Project Costs Tell the VoIP Story
July 9, 2007 Dan Burger
Technology is the sizzle, but it’s the bottom-line savings that puts the meat on the table. For voice over IP (VoIP), the starting point for asking questions comes from one angle: How can it save money? Budgets are tighter than a rusted lug nut on a ’51 Mercury. Projects are prioritized by how much money they can either make or save. And generally it’s easier to pinpoint how money will be saved rather than how it will be made.
VoIP will almost assuredly save money when compared to the PBX phone systems that are standard equipment at almost any organization you’ve ever thought about. In many instances, the difference can be startling. During a conversation last week with Terry Boulais, the director of business development at Key Information Systems, Boulais passed along a comparison based on the real world VoIP implementation at an unnamed U.S. bank with 12 branches and 500 employees.
In this example, Boulais points out that the bank is projected to save, over the course of five years, $654,000 in long distance tolls, conference call charges, and maintenance fees. In addition, there’s a $487,000 savings related to charges for PBX hardware and software that no longer exist. Not exactly a nickel and dime waste of time. I will mention, however, that the source of this information is a VoIP sales department eager to make to a strong impression. As such, it is the starting point for a conversation about the costs involved in VoIP projects. Costs are only part of the story, and only part of the story can be determined by what’s included in the costs. IBM likes what the System i brings to this discussion because it is highly reliable, highly scalable for growth, and offers logical partitioning for setting up backup and recovery options.
If you haven’t already read the VoIP articles published in the recent issues of The Four Hundred, please see VoIP and the Search for Single Points of Failure and Adoption of VoIP Tied to Relief from Phone Expenses as background to this piece, which is going to go further into cost scenarios and project planning considerations.
Dan Burger: What can we take away from the real world example of the bank saving over a million dollars during a five-year span? Is it that simple in black and white terms?
Terry Boulais: Because this is a bank, the move to VoIP is easier because banks are advanced in terms of their network infrastructure. They also have an emphasis on security, which could translate into encryption on voice calls if it was determined to be necessary. But the main thing is they have the pieces in place that enable voice over IP to be implemented without spending on infrastructure.
According to the figures (which were supplied by VoIP software provider 3Com), it shows that after 24 months the bank will have gained a return on investment based solely on moving its basic phone system to voice over IP. The formula takes into account reduced telephony costs plus productivity improvements.
The biggest savings relates to conferencing calls due to the long-distance charges. With voice over IP, everyone is on the same LAN (local area network), you can use conference calling without incurring extra costs. At the end of the month there is no check to write at the end of the month for a separate phone network because that is all built into the existing LAN with voice over IP.
The PBX box is gone. It’s that huge piece of hardware, like a giant switch, that all the phone wires feed into. There is a monthly service charge for that piece of hardware.
These are the reasons why somebody takes a look at VoIP out of the gate. But there are other benefits that come into play are realized when the system is tied in with other applications. And that is the real reason to go in this direction, but most people look at it to save pure dollars and that includes productivity benefits.
DB: What are your opinions on whether a company would buy VoIP and run it in his IT department or contracting it out as a service and maintenance deal?
TB: I think every contract we’ve done has the maintenance done on the software by 3Com. That’s part of the contract with 3Com. Smaller companies–maybe 30 to 50 people–may want to outsource the entire project–to a managed service group–because they don’t have the infrastructure or the hardware. It’s a convenience thing for them with a monthly fee. Most of the larger companies that we’ve been dealing with find it more cost efficient to handle it internally.
DB: Moving the voice communications piece from PBX to VoIP is one thing, but if you want to add data capability to the project so that telephone staff have, for instance, ERP interfaces that were not available to them before, what does that add to the price of a project?
TB: If the network structure is in place, then it is simply adding the software to interface the ERP and the phone system. So when the person calls in and the system recognizes the phone number it grabs the appropriate database and pulls it up on the screen. The interfaces could be a few or they could be many, and that has an impact on the total cost of the project. Depending on the complexity of the application, there could be as little as one month of implementation costs involved or as much as six months to get these interfaces right.
If it’s a home-grown application, it will take longer than one of the top five ERP application vendors. There’s a kit for the top five that allows it to be easily tapped into.
A customer can get the voice over IP, the call center, and other features up and running rapidly because they are software modules. When the request involves dipping into databases and pull info up on the screen, and the applications are home grown, it’s going to involve determination of file names and file types, sometimes in home grown apps it requires going to multiple files to get all the necessary info. The more files, the longer it takes.
DB: When you add features how much impact does that have to the price?
TB: They are all flat fee items that are sold as off-the-shelf modules to the 3Com software. Things like integrated fax and auto call routing. The costs are based on the number of users. It’s actually sold in blocks such as up to 100 users or up to 1,000 users.
DB: What are the intangibles in planning a PBX to VoIP system that can cause cost overruns in the final project?
TB: Starting out with an incomplete picture of the goals and objectives seems basic, but people will start out with the idea they only want basic phone service and later decide that they want call routing or integrated fax. It’s a desire to have additional features.
Cost overruns can occur when infrastructure does not allow the system to work properly. That’s why an assessment at the beginning is necessary. We do an infrastructure analysis. We wouldn’t start a project though if the infrastructure was improper. Adding video conferencing capability later can add infrastructure costs because of bandwidth issues. Doing integrated phone onto Blackberry mobile devices requires interfaces.
What really affects the overall project costs are writing the interfaces, so by not factoring that into the beginning of a project will cause cost overruns. It could require months and months of labor.
DB: As we talked about in the articles that came before this one, a company needs to be thinking in terms of fault tolerance, high availability, and disaster recovery, too. If the main box running voice over IP goes out, the party is over. The cost saving figures that compare PBX and VoIP do not include a redundant system do they?
TB: That’s correct. These figures are based on how much money a company will save switching to voice over IP solution, not a disaster recovery solution. It is not how much will be saved if a fault tolerant system is part of the deal. Adding high availability, for instance, would be huge if left unaccounted for in the early planning stages of voice over IP. As we talked about in the earlier articles, it is important to think about this outside the scope of just pure phone.
A fault tolerance solution will likely be reflected in additional infrastructure and hardware costs. If it’s a disaster recovery solution, there will be extra LAN costs to take the network to another location and add hardware costs. Some can keep it on the same network if there multiple locations already exist as in the bank with 12 locations example we used as an example. If there’s an iSeries at that location, then it could be a matter of carving off a partition to use as the secondary box if one of the boxes at another location goes down due to a disaster, a power outage, or whatever. It’s one of the reasons voice over IP looks good on iSeries.