SOFTWARE DEVELOPMENT
The Contingency Allowance
To demonstrate, let us take a very simple, but realistic example. The development
time for a small project is estimated to be 6 hours. It has reasonably taken into
account all known factors (function points, user requirements, etc, etc).
This is the likely outcome for a number of such projects:-
|
On time
|
5% |
0% |
|
1 hour late
|
10% |
17% |
|
2 hours late
|
20% |
33% |
|
3 hours late
|
40% |
50% |
|
4 hours late
|
20% |
67% |
|
5 hours late
|
5% |
83% |
Optimistic Estimations
These figures highlight the problem of estimation. The initial estimate is based
on an optimistic viewpoint - if all goes exactly as predicted, the project will
come in on time. But 95% of the time, unanticipated problems will cause a delay.
For a small project, the consequences of an underestimate may be minimal. But for
a large project, when the very existence of a company is threatened, much more conservative
estimates are needed.
What is required is a Contingency Allowance to be applied to the initial "most reasonable"
estimate. This Contingency Allowance should, at a minimum, result in a project completing
on time, at least 50% of the time. This means that half the time the project estimator
will be a hero, and half the time a villain.
A more professional choice would be a Contingency Allowance resulting in success
95% of the time. So, in this example, the estimate should be increased by 5 hours,
an increase of more than 80%.
The Expanding Funnel of Doubt
The Contingency Allowance must take into account the unpredictable nature of complex
projects.
The graph shows that the more complex the project, the more the funnel of doubt
expands. In other words, the completion date becomes less and less predictable.
Setting the Contingency Allowance
The figure that is to be used is completely subjective. My advice - the Contingency
Allowance must be set conservatively to ensure the success of the project. Automation
of administration processes must provide benefits to more than cover the estimate.
If not, go back to the drawing board, and design a less ambitious project.
The more conservative the estimation, the greater the probability of success. And
of course, the greater the resistance by management.
|