the mobile world. Will popular app makers have to worry about paying as well if their apps are consuming bandwidth?

If the wireless operators are peeved at Google and others making money off their pipes — and they are no doubt mad — they should build their own services that appeal to users. Otherwise, the carriers shouldn’t argue that they’re more than dumb pipes. The carriers may think they’re getting out early on charging content providers and may have more leverage than their wired counterparts. But pushing an unpopular model again doesn’t reflect bold thinking; it really shows that the idea cupboard is bare. Let’s hope their American wireless counterparts don’t try the same trick here.

Related content from GigaOM Pro (sub req’d):

• 4G State of the Union
• Metered Mobile Data Is Coming and Here’s How
• Mobile Broadband: pricing for profits

Post and thumbnail photos courtesy of Flickr user abbynormy

Indeed, carriers weren’t prepared for the explosive growth in mobile broadband use over recent years. Jump back in time prior to the debut of Apple’s iPhone in 2007, and you’ll see an industry focused on voice and messaging revenues; data was only useful to the mobile enterprise worker and early adopters using smartphones with clunky browsers that made mobile web surfing less appealing than a root canal.

With the first iPhone and credible competitors that followed came a growing appetite for data: first in the browser, later with mobile apps and the desire to use phones as mobile hotspots. Indeed, as noted in a GigaOM Pro report on the data traffic tsunami (subscription required), iPhones were the primary driver behind the 5,000 percent increase in data usage on AT&T’s network from 2007 to 2010. With the relatively recent addition of 3G tablets and MiFi devices, demand for data will only keep growing, yet the problem remains: How can carriers predict data use and manage the service revenues against operating costs?

Tiered data plans are one answer when compared to the old model. Until this year, consumers generally chose one of two data plans: a low-capacity option at roughly $25 to $30 (200 – 250 MB) or a $50 to $60 “unlimited” plan that typically provided 5 GB of data. The latter plan appears to be a better value on a cost-per-MB basis, but it also introduces a problem for network providers, because the capacity variance between the two plans is very wide and gives little room for operators to predict demand accurately. As a result, AT&T began the shift away from unlimited plans earlier this summer, and others are following.

Industry analyst Chetan Sharma sums up the problem best in his report, Managing Growth and Profits in the Yottabyte Era:

Operators typically have great intelligence on voice usage but for data, the infrastructure and efforts are generally not on par. There is little understanding of what consumers are doing, which applications and services they are tuned to at any given instant, forecasting traffic spikes, etc. As cost of supporting data services exceeds the cost of managing voice services and as the revenues from data services become more prominent than those from voice services, operators will have to pay much more attention to the specifics at a very granular level and design business models and pricing plans per the trends and forecasts.

Consumers may want unlimited plans, but carriers can’t provide them due to limited infrastructure. Increasing demand will outpace fixed supply as carriers invest more to keep up with the growing wave of data demand. The investments to try to do so aren’t outpacing revenues from demand; if you can get past the jumbled look of the iSuppli chart below, it shows a declining trend of network operating margins over the past year. Consumer demand is the one variable that unlimited plans can’t account for.

A “pay for what you use” model might be the second-most desirable data offering on a consumer’s wish list, but like the unlimited model, it does little to help the operators predict demand in advance. Only by looking at prior usage habits can the carrier even try to predict how much data a consumer may use in the future. And all it takes is a new, hit social networking service or popular, must-have bit of software to throw that predictor model out of whack.

This topic of managing network capacity and data demand is particularly timely given that T-Mobile is the latest U.S. carrier to offer tiered data plans. Om recently spoke with Neville Ray, T-Mobile’s CTO, about the plans and Om’s vision of the Gigabyte phone. Ray agreed that it’s just a matter of time, saying, “You’re seeing some of these very capable smart phones emerge, and in several cases, we’re seeing devices where data consumption will be greater than a gigabyte per month.”

It’s good to hear that carriers are thinking forward, both in terms of demand and the best plans to offer and predict usage, so a migration to tiered data could fit the bill for all parties involved with both 3G and 4G networks. (Related: see our mobile broadband guide to understand how data networks are currently migrating to 4G). The only remaining variable is how will consumers take to such limits? Yes, the plans will be marketed as potential cost savers for some, but the heaviest of users will balk at having to refill their data bucket several times a month. Of course, as that happens, we should see a very different margin chart in as little as six months from now, much to the carrier’s delight.

Related GigaOM Pro Content (sub req’d):

• Everybody Hertz: The Looming Spectrum Crisis
• The Internet of Things: What It Is, Why It Matters
• 4G: State of the Union

Mobile Business Briefing reports that Francisco Varela, head of platform partnerships at YouTube, detailed the dramatic growth YouTube has seen on mobile platforms and also talked about the future of YouTube’s new web-based mobile experience. According to that report, Varela also said that a new version of its mobile website, called Blazer, would provide mobile users with a faster and better search and viewing experience on mobile devices.

The Blazer mobile web platform, which was launched earlier this summer and available on Apple iOS and Android mobile devices, seeks to shift viewing away from dedicated mobile apps on those operating systems.

But YouTube can’t go it alone. It wants to make its videos available to users in a way that doesn’t crush operators’ data networks or drive them to institute tiered data plans like the one AT&T pushed on its users earlier this year. As a result, Varela was there at the conference seeking partnerships with mobile operators, according to RCR Wireless.

YouTube is clearly one reason Google veered from its net neutrality stance for wireless network when it struck a compromise with Verizon early this year that said carriers may need differentiated traffic management on mobile networks. YouTube requires a higher quality of service than other applications to ensure that users get a good video experience, and Google is trying to solidify its ability to provide that user experience by striking deals to manage traffic.

Another key to YouTube being able to continue offering a high-quality user experience is ensuring that users watching its videos won’t get hit with excess charges due to new tiered mobile data plans. Mobile Business Briefing quotes Varela as saying:

“Mobile broadband services need to echo and support the growing demand among users for immediate mobile access to high quality video from a vast library of content at an affordable price. YouTube –- together with a compelling operator data plan — can provide that through the Blazer platform.”

In other words, Google wants to ensure that mobile operators price their data plans in a way that won’t be prohibitive to users actually viewing video on YouTube. To do this, Google may seek to have traffic originating from YouTube excluded from wireless customers’ data usage. Welcome to the fragmentation of the mobile web.

Of course, partnerships with mobile operators won’t just benefit YouTube or its mobile viewers by rolling out traffic management on their mobile networks. Between the first and second half of 2009, mobile video traffic rose 99 percent, according to data from Allot Communications released earlier this year. The thinking goes that traffic management could help alleviate some of the pain that has arisen not just from YouTube, but other mobile video services as well.

Related content on GigaOM Pro: (subscription required)

• What Mobile Learned From the World Cup
• Mobile Overview, Q2 2010
• Why Carriers Can’t Afford to Wait for New Spectrum