/

Cisco: Are the deals worth it?

Most of Cisco's acquisitions have been successful, but there have been a few duds. 8/03/98

By Jim Duffy

Over the past five years, perhaps no other vendor has been on a buying spree as zealously as Cisco.

Cisco has spent about $7 billion to acquire 25 companies, from LAN switching pioneer Crescendo in 1993 through WAN switch leader StrataCom in 1996 to NetSpeed this year. The purchases are intended to broaden Cisco's product portfolio and deepen its research, development and marketing expertise - especially in the service provider arena. The company is attempting to jettison its "router only" persona and solidify a position as the leading supplier of network products for the Internet and intranet markets.

For the most part, Cisco's acquisitions have been successful. The company receives kudos from Wall Street and other observers for striving to maintain the independence and entrepreneurial cultures of the companies it buys. Cisco also targets companies close to its San Jose, Calif., headquarters, believing that adopting close to the mother ship enhances synergy. For companies on the East Coast, Cisco's expanding campus in Research Triangle Park, N.C., serves as "Cisco East."

"The first level we are interested in is the teams and the technologies they're working on," says Ammar Hanafi, manager of business development at Cisco. An acquired company is expected to prove its worth to Cisco within six to 24 months, Hanafi says. Longer term goals include tightly integrating the acquired team and technology into the operational fabric of Cisco.

And two-thirds of the CEOs from the acquired companies are still with Cisco, Hanafi adds.

"We don't call out particular transactions as being successful or unsuccessful," he says. "We don't go into these transactions expecting [a] 100% success rate. If we did that we'd be pretty cautious and ineffective. I can't think of a particular situation that was a disaster, mistake or bad deal." To date, Cisco's LAN switching acquisitions are considered the most successful by analysts and other observers. Crescendo and Kalpana, which Cisco acquired in 1994, are the sources for most of Cisco's popular Catalyst line of switches. These two acquisitions alone have helped propel Cisco to leadership in Ethernet switching, with 37.3% of the $4.7 billion 1997 worldwide market. Cisco ranks far ahead of 3Com, which is No. 2, with 15.7% market share, according to The Dell'Oro Group in Portola Valley, Calif.

"Those acquisitions have just been home runs," says Scott Heritage, an analyst at Warburg Dillon Read LLC in New York.

And the $4 billion acquisition of StrataCom in 1996 gave Cisco instant leadership and credibility in WAN switches for enterprises and service providers. Yet analysts say growth of the StrataCom business has not met expectations.

Sales of StrataCom gear to service providers lagged in 1997, analysts say. Some of that lag may have been due to a lack of cohesion between Cisco and StrataCom products and operations; and some was due to sales lost to Ascend Communications and Newbridge Networks.

"[Router/WAN switch integration] is not what [Cisco was] saying two years ago when they bought [StrataCom]," says Craig Johnson, an analyst at The PITA Group in Portland, Ore. "There are still two heads inside that beast."

Some of that is intentional, Johnson believes, because Cisco does not want to upset sales of its higher margin 12000 and 7500 series routers. Cisco's StrataCom business will face even more pressure when shipments of WAN routing/switching products from Lucent Technologies and Nortel ramp up, he says.

Indeed, not all of Cisco's acquisitions have been rousing successes.

Cisco killed LightStream's only product, the LightStream 2020 enterprise WAN switch, shortly after acquiring StrataCom. StrataCom's IGX switch quickly replaced the LightStream 2020 despite Cisco's insistence that there was no overlap between the StrataCom and LightStream product lines.

Cisco's purchase of Gigabit Ethernet pioneer Granite Systems in 1996 is also one of the industry's great mysteries. Many observers believe Cisco has been less than candid in its status reports on the benefits of the Granite acquisition and on Gigabit Ethernet product development. Suspicion was heightened when Cisco, after repeatedly stating it would wait until the Gigabit Ethernet standard was complete before shipping products, released products based on Granite MAC and PHY chips well before the standard.

Those products - switch modules and uplinks for the Catalyst 5000 line - disappointed industry analysts. Many were expecting the Granite technology to result in a new generation of Layer 3 switching Application Specific Integrated Circuits (ASIC) for wire-speed Gigabit Ethernet crossbar routing switches that could scale enterprise networks up to and beyond 100G bit/sec - a project Cisco sources referred to as "Milan."

But when Cisco rolled out the Catalyst 8500 line in April, which uses Layer 3 switching ASICs from MMC Networks, some observers saw it as a confirmation that the Granite acquisition was a wash and maybe Milan was too.

"[Granite was] never intended to [provide] the full-fledged Layer 3 ASICs for current or future products," says Jayshree Ullal, vice president of enterprise marketing at Cisco. So even though it took awhile for Cisco to deliver the goods from Granite, that acquisition and the StrataCom acquisition have been successful, according to Hanafi.

Observers are also waiting for Cisco's acquisition of the Dagaz business of Integrated Network Corp. to bear fruit. Dagaz makes digital subscriber line gear for enterprises and service providers in international markets. Yet Cisco acquired another DSL company - NetSpeed in Austin, Texas - after Dagaz. Yet Cisco released NetSpeed-based products before Dagaz-based gear.