By: Stephen Callahan, PMP

Do you often feel like you are chasing your tail when asked for project estimates? Is your usual estimating technique pulling a number out of the air? Do you find yourself working evenings and weekends to meet unrealistic due dates? If you answered yes to any of these questions read on to learn several project estimating techniques and enhance your ability to produce realistic and achievable estimates. In this article you will learn a particularly valuable technique (i.e., Decomposition / Detail Definition) that produces the most accurate estimates AND is defensible if challenged. People using this technique often discover they need two to three times more time to complete an activity than their intuitive/rough estimate. The only way you can get a better estimate than using this technique is to wait until you have finished and use your actuals!

There are several methods for developing duration estimates:

1. Analogous Estimating (Prior Experience) – An estimating technique that uses the values of parameters, such as scope, cost, budget, and duration or measures of scale such as size, weight, and complexity from a previous, similar activity as the basis for estimating the same parameter or measure for a future activity.

• Uses past similar projects as a base

• Used when there is limited details about the project (ex., early phases)

• A form of expert judgment

• Team members providing analogous estimates must have relevant experience and expertise

• Must factor in complexity, similarity, new technology, etc.

2. Parametric Estimating – An estimating technique that uses a statistical relationship between historical data and other variables (ex., square footage in construction, lines of code in software development) to calculate an estimate for activity parameters, such as scope, cost, budget, and duration.

• Relies on documented estimating relationships

• Produces more accurate estimates (assuming estimates based on sufficient and accurate historical data)

3. Expert Judgment – is judgment provided by a subject matter expert in a specific application area, knowledge area, discipline, industry, etc. as appropriate for the activity being performed.

• People who are expert in this type of work effort are consulted

• Greater accuracy as more “experts” are consulted (Wideband Delphi method – Ask several experts and take the average of their estimates)

4. Decomposition / Detail Definition (Bottoms Up) – Activity owners further decompose their assigned activities into steps. See the example below.

• Subdivide into detailed steps that can be (more) accurately estimated

• Largest investment in time to develop

• Most accurate estimates

All the above methods should be used to some extent in developing activity duration estimates. Since all projects by definition are unique ventures, decomposing activities into steps will provide the best estimate for the current project, as well as evidence for the basis of the estimate.

The activity owner, along with any others who are familiar with the activity, is responsible for developing the estimate.

PERT (Program Evaluation and Review Technique) method – A technique developed by the Navy is called the PERT technique. It is a 3-point weighted average estimate and is used to overcome a natural bias toward optimistic estimates. It uses the following three estimates:

• Optimistic (O)

• Most Likely (M)

• Pessimistic (P)

The formula used in the PERT method is:

(O + 4M + P) / 6

The resulting estimate adds a little more time to the most likely estimate.

Three-Point Estimate – is a new technique similar to the PERT that improves the accuracy of activity duration estimates by considering the amount of risk in the original estimate. It uses the same three estimates as the PERT technique but does not weight the “most likely” estimate. After determining the “optimistic,” “pessimistic,” and “most likely” estimates one simply takes the average of the three:

(O + M + P) / 3

Detail Definition Method – The activity duration estimating process requires:

• That the amount of work effort to complete the schedule activity is estimated
  

• The assumed amount of resources to be applied to complete the schedule activity is estimated, and
  

• The number of work periods needed to complete the schedule activity is determined.
  

• All data and assumptions that support duration estimating are documented for each activity duration estimate.

1. Break the activity down into steps.
2. Assign effort estimates to each step.
3. Total up the step effort estimates to determine the activity effort.

4. Remember to record any assumptions relative to the activity that influence the effort and/or duration estimates, for example:

• Number of people who will work on the activity

• Skill level of the doer(s)

• Availability of the doers to work on the activity (i.e., percent availability)

• Productive project hours in a typical day (vs. non-project work)

• Availability of necessary resources.

5. To determine the duration estimate:
a. Determine the number of productive project hours in a typical day (Note: consult with management to identify their expectation for productive project hours/day vs. other non-project business activity)
b. Determine how many of those hours are available for working on the new activity (taking into consideration other project work the resources are already committed to.)

6. Divide the activity effort by the number of hours available to work on it to get a preliminary duration estimate.

7. As a last step determine if there are any outside influences for the steps (ex., legal review, management review, etc.) and add the delay time to the duration estimate.

8. Determine if the activity has a variable (Effort-Driven) or a fixed (Duration Driven) duration:

• Variable duration: adding more resources will reduce the duration. The assumption with a variable duration is that with additional resources some steps can be done in parallel thereby finishing the activity faster. The activity owner would create a network diagram for the activity steps.

• Fixed duration: adding more resources will not reduce the duration.

Happy estimating and enjoy your improved work/life balance!

Matthew Sheaff currently serves as Director for the Center of Project Management at The Performance Institute. For questions and or comments he can be reached directly by emailing Sheaff@PerformanceWeb.org.