If you rely on News Feed in Facebook to find my posts, you're missing most of them. On average, only 16% of updates in Facebook make it into News Feeds. Let me suggest that you subscribe to me in Facebook, follow me on Twitter (@ccengct), or use an RSS reader.

Readers in the European Union are advised that I don't collect personal data, but the same cannot be said of Google.

Tuesday, December 7, 2010

iPhones and iAngst

Consumer Reports has sharply criticized AT&T Wireless and says that iPhone users, in particular, are dissatisfied with AT&T. Let me explain the inevitability of this.

Until about five years ago, mobile networks -- what we used to call cellular networks -- were designed primarily for telephone calls. Voice can be compressed to a very slow data rate without losing intelligibility. Also the spoken word has a high degree of redundancy, so that occasional drop-outs and impairments can be tolerated.

Data transmission, on the other hand, doesn't tolerate impairments. Worse, the bit rate of a streaming video can be orders of magnitude faster than the bit rate of a voice call. This puts an enormous burden on the mobile carrier to increase capacity if many customers begin to use data.

There are three ways to increase capacity. One is new technology. The migration from 2G to 3G increased capacity, although not without some technical limitations that I won't go into. The second is to acquire additional spectrum; that's a zero-sum game controlled by the FCC. The third is to "split cells", installing new radio equipment, new towers, etc to make the network increasingly dense.

All three of these cost money... big money, as in billions of dollars when extrapolated nationwide. Even a company the size of AT&T has financial constraints. (The wireline side of AT&T is simultaneously rolling out a U-verse product that consumes billions of capital too.) Moreover, AT&T can grow their network only so quickly; they have only so many engineers to do the work, and only so many trained and experienced engineers are in the labor pool.

Meanwhile there is an incessant demand from consumers for data services. Why? Because companies like Apple want to sell new phones into what would otherwise be a stagnant market. Apple doesn't care about AT&T's problems, as long as they don't impact the sales of iPhones. Apple intentionally promotes these data-intensive features to you because they increase differentiation, perceived value, and brand loyalty. So, when you see an advertisement from Apple, you see some really cool stuff and assume that the AT&T network has virtually unlimited capacity. It doesn't.

Companies like Netflix are also anxious to stream video to smartphones; it bypasses the choke-hold that the cable companies and wireline telephone companies have on residential broadband connectivity.

Of course AT&T themselves are not immune to commercial pressures. They signed an exclusive deal with Apple for the iPhone so that AT&T would have an advantage over Verizon Wireless and Sprint, whose networks at this time are even more constrained with respect to data than AT&T's. AT&T also wants to increase revenue from what would otherwise be a stagnant market. Thus they price their data services at something like $30 per month for a residential smartphone. Every new iPhone drives incremental ongoing revenue for AT&T. Selling the phone itself is not profitable, but it does increase "stickiness" because a user must extend his or her contract with AT&T to purchase the highly subsidized phone -- and then gets a phone that cannot be used on another network.

The problem: AT&T cannot increase its capacity fast enough to handle all these iPhones if they are used in a data-intensive way. How data-intensive usage would turn out to be was a matter of conjecture during the product planning cycle. Further, there are concerns that revenue of $30 per month is not compensatory for the capacity that the data features of these phones are actually consuming.

AT&T can bail itself out of this dilemma in two ways. One, it can encourage the deployment of "femtocells" that offload AT&T's network. AT&T is trying to get customers to spend their own money for these femtocells, and to some degree the strategy is working. Two, it can migrate to 4G which will increase capacity by an order of magnitude -- assuming that people dump their 3G smartphones and buy new 4G smartphones.

4G is still a few years away. In the U.S. it's being pushed most aggressively by Verizon Wireless who are desperate to get back into the race. Unfortunately, the slow progress of 4G was a primary factor in Nortel's bankruptcy.

In short, there are understandable reasons why AT&T's smartphone customer base is not happy. There is light at the end of the tunnel, but the tunnel is long and conditions inside the tunnel are not pleasant. My advice:
  • be tolerant,
  • consider getting a femtocell in your home or office, and
  • be ready to switch to 4G when it becomes available... just don't expect it to cost $30 per month.
Or just be content with 2G. I am, even though I'm a geek.