Best Practices Structure

The first three entries under the Matrix for Best Practices describe the three kinds of schedule improvements:

  • Potential reductions from nominal schedule
  • Improvement in progress visibility
  • Effect on schedule risk

The following paragraph explains these three entries and the rest of the entries in the best-practice matrix section.

  • Potential reduction in nominal schedule
    This entry provides an estimate of the effect that use of the practice will have on a project’s schedule. The potential reductions all assume that the practice is executed properly. Projects that use a practice for the first time will usually make mistakes that will cut into the practice’s efficacy.
  • Improvement in progress visibility
    It is difficult to pin down something as amorphous as “improvement in progress visibility”, and I have created the best approximation I can think of defining the improvement as the percent of the project that a practice makes more visible than a traditional waterfall lifecycle model would.

The ratings for this category arise from my best estimate. Once again, there is an approximate correspondence between the verbal ratings and the underlying quantifications but the verbal ratings better convey the inexactness of the data.

  • Effect on schedule risk
    Some practices such as evolutionary prototyping generally decrease development time compared with traditional methods, but they make it more difficult to predict specifically then the project will be finished.

A traditional development effort might take an average of 3 months and vary by plus or minus 2 weeks. An evolutionary-prototyping approach to the same project might take an average of 2 months and vary by plus 5 weeks/minus 2 weeks.

I have included some practices as Best Practices specifically because they have a strong effect on schedule risk. They might have little or no effect on average schedule length but they will dampen wild schedule fluctuations and help to bring-of-control schedules under control.

  • Chance of First Time Success
    Some the practices are more difficult to learn than others. We can expect to be immediately successful with some practices; for others we’ll probably have to settle for delayed gratifications.

A few practices (such as Reuse) require a substantial investment in infrastructure before they begin to pay off. Those have virtually no chance or “first-time” success and are rated as low or Poor, even though they have terrific long-term potential.

  • Chance of Long-Term Success
    Even when we factor out the learning curve effect, some practices are simply successful more often than others. This rating describes the chance that a practice will be successful if we stick with it long enough to become proficient at using it.

A comparison between the rating in this category and the “Chance for First Time Success” category will give us an indication of the steepness of the learning curve associated with the practice.

Some practices such as “Top-10 Risk”, have the same rating for both first-timers and long-term success. Those practices are exceptionally easy to learn to use. Others such as “Designing for Change”, have a difference of more than one rating level, such practices are relatively difficult to learn to use.