We've already looked at the fundamentals of IP LAN Telephony. But the days of choosing a technology for its own sake are gone now—there must be a good business reason for any technology deployment. Which brings us to why you would—or wouldn’t—want to give IP telephony a try. In future articles we’ll look at some specifics of network design if you do decide it’s for you, but let’s pick off the major issues.

The prime (sometimes only) reason to deploy anything is to save money, whether capital or ongoing opex costs. The main ones you’ll need to consider as applicable here are:

Switch purchase price—sub-100 users, a small PBX is probably cheaper (though manufacturers are addressing that market space now with cut-down softswitches specifically aimed at smaller organisations). For larger companies, the softswitch will tend to be significantly cheaper.
Switch maintenance costs—typically based on a percentage of the capital outlay.
Handset costs—while IP telephony handsets are priced comparably with digital feature sets for PBXs, most people don’t have these. There’s no question that adding up the cost of lots of IP telephony handsets will come to a very big number. On the plus side, this may be offset by the cost of the switch, and at least you won’t be counting them towards a maintenance cost.
LAN upgrade costs—this could be the deciding factor. If you don’t have a modern, switched, QoS-enabled, Power-Over-Ethernet-ready LAN, then you will probably have to install one. If you’re due for an upgrade anyway, then that may not be an issue. This is what makes IP Telephony installations more prevalent in greenfield sites, where the cost of the infrastructure needed can be lost amongst the building costs that you’ve got to do anyway.
Ongoing management—Moves, Adds and Changes. Because the phones have intelligence in them, they can dynamically tell the softswitch where on the network they are. This more or less removes the need for manual Moves, Adds and Changes (MACs). Some companies have found that the savings they’ve made in not paying out-of-hours rates for maintainers to carry out these pretty mundane patching and reconfiguration tasks have paid for the new system alone
Training—don’t ignore this one, because you may find that not just your IT staff, but also the switchboard operators and even users will need trained up on a new system. This may apply for any new telephone system, or may just apply if you go for an IP-based system with new user interfaces and features.

There will be others applicable particularly to your organisation. There have been companies where the cost of extending the comms room to get another rack of PBX line cards in is greater than deploying a softswitch system for the extra users

Remote site support
If you have multiple remote sites, each with its own PBX, then the idea is to replace them all with one central softswitch. This is more feasible than with a traditional PBX implementation, since only the voice signalling, not every call, needs to traverse the WAN back to the main site. It does potentially offer large cost savings, and much easier management. But you do need a WAN that can support this traffic, you’ll probably still need local PSTN breakout, and you need to provide resilience for if the WAN fails—although that’s not difficult, it will add to the cost. Then again, saving support and maintenance costs on a few dozen small PBXs and key systems is something to be taken very seriously. And if some of your remote sites are in particularly inhospitable places, it may be an awful lot easier getting a circuit of some kind out there and just hanging half a dozen handsets off a LAN switch, than trying to get a PBX set up where you can’t even get the local telco to give you a PSTN line.

We’ve already mentioned reduced Moves, Adds and Changes effort. Since these handsets can be pretty smart, one benefit you can get is if your company is a strong proponent of hot-desking or has lots of staff who move from one office to another. Allow users to log into their phones like they do their PCs, and you no longer have to remember to call forward your calls. Your phone appearance—complete with your dedicated DN, personal speed dials and access rights—will follow you wherever you are.
You can extend this so that, for example, if you have phones that haven’t been logged into by a member of staff, they can be used to make internal calls, say, or local calls or calls just to the operator, so that you control calls made by visitors, pupils or maintenance personnel.
Your phone is now an IP device, so to an extent it can do some of the things your PC can. You’ll hear a lot about web browsers on phones and XML applications to increase productivity. Logging into your Outlook calendar from your phone, so you can arrange meetings without needing your PC, that sort of thing. There are undoubtedly a plethora of applications that might be useful—it’s unlikely there’s anything that in itself is going to force the move to IP handsets though.

Availability and reliability
Your traditional phone system is always there and you’re constantly having outages on your data network, so how can you expect to rely on a LAN-based telephone system? If that’s you, you’re not alone. This is a serious consideration—can you expect your LAN to perform to meet the needs of voice services, in terms of uptime, speed, and latency. Not only do you need an infrastructure that provides resilience and reliability, but also offers the QoS functionality to make sure the voice quality is consistently acceptable to your users.
It is possible to design and build a network that will enable you to provide an IP telephony service that exceeds the performance of your traditional PBX. However, depending on your starting point, that might not come cheap. As will all LAN/WAN designs, you let the levels of redundancy and functionality that you pay for. Do not ignore the data network design when you’re looking at voice network options, as it is entirely possible that getting your network up to spec will cost you more than the voice hardware and software.

If you’re moving into a new building, and it all needs kitted out from scratch, then IP telephony is particularly attractive. If your PBX is on its last legs and needs replaced, IP telephony may be a cost-effective answer compared to a fork-lift upgrade from the PBX vendor. If you can see a productivity saving that can be made by using IP handsets to allow better communication for your staff, then it’s worth a look. One thing is for sure. IP telephony has a major foothold now, as can be seen by the fact that nearly all traditional PBX manufacturers also offer an IP telephony product too. As convergence becomes more and more the norm, it is looking increasingly likely that it’s not so much if you will ever give it a try, just when.