Merging your telecom and data network management? Some things to think about

Today it is estimated that between 10% and 15% of the installed base of PBX systems is now more than 10 years old. In addition, many of those systems need to be either totally replaced or significantly overhauled. For these PBXs, the estimated cost of just adding functionality such as Automatic Call Distribution may be as high as a few thousand dollars per seat. If messaging, Interactive Voice Response and basic Computer Telephony Integration options are also added, the incremental cost per seat may approach the $5,000 range. Ouch!

Perhaps this level of add-on investment may have been justified five years ago. However, many industry observers expect that by mid-2000 the state of PBX product development will have shifted dramatically. At that point, it is expected that many of the traditional PBX vendors will have shifted away from the classic monolithic architectures of the past and instead embrace active interworking with a variety of server-based architectures and systems.

By no means is this a projected event. The fact is that it is already happening. In October of 1998, a significant event in the PBX industry occurred at NetWorld+Interop when two of the three largest PBX vendors - Lucent and Siemens - each announced new LAN-based PBX products at about the same time as Cisco. All three vendors stated that their systems would currently support as many as 100 users, with plans to scale to 1,000 users in 12 to 18 months.

This event is particularly noteworthy because the last major PBX architectural shift - the transition from the analog to the Digital PBX architecture that began in the late 1970s - was a result of the major market leaders at the time (AT&T and Northern Telecom) deciding to endorse -- rather than fight - the concept by introducing their own products.

In light of these industry events, data net managers who are broadening their scope to include telecom management are well advised to question more than ever their vendors' long-term architectural plans for today's current circuit-switch-based PBXs. In addition, users should also question their vendors' plans to deliver and install IP telephony and server-based solutions over the coming 12 to 18 months. There are many (including myself) who believe that a significant implementation spike in campus telephony products may begin to occur in the latter half of next year. If that indeed turns out to be the case, those who intend on reaping the benefits of this early majority phase will begin some serious planning efforts in the latter half of this year.