The fastest way to get media attention in the telecommunications business these days is to offer cheap Internet telephony rates. It worked for Denver-based Qwest Communications Corp., when it announced plans to roll out nationwide 7.5 cents-per-minute long-distance calling over its Internet protocol-based, fiber optic network in December 1997.

In the wake of that announcement, the company has struck deals to acquire LCI International and EUNet International. The deals mark the start of an acquisition strategy that analysts believe will position the former carrier's carrier to compete globally with the likes of WorldCom Inc. and AT&T.

"They are a WorldCom wannabe," says Sim Hall, vice president of research for Fairfax, Va.-based research firm Action Information Services. "They have the same strategy-keep the stock price high and grow through rapid acquisition."

Alternatively, Qwest would make an attractive takeover target for any of the Bell regional holding companies once they meet federal requirements to compete in long-distance.

Qwest's bread-and-butter business is constructing fiber optic communications systems for established and emerging carriers. Its long-haul capacity clients include GTE and MCI.

But the company and its billionaire chairman, Philip F. Anschutz-owner of 87% of the company's stock-are no longer satisfied with being a carrier's carrier. The start-up has retained most of the rights of way needed to build its 16,000 mile, coast-to-coast, fiber optic network. The $2 billion network is to connect 125 metropolitan areas by the end of the first quarter of 1999 (see table on page 30).

Qwest's ambitions are not limited to the U.S. market.

Its acquisition in March of Amsterdam-based EUNet International-one of Europe's biggest Internet access providers with operations in 14 countries-marks the start of what is expected to be an aggressive international expansion. Plans are also in the works to extend service to the U.K. More than 3500 miles of the network there are operational.

"This is a global business model," says Lewis Wilks, president of Qwest's Business Markets group. "By the nature of a network that leverages IP technology as one of its inherent characteristics, you are defining a model that has no geographic boundaries."

In building its network, Qwest may have an initial edge over existing carriers because the company will use the latest OC-3 (156 Mb/s), OC-12 (622 Mb/s) and OC-192 (10 Gb/s) networking technologies, one analyst says. Although more established carriers also are moving to these infrastructure platforms, they must contend with legacy systems in which they have significant investments.

"Qwest is working with a brand new network that has been built from scratch with all new fiber and high-speed equipment-switches and software," says Robert Gadient, senior consultant with South San Francisco-based Ryan, Hankin and Kent.

The Qwest network consists of 48 fibers with the capability of adding 10 times that many through additional in-place conduits. Each fiber carries eight wavelength division multiplexing (WDM) windows, each with a bandwidth of 10 Gb/s. Qwest officials like to boast that the 2 Tb network will have the bandwidth to transmit the entire Library of Congress coast-to-coast in 20 seconds.

Qwest officials also point out that the company's network is contiguous-it never uses a third-party carrier to bridge any nodes. This will be a marketplace differentiation because other carriers inevitably will have to patch some traffic over the legacy networks, Wilks says.

"The weakest links in legacy networks will constrain mass deployment of extreme high capacity on a national level," he says.

The net result of Qwest's strategy could wrest long-distance avenues from incumbent interexchange carriers.

"They can really play the role of the spoiler because AT&T, Sprint and MCI were banking on long-distance being the cash cow for investments in global, wireless and the local loop," says Cathy Gadecki, senior broadband analyst with TeleChoice.

That is just not true, says Frank Iana, AT&T executive vice president for network and computing services.

AT&T has upgraded its fiber infrastructure to eight wavelength, eight division multiplexing technology. This upgrade has increased the capacity of the 41,000-mile network by a factor of six.

Although Iana acknowledges that AT&T has faced some short-term capacity problems while its Sonet rings are put in place, he expects those to be resolved soon. In the meantime, AT&T can boast 616 points of interface, and by 1999, it will be able to use resources from its acquisition of Teleport Communications Group. This will give AT&T local distribution in 69 major cities.

By then, the carrier will have 143 switching nodes, the largest frame relay network in existence and an asynchronous transfer mode network that is growing by a factor of 10.

The quality of Internet telephony transmission is still debated, but industry observers are betting that acceptable levels will be achieved soon, at least by companies such as Qwest that can manage the quality of their own networks.

By running IP telephony over a dedicated network, "you can simply allocate enough bandwidth so you don't have any noticeable delay," says Action's Hall.

Of course, Qwest is not the only company with ambitious Internet calling plans. ICG, Englewood, Colo., has just announced long-distance service over its IP network for 5.9 cents a minute, while WorldCom's Internet division, UUNet, is attacking the $92 million Internet fax market.

Hall estimates the traditional carriers could lose millions of customers and as much as $8 billion to Internet telephony between 1998 and 2001 (see figure).

Although the new IP networks will help cut domestic long-distance rates, the big impact likely will be felt on international calls. The average long-distance call in the U.S. costs about 13 cents a minute, but the average international price is 89 cents. Telco revenues per minute on international calls will fall more than 20% annually through 2001, Hall predicts.

Qwest's Wilks agrees that a great opportunity exists for IP telephony in the international market, but he disagrees with the view that it has limited potential for sustainable advantage domestically.

"Any time you can add quality, reliability and a fundamental shift in the cost dynamics associated with the way people communicate, [that] changes almost everything [for] almost any geographic [area], including the U.S. market," Wilks says.

Of the many obstacles facing Qwest as it tries to implement its aggressive growth plan, the most onerous will be overcoming internal infrastructure weaknesses.

Its $4.4 billion merger with LCI will help smooth the transition (Telephony, March 16, page 6). Assuming that the two companies manage the merger effectively, the new company will be the fourth-largest telecommunications provider in the U.S. More important, Qwest will gain access to LCI's large, experienced sales staff and back-office billing operations-an integral part of the deal they struck.

That still leaves gaps in Qwest's grand strategy that both Hall and Gadecki say likely will be filled through acquisition. The combination still has no wireless or other local presence, and LCI has focused strictly on voice services.

But Qwest is fleshing out its retail services line.

Qwest announced in late April that it will provide remote access virtual private networks (VPNs), which will give telecommuters dial-up access to their companies' VPNs.

At the same time, Qwest announced that it will provide advanced frame relay and ATM services. The company will have dual OC-48 (2.5 Gb/s) backbones, one native IP and the other native ATM, and the latter will carry frame relay and ATM traffic.