Menu
Calculating the cost of communicating

Calculating the cost of communicating

At a recent trade show, I happened to mention that some of my more forward-looking clients have created "communications calculators" that enable them to predict -- with a high degree of accuracy -- what the communications costs will be for moving or adding employees. Let's say a business unit owner wants to shift 5,000 employees from site A to site B -- how will that affect communications costs?

Unsurprisingly, the first question I got was, "How do I build one of those?" Here's what I suggest:

Start by investing in products that enable you to accurately identify application flows through the network -- by type, site and user. Track this over time, to determine long-term trends, so you're aware of what percentage of the WAN is being consumed by which applications.

While you're gathering that data, make sure you're investing in asset management technology. Make sure you're aware of which hardware and software assets you've got, what's being used and what's idle, and by whom it's being used. Count everything: licensing fees for unified communications applications, hard and soft IP and TDM phones, and of course, any and all data networking gear.

Also audit your WAN service contracts. Try to arrive at a consistent dollars-per-megabit-per-second metric for different traffic and service classes. In other words, if you're still using POTS, convert your voice traffic from cents per minute to dollars per megabit. Rank your MPLS, private-line, and legacy frame/ATM data traffic the same way (you may need multiple ratings if you're paying for differential class of service). And don't forget wireless -- it's an increasingly important component of your overall communications spend.

Figure out how many people you have supporting your users, and what they're doing. Separate out the folks who are doing architecture and planning from direct support and break-fix -- the former don't increase based on the number of users, but the latter do. A good ratio to have is users-per-support staffer

Take this information -- all of the above -- and start crafting "user profiles." You shouldn't need more than a handful, but you should be able to categorize users fairly simply based on the following: application portfolio, hardware configuration, LAN connectivity, WAN connectivity, support requirements, geographic location, mobility requirements, telecommuting requirements, and backup and recovery needs.

The user profile is essentially a map between users' job functions and their technical requirements. The idea here is to be able to fairly straightforwardly build out definitions like, "A back-office administrative worker has the following requirements. . . . and . . . A developer/engineer has the following requirements.

Next, build out your model. Watch for step-function increases -- places where adding one more user necessitates upgrading a pipe from T-1 to fractional T-3, or adding a new server or staffer. At the end of this exercise, you should be able to estimate the communications cost of each employee quite well.

Finally, loop in accounting. To perfect your model, you need real data -- so compare your estimated and actual costs, and investigate every discrepancy. Just a few iterations will clarify where your model needs beefing up.


Follow Us

Join the newsletter!

Error: Please check your email address.

Featured

Slideshows

Reseller News ICT Industry Awards 2017 - Meet the winners...

Reseller News ICT Industry Awards 2017 - Meet the winners...

Reseller News honoured the industry’s finest on a standout evening for the New Zealand channel, recognising the achievements of established and emerging partners on a memorable night in Auckland.

Reseller News ICT Industry Awards 2017 - Meet the winners...
Show Comments