Refine estimates at the end of major milestones.
by Neal Whitten, PMP, Contributing Editor
Most of us have been taught that, at the beginning of a project, we make commitments that carry us all the way to the completion of the project. That’s a reasonable directive if the project is up to three months or so in duration. However, projects that are longer should not have their end dates committed. You read correctly. Once again, I am going to go against conventional wisdom, so put your thinking caps on.
Projects should be managed by major milestones, and major milestones should be set one to two months apart. The closest major milestone is firmly committed. This means that the members of a project will do everything reasonable to achieve that committed date.
All remaining major milestones are estimates, including the ultimate major milestone: the delivery date. They are not firm commitments. They are, however, sincere estimates and every reasonable attempt should be made to achieve them. It is imperative to estimate these dates for planning and business purposes.
Just after the identification of each major milestone in a project plan, an activity called “resize project plan” is added. After a major milestone has been reached, the project manager (working with the project members) determines if the next major milestone and subsequent major milestones are still reasonable to achieve. At that point, major milestones can be re-estimated and the relevant portions of the project plan are updated and rebaselined. This approach to planning and estimating follows the rolling wave concept whereby estimates are refined as better data becomes available.
Many types of projects, but especially software projects, get a bum rap for missing schedules. But think about it: Who in their right mind would commit to an end date before knowing what it is that is to be built or how it will be built? Most of us! But this is old-school, archaic thinking.
I often hear executives say that they can see some benefit to this thinking for noncontractual projects but not for contractual projects. They tell me that there is no point in resizing the project plan at the end of major milestones because the deliverable date is fixed in a legal document. Why perform a useless, wishful exercise?
Projects that require resizing are exactly the ones that are defined in contracts. Why? Because of the culture surrounding contractual projects. When asked, “Are you going to complete on time?” a project manager typically answers, “I have to.” The next question is, “I know you have to, but are you going to complete on time?” On contractual projects, no one feels empowered to question or revise dates. Instead, hope and optimism spring eternal.
Hope and optimism are not enough to make a project successful. Good planning and judgment are required. Contractual projects particularly must be considered for resizing at the end of major milestones to avoid sweeping problems under the rug. Projects need a culture where all problems are visible so that they can be solved effectively. Note that if a project is resized and its problems are brought into focus for resolution, that still does not guarantee that the project can or will be delivered on time. However, not confronting the problems is certain to further delay any chance to achieve the committed delivery date. Resizing helps to minimize the collateral damage that can build up on a project.
Commitments should be viewed as sacred ground. However, if a person believes that a commitment cannot ever be revised–and the commitment is in jeopardy–then the tendency is to deny or ignore that there’s trouble and, therefore, avoid seeking the best corrective action. Better to admit a problem, create a fix and potentially be marginally late than to ignore the problem and suffer far greater damage. Commit to that which you reasonably understand. Estimate for that which you have a weak understanding, and resize when new information is available.
Now, go make a difference!