The article, which focuses on Cingular and its Austin, Texas call center, found an industry that is becoming interested in pleasing -- and holding onto -- its customers. The research firm Yankee Group told the Times that 20% of mobile phone customers (45 million) switch phone companies each year. Gartner's Esteban Kolsky noted that the old barriers to switching providers, like losing one's phone number, are gone. With fewer penalties for leaving, bad service is driving a lot of provider shuffling.

I asked VocaLabs' Peter Leppik to comment. His company has been doing quarterly customer service rankings of the big mobile phone companies for the last two years, and he told me he has noticed a mild trend toward better service, but he called it "very small and very gradual."

But he said it isn't just the phone companies: "It is true that mobile phone companies have a reputation for bad service, but on the whole they're not any worse than banks, travel companies, etc., all of which have deserved reputations for lousy service."

And while the New York Times found Cingular at the bottom of most customer service surveys, Leppik points out that they weren't so bad until they merged with AT&T Wireless. "Cingular is not actually bottom-of-the-barrel for bad service. That honor has gone to AT&T Wireless and SprintPCS in most of our quarterly reports. But with the Cingular/AT&T Wireless merger, AT&T Wireless is pulling the overall Cingular score down."

Both Cingular and Verizon plan on using automated post-call surveys soon. VocaLabs does live post-call surveys, which Leppik thinks are much more effective: "The end-of-call survey is a really bad way to survey callers, and we refuse to do that kind of survey. The problem is that unhappy customers usually hang up before getting to the survey, so you only wind up surveying the happy customers. This is actually worse than useless, since it makes it look like you're doing just fine even when the wheels are coming off the bus."

As far as Leppik's concerned, it's a start, but not a good one. "The simple fact that they're using automated end-of-call surveys is a pretty good sign that they still don't get the idea of better service, since the only reason to use that technique is because it's the cheapest way to do surveys. It's the kind of thing you do if your boss says 'Go do some surveys, but I'm not going to give you a budget.'"

Dr. Jodie Monger, a consultant and frequent contributor to our magazine has some similar thoughts. In an article called "Are You Guilty of Survey Malpractice?" she wrote for our upcoming May issue, she said: "Research is a science as is the application of the results. Garbage in equals garbage out. Dirty data is the equivalent of garbage for your measurement program. Considering the implications to personnel management and the overall management of the center and its ultimate value as the relationship manager for the company, it is better to do NO surveys than to do poor surveys."

Jodie Monger spoke about "dirty data" at our Seattle ACCE show last fall. Go here to watch a three minute video about difference between dirty and clean data.

Leppik concedes he may be too hard on the phone companies though: "I suspect most people who do end-of-call surveys simply don't realize the shortcomings of the method, and they get sold on it by a vendor. Several IVR companies have started doing these as upsells: 'Here's your new speech recognition system. You want surveys with that?'"