VoIP and snowflakes

The rational for VoIP implementation seems to be as varied as the companies that are going for it.

Columnists and other industry observers often make sweeping generalizations. That’s what we get paid for: to gather information from diverse experience points and boil it down to a focused set of concisely stated conclusions.

But generalizations can get you into trouble. You can find yourself saying that a given technology is going nowhere or that a market is poised to explode, when the truth is a bit more subtle—because between “nowhere” and “explode” lies a huge territory where reality usually takes up residence.

Voice over IP (VoIP) is a prime example. Several years ago, interest in VoIP grew as IP data networks took hold. The idea of bypassing relatively expensive public-switched, long-distance services seemed quite compelling, so prognosticators hyped “convergence” of voice and data over common infrastructure.

Then voice prices fell. With long distance costing pennies per minute, VoIP lost its appeal as a money-saver. Analysts began to sound the death knell, opining that VoIP only made sense for users who needed integrated voice/data applications.

So, is enterprise VoIP really dead? Or is it still a great way to cut costs? Should communications managers dismiss it as a fad? Or would they be foolish to do so?

“IT DEPENDS”

To answer these questions, I looked closely at several real-world VoIP implementations. To my surprise—and, to some extent, my frustration as an erstwhile pundit—I discovered tremendous diversity in these implementations and their business rationales. Every company seemed to have its own reasons for VoIP deployment. And their implementations were as unique as snowflakes.

One organization, for example, had just completed a major infrastructure overhaul that included new cabling. With all that spanking-new infrastructure, the last task the head of the project wanted to do was run multiple wires to each workspace. So, the organization decided that all services (e.g., data, phone, security, video) would share the same cable plant.

Another reason that this organization went to IP telephony was that it had been using its local carrier’s Centrex service. This service was expensive and clumsy for moves, adds and changes, so it was easy to cost-justify new IP PBXs and IP phones.

Not everyone can justify such a wholesale conversion. In another case, a company that had grown by acquisition faced the problem of incompatible PBXs at virtually every location. Rather than spend lots of money on new PBXs, the company installed relatively inexpensive VoIP gateways at each office. These gateways communicated with each other, allowing calls bound to any office to be routed over the company’s wide area network. This eliminated toll charges when people in different offices called each other. The system also allowed the company to centralize its customer-support operations. Customer-support calls taken by any individual office are transparently routed to the centralized help desk at corporate headquarters over the IP network—at no cost.

Another company was actually installing such gateways at customers’ offices so they could call the company for free over the Internet. This tactic was apparently worthwhile for big accounts, since the cost of calls in both directions typically runs into the low thousands over the course of a year or two.

WHAT'S THE SAME?

At first glance, none of these implementations have anything in common. One involved replacement of conventional equipment. The others didn’t. One created connections to other companies. The others didn’t. One leveraged a private WAN. The others didn’t.

The three companies moved ahead with VoIP for different reasons, and they’re using different architectures to do so. They’re all early VoIP adopters, but other than that, they have almost nothing in common.

Well, almost nothing. There is one aspect all three organizations—as well as most of the others I investigated—have in common: a single person is in charge of both the data and the voice network. This single manager may have had to do some work to convince upper management to go with VoIP, but he or she didn’t have to battle a peer or another internal department.

That’s very instructive. Eventually, the main obstacles that VoIP has to overcome may not be technical or economic at all. They’re probably political. Communications managers with full authority over all things networkable may choose to go whole-hog with VoIP or may find niche VoIP applications. However, chances are they will find VoIP to be applicable somehow to their organizations’ business needs in one way or the other. On the other hand, those who don’t have ownership over both the voice and data side probably will not want to hassle with the internal politics that VoIP engenders.

The bigger lesson is this: Those who view this industry in terms of technology and/or economics alone will miss the big picture. People make decisions based on many factors. And for VoIP, corporate politics is still one of those factors.

To comment on this month’s column, contact me at ll@exit109.com.