Joseph Phillips, PMP: Estimating Time and Project Resources

Once the WBS has been created the project manager and the project team can begin to create an estimate of how long the project work will actually take and what resources are required to complete the project work. Resources are generally thought of as people and the roles they play within a project (think network engineer, mechanical engineer, and architect). Resources, however, are also things like materials, equipment, and facilities that the project needs in order to be successful.

When the project manager first begins to estimate the duration of the project she should always look for historical information. If a project of this type has been completed before historical information will be the best input to the project duration estimating process. This approach is also known as analogous estimating – where the project manager is creating an analogy between the two projects. When using an analogous estimate the projects don’t, and often won’t, be identical. It’s an estimate based on projects that have completed in the past.

Real World Note: An analogous estimate is also known as top-down estimating. It’s not a very reliable estimate, but it’s a quick way to estimate the project duration.

Another project duration estimating approach is the parametric estimate. This estimating technique uses a parameter to determine how long the project work should take to complete. For example, a project to install 80 printing presses may estimate that one printing press will take 35 hours to install, so the parametric estimate would be 80 times 35 hours for 2,800 of labor.

This approach is somewhat reliable, assuming that the initial parameter used is reliable. A parametric estimate does not take into account the learning curve. Basically, when it comes to estimating time using a parametric estimate, the more times a person does an activity their efficiency at repeating the activity will increase, so the duration to complete future activities will diminish.

Consider a project to paint 4,000 hotel rooms. Rooms 1-10 may be somewhat slow but after a project team paints room 40, or 50, or even 100 their efficiency at doing that type of work will increase, so the parametric estimate would be flawed. As you can guess, the parametric estimate is a quick but somewhat unreliable approach to project time estimating.

Some project managers use what’s called a three-point estimate to predict how long the project duration will take. This estimate type uses three points for each work package: the optimistic estimated >time, the most likely estimated time, and the pessimistic estimated time to predict how long each work package will take to complete. The project manager then finds and uses the average of the three points to estimate how long the project work will take to complete. Figure 4 is an example of the three-point estimate in action.

Figure 4: A three-point estimate finds the average time for each activity duration .

Real World Note: The Program Evaluation and Review Technique (PERT) is similar to the three-point estimate. PERT uses a weighted average while the three-point estimate does not. Here’s the formula for PERT: (Optimistic + (4 x Most Likely ) + Pessimistic)/6.

Guess what? It’ll magically take 40 hours to complete the work. Suzie will either complete the work in 32 hours and not tell you she’s done because she reported it’ll take 40 hours or she’ll ease through the work and use all 40 hours to complete the 32-hour assignment. Another possibility is that Suzie won’t start on the actual work until hour eight and she’ll pray and hope nothing goes awry in the assignment. The worst possible scenario is that she’ll wait until hour eight to begin, things do go awry, and she’ll require more than the allotted 40 hours for the work.

There are multiple strategies to combat Parkinson’s Law, one of which is the three-point estimate. Another approach is to use a “management reserve” which is a pool of time that all activities may borrow from should their activities go over the allotted completion time. In other words, if a project is estimated to take ten months to complete the project manager will allot approximately 15% additional time for management reserve. The project will, in this instance, take eleven and a half months to complete. As activities are late the time is subtracted from the management reserve. To use management reserve, however, the project manager and the project team must agree to be accurate and as honest as possible when estimating how long activities may take to complete.

Sometimes it’s darn near impossible to use management reserve – or to provide an accurate time estimate. This is true if a project of this nature has never been attempted before. When a project has never been attempted before it’s called “The First-Time First-Use Penalty.” If you’ve never tried to use a particular type of material, create a particular type of software, or install a new appliance you likely won’t know how long it’ll take to complete the work until you actually get in and complete the work. It’s kind of like the first person that decided to eat a snail – he didn’t know they were tasty until he ate one.

Real World Note : The First-Time First-Use Penalty is one of the reasons why it’s critical to future project success to keep a lessons learned document and accurate project records of actual time and costs to complete the project work. You’re creating future historical information so other projects can glean from your experience.