History keeps trying to teach us lessons, but sometimes we are slow to grasp them. Speaking at the UBS Annual Conference last week, Ralph de la Vega, President and CEO, AT&T Mobility and Consumer Markets, said that "a small number of heavy users are hampering network performance" for the rest of its customers. ... About 3 percent of [AT&T's] smart phone customers are driving about 40 percent of the [wireless data] traffic." He wants to convince those heavy users to reduce their usage "voluntarily" or he will start charging them more: "Longer term, there's got to be some sort of pricing scheme that addresses usage"
Where have we heard this before? In 1991, early online services company Prodigy reportedly discovered that "20 percent of the users were sending 90 percent of the email messages, costing the company millions of dollars with a mail flow growing 20 percent a month." Prodigy characterized these heavy users as "abusers" and revised its formerly unlimited flat-rate pricing: for the $12.95 monthly fee, each household now got 30 emails per month and was charged 25 cents for each additional message. The result was probably the first nationwide protest organized largely online. (Prodigy later abandoned flat-rate pricing in favor of three options, ranging from $7.95 to $19.95, combining monthly fees and hourly usage charges.)
Mindful of that lesson, we (Sandy & Dave) spoke about the coming opportunities for high-speed data services at several conferences in 1996. We pointed to Prodigy's experience and suggested that while customers prefer flat-rate pricing, a single flat-rate plan would inevitably lead to "run away" applications with a few users consuming most of the resources. We suggested tiered pricing alternatives to accommodate the needs of both light and heavy users.
But marketeers love the simplicity of flat-rate pricing. Our message went unheeded, and flat-rate pricing became the norm for consumer data services. Over the years, we have followed the attempts of broadband service providers--both cable and telco--to recoup the costs for usage that they view as well beyond the norm. For example, in April, 2009, Time Warner Cable announced tests of "Consumption Based Billing" in several markets. This resulted in an uproar that drew the attention of Congress. Subsequently CEO Glenn Britt put the trials on hold, saying "we will not proceed with implementation of additional tests until further consultation with our customers and other interested parties". The company said it was postponing tiered pricing until it could better educate customers on the new fee structure.
So here we are once again, this time with AT&T and iPhone flat rates. De la Vega told his audience that AT&T will address their problems by first making customers aware of how much wireless data they're consuming and then charging heavy users more if they don't cut back their usage. It sounds a lots like what Time Warner decided--first educate users on how much they are using and then charge the very heavy users if they don't cut back. It will be interesting to see how customers who were sold iPhones under the premise of unlimited data usage react to this revised offer.
So what have we learned from this?
Why don't we ever learn?
The situation with the cable operators was not so surprising. In their world, it didn't matter if one person or a million were watching a TV program. They didn't think in terms of Erlangs -- indeed, most of them had never heard the word.
But AT&T is steeped in traffic engineering. It's hard to understand how they fell into this trap. They're likely to find that it's hard to get out of it.