Estimation Principles


You should remember the following principles when undertaking project estimates. These principles have been shown to improve estimation with any formal techniques that are being applied.

Avoid Single-Person Estimation

The estimates for a project should be undertaken by more than one person and preferably in a team-based estimation process as described in this chapter. An estimate from one person includes biases and does not benefit from the approach of the project from various viewpoints and expertise. If you estimate for another person you are making two estimation errors. The first is that you are making an estimate based on how much effort you would need ( generally flawed); second, you are estimating how different the other person is from you (seriously flawed).

Always Complete a Risk Assessment Prior to Estimation

As we discussed in Chapter 12, there are at least 70 factors that have been shown to affect project cost, quality, and duration. Any estimate undertaken without a formal risk assessment as described earlier is incorrect. At a minimum, you should never estimate without a good understanding of what we call the "Top 10" factors [4] (see Figure 14.4). These 10 factors have been shown by Jones (1994) to be the biggest influence on a project's productivity, cost, and effort.

[4] These factors are also useful as the basis of a quick and dirty risk assessment.

Figure 14.4. The Top 10 project factors

graphics/14fig04.gif

Where Possible, Use Relevant Experts

You should determine whether there have been similar projects in your organization or other organizations and whether there are experts such as vendors or consultants who could give you advice about project estimates.

Always Carefully Document Estimating Assumptions

You will always be making a number of assumptions when undertaking estimates. The Risk Assessment process covers many of the assumptions, but you should follow a careful process of documenting assumptions such as your access to key people, the relevant skill levels of your project team members , and so on.

Review the Work Breakdown Structures

As we discussed earlier, one of the most common causes of poor estimation is the missing of major tasks such as education, documentation, and so on. As we discussed in Chapter 13, you should always check and re-check your project task list.

Always Undertake Sensitivity Analysis

Your initial estimates for a project should be stated as a range from best case (i.e., everything going better than expected) to worst case (i.e., the project experiences major difficulty). The use of single-figure estimation and fixed price quotes should be avoided at Estimation Points 1 and 2 at a minimum. This is a major concern as most business and IT project managers have been taught to estimate in single figures. Other professionals involved in estimating such as accountants , economists, and quantity surveyors estimate in ranges. The use of sensitivity analysis is widely adopted in the construction, economics, and statistical areas and is well suited to IT and business projects.

Just Say No

In many cases, the best estimate is no estimate. As we cover later, it is very common (too common, in fact) for business and IT project managers to be expected to give "ballpark" estimates before they have had a chance to really understand the business requirements. This practice is a game that has been played by both business and IT management for years . The "ballpark" becomes the expectation and you have been made a sucker.



Radical Project Management
Radical Project Management
ISBN: 0130094862
EAN: 2147483647
Year: 2002
Pages: 136
Authors: Rob Thomsett

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net