Other Articles by David Willis

LMDS: Is It a Little Too Much, a Little Too Late?, Columnists, February 8, 1999 Switching On Frame Relay SVCs, Workshops, February 22, 1999 Putting Service Levels in Perspective, Columnists, March 8, 1999 Mariposa, 3Com Raise the Bar For Next-Generation ATM Access, Reviews, March 22, 1999 Wireless Phones: Untethered and Unreliable, Columnists, April 5, 1999

Other Colunistst this issue

Business to Business By Brian Walsh On the Edge By Art Wittmann

Company Directory

Browse our directory to get data, starting with a particular company.

Reader Service

Allows you to request additional product information from our advertisers.

Print The Full Article

ClickHere

E-mail this URL

ClickHere

Buy the Book

In 1998, the telecommunications industry saw 33 large consolidations valued at $220 billion, according to Clarus Research. That's created odd hybrid-service providers, where an energy company like Williams Networks is investing heavily in the communications business. Last year, Level 3 Communications, one of the up-and-coming IP service companies, made more money in coal mining than in networking.

And there are no signs of the consolidation slowing down anytime soon. Large companies are getting larger, buying up innovation as a sure way to excite investors. The theory goes that small companies come up with new products, make mistakes and sometimes succeed. Big companies buy out the small companies with their highly valued stock, making themselves look good in the process. The small companies gain the mature sales and distribution system of the larger companies, and principals make off with a large trunkful of cash and/or stock. Some companies appear to be deliberately built with the goal of getting acquired. Just look at all the terabit router companies, including Juniper Networks and Avici Systems. How can these single-product companies expect service providers to bet their networks on this stuff?

Consolidation is a natural progression for a mature industry, and networking is simply behaving like other market sectors. From banks to trash collection, from soft drinks to health care, from toys to tobacco, the market is controlled by a few large players. As conglomerates, these companies find themselves with a broad mix of products and customers. Yet customers are often confused about what has happened.

Now we've entered the era of the networking oligopolies. Cisco, Lucent, Nortel and 3Com have become the unstoppable forces on the equipment scene, with AT&T-TCI, Bell Atlantic-GTE, SBC-Ameritech and MCI WorldCom calling the tune in services. Conservative network managers are hitching up with these companies because they appear to be the safest havens in a dangerous and complicated world, in the same way that corporate strategists got lazy and began to standardize around Microsoft, for better and for worse.

Naturally, scale benefits both the vendor and the consumer. In communications, there is a compelling desire to hand over end-to-end responsibility to a single provider. But putting all of your networking eggs in a single service basket is foolish. As a buyer of services, you should always maintain a second service supplier as a hedge against lackluster service.

The Business of Big Business Mergers and acquisitions work is an industry unto itself, with thousands of consultants clamoring to help companies smooth the transition into a single organization. Online clearinghouses such as MergerNetwork.com match buyers and sellers like they were trading baseball cards. But the fact is that many consolidations fail to deliver on their promises. The American Management Association states that only 15 percent of deals achieve their financial goals, as measured by return on investment, profitability and share value.

Customers often take the hardest hit. They rarely get a clear view of what the future holds for strategic products and services. Occasionally, plum products just disappear. Take, for example, ProTools. Purchased by Network Associates (then Network General) for $40 million in 1994, the ProTools analyzer supported real-time, on-the-fly protocol decodes to a GUI. Cool stuff. But don't try to find anything like it from Network Associates. The company only recently produced a Windows-based analyzer (courtesy of another acquisition), but it still doesn't do packet analysis on the fly.

As merging companies try to understand and meld with each other, a heated behind-the-scenes battle ensues over which products are going to grow and which will fall by the wayside. They don't always make the right decision--for example, for two years in a row we rated 3Com's AccessBuilder 4000 higher than USR's Total Control products. Guess which product survived the 3Com-USR merger? Hint: Don't ask 3Com for the AccessBuilder 4000.

A Little of This, a Little of That When products do emerge from a merger or partnership, sometime they are a cobbled-together mess. Cisco and Livingston partnered on the Cisco 1020 Dial-on-Demand Async router and tried its best to build Cisco's hallmark IOS interface into the box, even though the underlying engine wasn't compatible. When system administrators complained that IOS commands didn't make the 1020 behave exactly the way other IOS-aware products did, Cisco retired the product.

If you look closely, you can tell which products are strategic to a vendor's mission and which are just placeholders. But the placeholder products often have loyal customers. As an example, I'll predict the popular Cisco 700 Series of ISDN routers will get the ax within the next 18 months or so. These devices are a legacy of Cisco's 1995 acquisition of Combinet and run a command language that is not IOS but close enough to it to be annoying when it diverges. With the new 800 series running IOS for a few dollars more, the 700s have likely lost their strategic value to Cisco.

The most frequent problem associated with an acquisition is the way that product improvement and support stall while the company overcomes integration shock. Last year, Lucent acquired Yurie Systems, an ATM access concentrator company, for $1 billion--a figure even Lucent might consider to represent real money. Yet, a year later, Yurie's flagship product, rebranded as the Lucent AC 10, hasn't changed. Our 1999 test units still ran code developed during the Yurie days, and a pack of competing products have left it in the dust. Lucent technical support passed us around like a hot potato because its database couldn't classify the AC 10 as a voice or data product (it's both). If I had been a Yurie customer, I'd be...well, a Mariposa customer by now.

Of course, acquiring companies isn't simply about purchasing products. Intellectual capital is the hot commodity. While engineers are the real asset, keeping them from being distracted during a takeover is next to impossible. The brightest stars often leave or are placed in odd positions in an effort to make something--anything--work.

I wonder if two years ago Marc Andreessen could have imagined himself as the CTO of America Online.

A renegade freethinker with a 20-person startup just isn't going to fit comfortably as a cog in a 20,000-employee machine. Small, simple, fast-moving organizations don't mix well with large, complex, bureaucratic ones. The battle to innovate among the big four networking vendors is heating up, and it will be fought and won based on how well they can integrate companies without screwing them up. Indeed, the best way to keep an acquired company moving is to keep the parent company out of the way.

Yet, more often than not, the top priority for the acquiring company is to indoctrinate and incubate. Products sit around while they get "rebranded." In the case of 3Com's recent acquisition of NBX, it will take six months to get NBX's IP PBX product into the SuperStack form factor and to retrain the sales force, as if any customers have been screaming for NBX to make the darn thing stackable.

As engineering priorities are reworked, products, such as Class Data's policy server, are likely to die or change entirely. When I met with Class Data officials a couple of years ago, their focus was on an end-to-end IP class of service management with a multivendor twist. It would happily work with a range of proprietary switch and router products, and expose a single interface on which you could manage it all. It was a kind of meta-manager for QoS. Since Cisco acquired the company, the focus is on Cisco-specific and standards-based QoS mechanisms only. So if I were a Class Data customer, I'd be...well, doing proprietary QoS management now.

What do we learn from all this? When the maker of a product that's critical to your company gets acquired, make yourself be heard--loudly. Don't settle for what the frontline salesperson tells you. He or she will be programmed to assure you nothing is wrong, though it's likely this salesperson doesn't really have a clue as to what's going to happen. Contact the acquiring company fast, as high up the ladder as you can get--let the top brass know you exist. Tell them how important the product is to your operation. If ever there were a time to show your hand and reveal your plans for future purchases, it's now. And just to be safe, start shopping around.

Send your comments on this column to David Willis at dwillis@nwc.com.