Project Planning and Estimating for Managers
Guide Project Manager to Good Project Estimation for Project Success
Dec 16, 2008
Software project management offers some guidance but no hard and fast rules, so consequently project managers have a great deal of discretion in determining what project estimating approach is taken. Those project estimates become central to producing the project schedule, which is hopefully not a work of fiction although inevitably many are exactly that. Managers need to understand project estimation and ensure project managers produce good project estimates and so increasing the chances of a successful project.
Approach to Project Estimation
Project estimating is very important and a number of different approaches have developed such as PERT [Program Evaluation and Review Technique], COCOMO [Constructive Cost Model] or Delphi Method. Whilst each technique has its strengths and weaknesses it is not always clear that any project estimating technique is sufficiently reliable to produce accurate estimates and schedules. This is primarily because:
- Each project is different in terms of objectives, scope, resources, constraints and risks
- Project planning often does not identify all of the tasks that need to be estimated
- Project estimating tends to suffer from over-optimism with estimators not allowing enough time to either do the work or solve the inevitable problems that emerge
Project Planning is too Optimistic
Project planning is based on the project scope and objectives, which should be clearly defined during project initiation, and project estimates of the tasks involved. Most projects take longer than expected primarily due to over-optimism from both the project manager and the experts producing estimates. Project managers should be encouraged to use PERT for the critical tasks of the projects if not all tasks to compensate for over optimistic estimates. All project managers will have been taught PERT using this formula:
- Expected Time = (Optimistic Time + 4x Most Likely Time + Pessimistic Time) / 6
It is simple to understand – each task is given three estimates of time to complete that task:
- Optimistic Time – everything goes well – for example 2 days
- Pessimistic Time – everything goes badly – for example 9 days
- Most Likely Time – everything goes OK with only minor problems - for example 4 days
Then the formula is used to take the average to produce a project estimate of time to complete that task – 4.5 days [(2+ 4x4 +9) / 6]
Project Estimating for Managers
Some project managers do not use any formal approach to obtaining project estimates for the project planning phase. Instead, they typically rely on task estimates from the “experts”. Unfortunately, this tends to produce the most optimistic delivery date. Managers may believe that this date is achievable or it seems a long way off and so must be easily achievable, however, often that is not the case. As a manager how do you influence the project manager to produce a better, more realistic, schedule?
Better Project Estimates for Better Project Scheduling
The key to better estimates is to look for the likely causes of estimates to be wrong and to get the project manager to compensate for them:
- Over optimistic project estimates for tasks especially for project documentation review and approval – use PERT
- Missing tasks – ensure that the project planning is comprehensive especially in defining boundaries of project and interactions with existing business systems
- Missing natural time delays: holidays, bank holidays…
- Contingency for key risks such as delayed delivery by third party suppliers
Producing better project estimates up-front is more likely to lead to more realistic project scheduling and consequently a successful project.
 |
 |
 |
