Estimating on the Fly
Project managers routinely feel a great deal of pressure to start work on a project quickly since many executives think planning and risk management are simply bureaucratic paper shuffling processes with no real-world pay off. There is some truth to that assumption, particularly in bureaucratic organizations where any activity like risk management is an opportunity for more papers, more procedures and more endless meetings. In addition, there is the fantasy that good project managers are good firefighters and so spending time and money on risk management is wasted. When bad risks flare up; we just fight the fires.
Increasing Certainty of the Estimate as We Advance Through Planning
The project manager needs to supply estimating data to decision makers but also needs to qualify it with additional information about how certain the numbers really are. Everyone is unhappy with estimates that are somewhat uncertain. But far better to trigger that unhappiness than to have people making decisions based on data that is nowhere near as reliable as they would like. In the table below, we talk about a sequence of estimates during the lifecycle of a project with increasingly greater accuracy. The following sequence is the best way for a project manager to give decision-makers the data they need without making commitments that are not realistic.
How to Estimate Project Duration |
||
Step # |
Responsibility of |
Actions to Take |
1 |
Project manager |
The foundation for accurate estimates of duration is tasks that are defined not as activities with no endpoint but rather as deliverables with acceptance criteria. |
2 |
Project manager and team members |
The project manager completes the work package, which details the deliverable of the task and any associated documentation, inspections or standards that have to be met. |
3 |
Team member |
Reviews the work package to ensure that it is complete and clear. The team member then adds information about the major risks that could cause the test to take longer and also specifies what they need before they can start work. |
4 |
Project manager and team members |
Meet to review the work package and determine the resource's availability to work on the project (i.e.; half-time, full-time, four hours a week). The project manager secures a commitment from the team member's organizational superior on the availability |
5 |
Project manager and team members |
Based on the work package information, the project manager asks the team member to make three estimates of how many hours of work (not days of duration) the task will take. The first estimate is an optimistic estimate assuming most things go right, the second estimate is a pessimistic estimate that says several things will go wrong. The third estimate is the best guess. The project manager then uses those three estimates to calculate the weighted work estimate for the project. |
6 |
Project manager - during initial meeting to discuss the project and its scope |
Using the estimate from the team member and the predecessor relationships between tasks, the project manager sequences the tasks. Then, using the availability of the team member to work on the project, the project manager calculates the duration the start and finish dates for each task. |
