During the press conference Apple called last week to defend the new iPhone, CEO Steve Jobs emphasized that "antenna-gate" was an industrywide problem. Lots of other phones, Jobs explained, suffer diminished reception when you hold them the wrong way. Unsurprisingly, Apple's dig prompted an immediate response from Jim Balsillie and Mike Lazaridis, the co-CEOs of Research in Motion, the company that makes the BlackBerry. "Apple's attempt to draw RIM into Apple's self-made debacle is unacceptable," the CEOs said in a statement. The executives went on to describe their company's extensive antenna research, and explained that RIM has "avoided designs like the one Apple used in the iPhone 4 and instead has used innovative designs which reduce the risk for dropped calls, especially in areas of lower coverage." They added, "One thing is for certain, RIM's customers don't need to use a case for their BlackBerry smartphone to maintain proper connectivity."
It was a strong, punchy statement from a company that still maintains the lead in smartphone marketshare. There was just one problem: RIM hasn't revealed any data to prove that its phone designs actually "reduce the risk for dropped calls." On Tuesday, I e-mailed RIM's press office to ask for that data. Would the company tell me how often some of its best-selling phones drop calls? Apparently not. RIM's media team acknowledged receiving my request, but as of Thursday morning, the company hadn't gotten back to me.
I also called Motorola to ask about the dropped-call rate for its phones. In a statement last week, co-CEO Sanjay Jha said, "In our own testing we have found that Droid X performs much better than iPhone 4 when held by consumers." That sounded promising—if there are actual tests showing that the Droid X drops fewer calls than the new iPhone, wouldn't the company be proud to show off the results? Again, apparently not. Motorola's media-relations team didn't acknowledge my repeated requests for dropped-call data.
OK, one more try: I reached out to HTC, the Taiwan-based company that makes a range of phones based on Google's Android operating system, including the EVO 4G. * Hui-Meng Cheng, HTC's chief financial officer, told the Wall Street Journal this week that the reception problems Jobs pointed to "are certainly not common among smartphones"; another company rep told the tech blog Pocket Lint that only 0.016 percent of customers had complained about reception on the Droid Eris, a lower complaint rate than that of the iPhone 4. But when I asked HTC for dropped-call data on its phones, Rickey Bird, a spokesman, told me, "We do not make those numbers public." When I asked why not, Bird said that the decision was "in line with the rest of the industry."
He's right about that. As far as I can tell, none of the leading phone makers publishes information about the number of calls its devices drop during actual use. This is not for lack of data, either. Bird acknowledged that wireless carriers provide HTC with the dropped-call rates on its phones. Representatives at several wireless companies confirmed this; the carriers constantly measure how well specific devices perform on their networks, and they send detailed dropped-call information to the manufacturers. But neither the phone makers nor the carriers want to make that data public—and they won't say why, either.
As I argued last year, this kind of secrecy is one of the main reasons wireless service in America lags the rest of the world. These days, it's possible to find accurate performance data for most of the major purchases we make in our lives. If you're shopping for a car, you can find out its gas mileage. If you're shopping for a plane ticket, you can look up each airline's on-time rate. When you go looking for a new cell phone, though, you enter a data-free zone in which every company is free to claim that its devices offer spectacular service. If customers or the media disagree, the companies can argue—as Apple did last week—that the critics are just carping, because nobody has any definitive data that can prove them wrong.
Should we believe RIM, Motorola, HTC, Nokia, and other phone manufacturers when they claim that their phones offer better reception than Apple's? Not blindly. After all, Apple was recently boasting that the "iPhone 4's wireless performance is the best we have ever shipped"—a claim that fell apart when Jobs, to his credit, brought out actual dropped call data at his press conference, data that showed the new iPhone drops more calls than last year's model, the 3GS. If we're now questioning Apple, we should also question everyone else. While I commend Apple's rivals for taking advantage of a good PR opportunity, I wouldn't put too much stock in their statements. If you want me to believe the BlackBerry Bold or the Droid X or the HTC EVO really drops fewer calls than the iPhone, prove it to me. Otherwise, just stop talking.
Bird, the HTC spokesman, suggested that I call up wireless carriers for more specific data on cell phone reception. So that's what I did, and, indeed, I had more luck here. Of the four major carriers, only T-Mobile ignored my request. The other carriers were slightly more forthcoming. Tom Pica, a spokesman for Verizon, declined to tell me the rates of dropped calls on the company's service, but he did discuss Verizon's elaborate testing procedures for its network. Verizon has more than 100 test vehicles that are packed with phones that work on Verizon and other companies' networks; the vehicles drive around the country, and the phones constantly place calls and make data connections to all major networks. (It's basically a more sophisticated version of Verizon's "Can you hear me now?" ads.) "But we don't get into the specific results of those tests," Pica told me, other than to claim that Verizon is "the nation's most reliable wireless network."
Both Sprint and AT&T went further. A Sprint spokeswoman told me that its average "call error rate"—which combines the rate for dropped calls with the rate for "blocked calls," meaning calls that a user tried to place that never went through—is lower than 1 percent. That rate comes from "internal performance metrics," the spokeswoman said. Meanwhile, Mark Siegel, a spokesman for AT&T, said that "on average, our dropped call rate is in the low 1 percent range—in other words, slightly greater than 1 percent." (In a statement to the tech blog Boy Genius Report, AT&T said the precise rate was 1.44 percent.) This dropped call-rate was derived from drive-tests like the one Verizon conducts, Siegel said. He went on to explain that AT&T's dropped-call rate was within two-tenths of a percent of the rate of the "market leader" in the industry, which is generally recognized to be Verizon. (Siegel dismissed the findings of a recent customer survey that showed AT&T dropping 4.5 percent of calls, much higher than others in the industry; he argued that because the survey was based on customers' self-reported impressions of the network, it was likely less accurate than the data found in AT&T's drive tests.)
Neither AT&T nor any other company would give me dropped call rates for specific devices. Those stats, Siegel said, were "very competitive information," and he argued that they probably wouldn't even be very helpful to customers. "We're talking about very small gradations of difference," Siegel said. "I'm making these numbers up, but if you said, this phone has a call-completion rate of 99.2 percent, versus this one which has a call-completion rate of 99 percent, is that a meaningful difference to customers? I'm not sure how useful that would be for someone to know."