Next Step in the Schedule Development Process

Before we get into the details of estimating, let’s make sure we are clear on where estimating falls in the schedule development and planning process. If someone stopped you on the street and asked you for an estimate, what is the minimum information that you would need? You would need to know what the estimate is for—what work is to be done. And you would need to know who is going to do it—what type of resources will be involved in performing the work. This flow is shown in Figure 7.1. Estimating the work should occur after you have identified the work and after you have thought about what resources are needed for the project.

It sounds so simple, doesn’t it? Then why is this so tough? Well, we cover this in more detail later in this chapter, but these two basic prerequisites are where most estimating woes originate. There is often not a clear or complete understanding of the work to be performed by the person doing the estimate, and the relationship between the work estimate and the resource doing the work is not defined or communicated. In addition, there is the challenge of estimating work that has not been done before in exactly these conditions.

Yet, estimating the work effort is a cornerstone activity for planning the project. From these work estimates, we determine the project costs (see Chapter 9, “Determining the Project Budget”), develop the project schedule (Chapter 8, “Developing the Project Schedule”), and identify key project risks. This relationship is illustrated in Figure 7.2.

Managing the Risk, Managing the Estimates

There lies the key challenge. How do you manage the uncertainty that is naturally involved with the estimating process? Because these estimates form the foundation for the project schedule and the project budget, you must implement techniques and approaches that enable you to properly manage this risk and the expectations of your stakeholders.

Although this subject of estimating and risk could easily slip into a review of statistics, probability, standard deviations, skewed distributions, and Monte Carlo analysis, we will not go there. In many real-world environments, these advanced concepts and techniques are not utilized to estimate work and to manage the associated risk, and these topics would be outside the scope of this book. Our focus is understanding the impact that estimating work has on our overall risk management approach and what we can do to minimize those risks.

Estimating the work is a fundamental risk analysis step. Not only do you estimate work efforts, but you also identify the assumptions that support the estimate and the key risk factors that might affect the accuracy of those estimates. These key outputs are depicted in Figure 7.3.

Reasons for Estimating Woes

Before we review the key estimating techniques and methods that we need to know to best plan our projects and manage our risk, let’s first take a deeper look at the common reasons for estimating woes on many troubled projects:

• Improper work definition—The number one reason for inaccurate work estimates is inadequate definition of the work to be performed. This includes the following:

  • Estimates based on high-level or incomplete requirements.

  • Estimates based on incomplete work—work elements (packages) not accounted for in the WBS.

  • Estimates based on lack of detailed work breakdown.

  • Estimates made without understanding the standards, quality levels, and completion criteria for the work package.

• Wrong people estimating—Another key reason for inaccurate work estimates is that the wrong people make the estimates. Although it might be appropriate for management to make ballpark estimates during the early defining and planning stages, when firm commitments must be made, it is best to have the people who have experience doing the work make the estimates (or at least review and approve any proposed estimate made by someone else).

• Poor communications—This reason hits on the process of facilitating estimate development. This category includes events such as

  • Not sharing all necessary information with the estimator.

  • Not verifying with the estimator what resource assumptions and other factors the estimates are based on.

  • Not capturing and communicating the estimate assumptions to all stakeholders.

• Wrong technique used—We cover this in greater detail in the next section, but this category includes events such as

  • Making firm budget commitments based on top-down or ballpark estimates rather than bottom-up estimates.

  • Not asking for an estimate range or multiple estimates.

  • Not leveraging the project team.

  • Not basing estimates on similar experiences.

• Resource issues—This is a specific case where it’s not really an estimate issue but an issue with the assumptions supporting the estimate. This is when the person assigned to do the work is not producing at the targeted level or when there are performance quality issues with any of the materials, facilities, or tools. Without documented assumptions, these issues can appear as inaccurate estimates to stakeholders.

• Lack of contingency—In many cases, especially on projects involving new technologies and new processes, the identified risk factors are not properly accounted for in the work estimates. The uncertainty level in specific work estimates needs to be identified and carried forward into the project schedule and budget as part of the contingency buffer or management reserve.

• Management decisions—In many situations, senior management influence and decisions impact the estimating accuracy level. This category includes events such as

  • Senior management making firm budget commitments based on initial, high-level estimates and not accounting for accuracy ranges.

  • Senior management not willing to invest time or resources to get detailed, bottom-up estimates.

  • Estimators factoring their estimates for senior management expectations rather than the actual work effort.

  • Management requesting that estimates be reduced to make the work meet the budget or schedule goals.

  • Management decisions to bid or accept work for less than estimated cost.

  • No use of management reserve to account for risk or uncertainty.

Powerful Estimating Techniques and Methods

There are several key estimating techniques you should know about. Table 7.1 lists these techniques and summarizes the key characteristics of each.

For each estimating technique (approach), there are one or more methods that can be leveraged. Table 7.2 lists these methods and summarizes the key characteristics of each.

As with all other planning activities, work estimates are refined and improved as more is learned about the project. At a minimum, each project (or project phase) should be estimated three times. Each estimate provides a greater degree of accuracy. To better understand this concept and to better educate others in your organization, see the three levels of estimate accuracy recognized by PMI in Table 7.3.

Best Practices

Now that we have an overview of the estimating techniques and methods that are available to us, and we have a feel for the estimating mistakes that are commonly made, let’s review the estimating best practices of successful organizations and projects:

• Should be based on the work breakdown detailed in the WBS.

• Estimating should be performed (or approved) by the person doing the work.

• The work estimates for lower-level WBS items should be less than the standard reporting period for the project (typically one or two weeks).

• As discussed in Chapter 6, “Developing the Work Breakdown Structure,” if the work estimate is not less than this, it is a good sign the task needs further decomposition.

• Estimating should be based on historical information and expert judgment.

• Estimates are influenced by the capabilities of the resources (human and materials) allocated to the activity.

• Estimates are influenced by the known project risks and should be adjusted accordingly to account for those risks.

• When estimating effort for something unprecedented or high risk, where the confidence level in any estimate is low, leverage timeboxing and proof-of-concept models to improve knowledge quickly, so you can derive better estimates. This approach is a key aspect of agile projects.

• All bases and assumptions used in estimating should be documented in the project plan.

• When asking an SME for an activity estimate, make sure to provide the following whenever possible:

  • Project definition document (context, approach, assumptions, and constraints)

  • WBS

  • Applicable standards, quality levels, and completion criteria for the work package

• When asking an SME for an activity estimate, make sure to ask for the following at a minimum:

  • An estimate range (not just a single value)

  • Factors driving that range

  • Assumed resource level, skills, and productivity

  • Assumed quality level and acceptable completion criteria

• Estimates should be given in specific time ranges.

• To manage high-risk projects, the following estimating techniques are recommended:

  • Use of phased and bottom-up estimating techniques

  • Use of the average weight and team consensus estimating methods

• For high-risk projects where the organization lacks significant previous experience or process knowledge, consider outsourcing the planning phase to an outside firm as an assessment engagement.

• A project’s time and cost estimates should be based on project needs and not dictated by senior management. The project manager should work with senior management to reconcile any differences.

• Reserve time (contingency and buffer) should be added to either the project schedule or to individual activity duration estimates to account for the level of risk and any uncertainty that exists.

• Historical information is vital to improving estimates. If you don’t measure actual performance, you will not have the feedback to improve estimating accuracy.

The map in Figure 7.4 summarizes the main points reviewed in this chapter.