Service provider special: How to upgrade to VoIP

Resellers helping their customers upgrade to VoIP will do well to spread telephony migration problems to reduce risk and increase opportunity.

It's always smart to think long-term and have a strategy that guides not only your current initiative with a customer, but the next one as well. That is, after you complete the project at hand, where will that leave you? Do choices you make in this project eliminate future choices? Advantages of foresight and long-term planning include consistency, more efficient utilisation of your resources, and project budgets that help instead of hurt you.

But for VARs doing telephony migrations to VoIP, there are two additional important reasons to take a long-term, strategic view. First, spreading the work over multiple projects reduces the risk of any given project. Second, it's more efficient and thus profitable for you to use a few resources over a long period of time as opposed to a large number of resources in a very short period of time to accomplish the same scope. For instance, keeping a couple guys on a project all year instead of using 20 guys for a month. The combination of these also mean that it's easier to sell a small first-step project than a high-dollar comprehensive project, and success in the small, lower-risk project will usually result in being awarded subsequent projects with well-defined objectives.

In light of this, let's look at two long-term strategies for migrating a customer's telephony set up from a legacy PBX to a VoIP system.

Option A

Many legacy vendors like Avaya, Nortel and Siemens offer IP modules that attach to their older PBX that provide signalling over IP to the phones and gateway services between IP and the PSTN. So one option is to leverage the existing PBX by upgrading it with IP functionality to provide immediate VoIP benefits while planning to phase out legacy systems and be 100% IP at a later time. Some of the potential advantages of this approach are:

  • Financial, depending on depreciation schedules, etc.
  • Preserves some look and feel, so less retraining is required and there are fewer helpdesk calls
  • Leverages old cable plant (typically CAT3), which is often available in more places than the Ethernet LAN cable plant (typically CAT5)
  • Avoids changes to PSTN circuits
  • Higher comfort-level with existing voice technicians in legacy product means less training and better architectural decisions
  • Avoid disruption to call centres or other special telephony requirements like faxing

An additional advantage of this approach is that if you have some users on legacy phones and other users on IP phones for some period of time, it's OK. It won't be perceived as a stalled or failed project. Because your customer isn't paying for complete legacy and IP infrastructures simultaneously, they can migrate users as they can afford new phones instead of being under tremendous financial pressure to complete the project to avoid infrastructure costs.

Option B

The other option is to deploy IPT softswitches immediately and in parallel with legacy PBX. Then migrate users individually from one system to the other, decommissioning the legacy infrastructure once the last user is migrated. The advantages of this approach are:

  • You're not locked into the current vendor and can shop around for more features or cheaper services
  • More flexible design options without concern about backward compatibility
  • Can often avoid expensive PBX software upgrades or maintenance contracts
  • Can often take advantage of competitive pricing when changing vendors
  • Faster path to an IP-only telephony system

Understanding why your customer wants to migrate to VoIP is the key to navigating these two options. There are many possible drivers, and being able to identify the important ones will often make the choice of options fairly obvious. For instance, if the key driver is that the current environment is so old, that it's experiencing a lot of failures, that clearly favors option B. If the key driver is toll-bypass, then option A is probably a lot faster and less expensive. In any case, look for the fastest path to satisfy the driver while still keeping a long term view of the overall strategy, keeping in mind that the various activities described above are really more about strategies than projects, as they can be sorted into one or many projects, depending on the sise and complexity of the environment.

About the author: Tom Lancaster, CCIE# 8829 CNX# 1105, is a consultant with 15 years of experience in the networking industry. He is co-author of several books on networking, most recently,CCSP: Secure PIX and Secure VPN Study Guide, published by Sybex.

Read more on Voice networking and VoIP