Each week, project management veteran Tom Mochal provides valuable advice about how to plan and manage projects. Tom first describes a common-problem scenario, based on real-life situations. He then describes a solution, using practical project management practices and techniques.

Estimating the cost of an IT project can be a hit-or-miss proposition, especially if the task involves unfamiliar technology. Here is how one project manager used two methods to make sure his estimate was on target.

The dilemma
Sean works in the IT Research and Development (R&D) Department at Blue Sky Manufacturing. He was asked to create a proof-of-concept for sending information from the company intranet to new handheld devices. His boss had returned his first estimate for the project.

“I gave my boss a high-level estimate,” Sean said. “But he said there were too many guesses in it. He asked me to get more facts and estimate the work to a greater degree of certainty. I agreed that my first attempt was a guess, but how can it be more reliable than that—we have never done a wireless project before.”

“There are usually ways to estimate work that are better than guessing,” I said. “Let’s start at the beginning. Do you have a project definition?”

I was pleasantly surprised to learn that Sean did have an abbreviated project definition and a template that described what needed to be done on a proof-of-concept project.

“We have much of the information we need,” Sean said. “But now what should I do? This technology is completely new to us.”

Mentor advice
“First, the fact that you have a template should help us put together an initial estimate,” I said. “Second, although your point about the technology being new to us is valid, handheld devices are not brand new. We should be able to validate our first estimate by talking to some experts who have experience in this area.”

Many people struggle with estimating because they are not really clear about what they are trying to do. This may be especially true in R&D work. However, good estimates are the result of good techniques.

In Sean’s case, he had the advantage of a project definition and a template to guide him through the proof-of-concept project process. From there, he should be able to estimate the work separately using two techniques:

  • A simple work breakdown estimate
  • Expert opinion

Break down the work
For the work breakdown technique, Sean needs to start with his work-plan template and estimate each activity. Activities that have been done before can be estimated reasonably well. New activities are more difficult to estimate, but the work breakdown structure, which defines activities at a more granular level, should help produce estimates that are reasonably close.

The lower-level estimates are added up to a summary number. Project management time is estimated at 15 percent of the summary total. Sean should also add a 20 percent contingency to cover the remaining estimating uncertainty and risk.

Consult the experts
The second technique is expert opinion. Even though Sean has not done this type of work, he can talk to a research analyst or an industry expert to gain an understanding of what is involved. These experts will know what other companies are doing and can validate or challenge the previous estimate.

If the two techniques do not yield similar results, Sean can go through the work breakdown estimate with the expert and determine which one might need revising. When Sean is comfortable that the two techniques are yielding similar estimates, he can document the estimate package and send it on to his boss—this time with substantial supporting detail.

Project management veteran Tom Mochal is director of internal development at a software company in Atlanta. Most recently, he worked for the Coca-Cola Company, where he was responsible for deploying, training, and coaching project management and life-cycle skills to the IS division. He’s also worked for Eastman Kodak and Cap Gemini America and has developed a project management methodology called TenStep.

Tell us about your project management problems. Tom Mochal will answer those that affect the widest number of readers. Post a comment below or send us a note.