There's an article on Computerworld that caught my eye several days ago. I've talked to several consultants, resellers and vendors over the past few months who've all alluded to this problem in one way or another

Survey: VoIP planning, installation take more time than expected
Upfront time is double that of an earlier report
News Story by Matt Hamblen


MARCH 17, 2006 (COMPUTERWORLD) - Despite significant benefits, voice-over-IP projects take more than twice the time for planning, installation and training than once expected, according to recent research.

In annual surveys of companies conducting VoIP deployments, the average time a company devoted to upfront planning, installation, troubleshooting and training in 2004 was 52 minutes per user, which increased to 133 minutes in 2005, according to Nemertes Research Inc. in New York.

First, let me once again lay to rest a misconception that's plagued the combined voice and data industries for years. There's an old sales argument that's been used for the past twenty years. It takes on one of two forms - "Pay for the voice network and the data rides for free" or "Pay for the data network and the voice rides free." Speaking as someone who's worked in both sales and engineering for both voice and data services, let me give you a tip - If your sales person, voice or data, ever utters those words, escort them to the door. They don't know what they're saying, they don't understand your needs and they are doing you a disservice.

Those are high level sales-y arguments to the economy of scale of converging networks, but in the context they're often used, they are a fantasy. There's a more accurate quote you're all familiar with - There ain't no such thing as a free lunch.

There are absolutely economies of scale to be gained by converging voice and data. The people who will gain the most are those who are the most prepared.

Too often when exploring a VoIP deployment, the narrow, or simple, approach is taken. The idea that we have a data network that performs really well lures us to believe we can simply add another appliaction (VoIP) and it will work flawlessly. The truth is far more complex. The addition of VoIP to a data network is a significant change. It changes the entire security posture of the network. It changes the data delivery model of some networks. Quality of Service (QoS) to provide the appropriate call quality parameters may be necessary.

When you explore VoIP as a service in your corporate network there are some steps that can't be overlooked. That's the pain point mentioned in the article I referenced. It's too easy to overlook the need for comprehensive planning. Many network managers incorrectly believe their networks can support VoIP without performing a complete readiness assessment test. Too often VoIP services are implemented without considering the impact of security enhancements on call quality.

Those who suffer, and wind up being quoted in articles like this one, are often those who failed to plan ahead. It's absolutely true that adding VoIP is not without costs. It requires a significant labor effort to be successful. The better you plan, the more aware of the labor effort you'll be up front in the project. Set your own expectations reasonably and you'll be successfully pleased with your outcomes. Fail to invest the effort in planning, readiness assessment and preparation, and you will be disappointed in just how much work you create for yourself.

It is always harder tp go back and rework an implementation for success than to do it properly the first time. Protect yourself by understanding what you're doing each step of the way.

Note: I'm posting this in both the community forums and on the blog.

Technorati Tags
VoIP
IP Telephony
Voice Over IP
VoIP plannins
readiness assessment