Information Processing: TELECOMMUNICATIONS
QUALCOMM: NOT EXACTLY AN OVERNIGHT SUCCESS
After a decade, Qualcomm's wireless standard takes off
Qualcomm Inc. Chairman Irwin M. Jacobs jokes about it now. But at the time, the rocky debut of his pioneering wireless technology was anything but funny. He had cajoled 300 industry leaders into showing up at his tiny San Diego company while outside, workers were loading a 30-pound, PC-like gizmo with a phone receiver dangling off the side into a big white van.
As Jacobs wrapped up his talk on the wonders of his digital wireless technology, he saw a partner waving frantically. The message was unmistakable: Keep talking. So Jacobs did--for 25 minutes, while engineers worked furiously to get the massive phone hooked up. "We were sunk once we let people outside that room," recalls Jacobs.
The demo, nearly a decade ago, did work eventually. But Jacobs has faced countless gut-wrenching moments as Qualcomm emerged the unlikely leader in the market for digital handsets for the new wireless personal communication services (PCS). As master pitchman for Qualcomm, Jacobs has spent the past decade stumping for his baby, a technical standard called code-division multiple access (CDMA) that has managed to far outpace the original digital standard chosen by the cellular industry, time-division multiple access (TDMA).
INTERFERENCE. Qualcomm's battle for industry acceptance is paying off. CDMA was long criticized for being too complex in the way it divides digitized voice and data into about 10 different time slots on the same radio wave. TDMA uses three time slots, trading greater capacity for ease of design. But the market went for capacity: Some 57% of a new generation of digital wireless systems being built will use CDMA. Another 24% will use a European standard called GSM (for general standard for mobile communications)--leaving TDMA with just 14%. Jacobs has "overcome what should have been fatal disadvantages with brilliant public relations," says Herschel Shosteck, an industry economist.
The numbers tell the story. Qualcomm's revenues more than doubled, to $814 million, in 1996, and analysts project sales will pass $2 billion this year. PrimeCo, an alliance of three Bell companies and AirTouch Communications, and Sprint PCS, a partnership of Sprint Corp. and three cable companies, have agreed to buy a combined $850 million worth of handsets from Qualcomm and partner Sony Corp. over the next two years. Orders for handsets and other equipment from China, Korea, Russia, and Chile should add at least another $500 million in sales.
But Jacobs isn't home free. Qualcomm only makes CDMA phones, while its much larger rivals will soon start building to all three standards. And Qualcomm is hitting legal hassles. Just hours after Qualcomm introduced its sleek new Q phone in March, Motorola Inc. slapped a patent-infringement suit on the company, claiming the phone copied Motorola's pocket-sized StarTAC. Ericsson filed a similar suit. Qualcomm denies wrongdoing, but its stock plummeted from 63 to 45, only recently rising to around 52.
No question, Jacobs is playing with the big boys now--and that's a long way from his roots as a defense-industry engineer. Jacobs, 63, earned an engineering degree at Massachusetts Institute of Technology. In the 1960s, he and MIT classmate Andrew J. Viterbi formed Linkabit Corp. to develop a military satellite communications system. The pair founded Qualcomm in 1985. They soon focused on the cellular industry's conversion from analog to digital. But the U.S. industry had already agreed to adopt TDMA as its digital standard. In 1989, Jacobs walked into the Washington offices of the Cellular Telephone Industries Assn. (CTIA) to a frosty reception. "We had just finished a bruising debate and had made up our minds on another standard," recalls Kevin Kelley, a former CTIA member and now a Qualcomm senior vice-president. Says Jacobs: "They threw us out on our ears."
FLAME WARS. Jacobs persevered, convinced his compression technology would increase cellular networks' capacity as much as fortyfold, vs. the three or four times offered by TDMA. One of the few to listen was Pacific Telesis' wireless division, now AirTouch Communications Inc. It gave Qualcomm $2 million to build a trial network in San Diego.
The results of the 1991 trial were impressive enough to convince the CTIA to reopen the standards debate. In 1993, the group chose CDMA as a second standard, setting off a firestorm. McCaw Cellular Communications Inc. (now AT&T Wireless) and SBC Communications Corp. had already sunk millions into TDMA. A torrent of attacks quickly showed up in the trade press, at conventions, and on Internet forums: CDMA was too expensive, CDMA would be jammed, CDMA was too complicated. Jacobs was labeled everything from a "snake charmer" to a fraud.
Despite the brouhaha, Northern Telecom Ltd. and Motorola decided to license CDMA technology. The move was little more than a hedge, but it gave Qualcomm credibility. Jacobs and his team next went to Asia looking for partners. But wherever they went, they were preceded by a flood of Ericsson "white papers" warning of CDMA's problems.
Critics insist Qualcomm deserved the scrutiny. "They promised way beyond what they could deliver," charges Shosteck. True enough, CDMA's compression abilities are proving to be only about one-seventh of Jacobs' early claims. But it still offers more than twice the capacity of TDMA and GSM. That advantage won over Sprint PCS and PrimeCo: both signed on with CDMA in 1995 after extensive testing. "Irwin was very convincing," says Keith Paglusch, a senior engineer with Sprint PCS.
Jacobs immediately faced another problem: No one made CDMA handsets. But Sprint and PrimeCo needed huge numbers of them for their national rollouts, scheduled to start that fall. Jacobs saw an opportunity. Sony agreed to team up with Qualcomm, investing 49% in a phone-making venture in exchange for its mass-production knowhow. By mid-1996, Qualcomm was scrambling to fill orders from a factory created from a former medical-supply facility.
As deadlines approached, plenty went wrong. A tractor-trailer bound for the East Coast with thousands of Qualcomm phones was halfway across the country when Bernie Bianchino, Sprint's chief business-development officer, received a late-night call: A software glitch meant every phone had a defective menu screen. The truck made a U-turn and raced back to San Diego, where each phone was reprogrammed.
For PrimeCo, the problems were worse. Ten days before its national rollout, engineers discovered an ear-piercing noise. "You'd hit the send button, and this god-awful screech would just blow the phone right out of your hand," recalls Lowell McAdam, PrimeCo's chief operating officer. The problem was traced to the handset's software. But with 40,000 defective phones sitting in a Florida warehouse and the national rollout just days away, there was no time to ship the phones back. Instead, a Qualcomm engineer flew out with the fix.
CRASH LANDING. When he arrived, "he was like the guy showing up with the ice chest for an organ transplant," recalls McAdam. For four days, technicians worked around the clock reprogramming 40,000 handsets with the software fix before shipping them to retailers via Federal Express.
These days, the Qualcomm factory is humming and Jacobs has plans for factories in Asia and Latin America. He can't afford any more glitches: Motorola and a host of other phone makers are scheduled to flood the market with their own CDMA products later this year. Jacobs vows to stay in front of the pack with feature-rich handsets. Given his track record, he may do it--but don't assume he'll do it easily.By Eric Schine in San Diego and Peter Elstrom in ChicagoReturn to top