This Guest paper was submitted for publication and is copyright to Gary J. Summers © 2009
Published October 2009

Introduction | The Bayes' Law PPM Model | Proposals and Selection
New PPM Metrics | Management | Improving Your PPM Situation | Conclusion

Proposals and Selection


The quality of your proposals, P-proposals, affects the difficulty of and value created by your project selection. To consider the difficulty of project selection, suppose you have fifty proposals to choose from. If forty-five of them are Bad proposals, creating a successful portfolio is difficult. If forty-five of them are Good proposals, even random selection makes money.

Figure 2 illustrates this relationship. The horizontal axis shows P-proposals, and the vertical axis on the left shows the QPS needed to produce P-results = 80%. The dark curve shows the relationship. When P-proposals is small, achieving P-results = 80% requires tremendous skill, which no company can achieve. When P-proposals > 40% the goal becomes attainable, and as P-proposals increases further, achieving the goal becomes easy.

Now consider how P-proposals affects portfolio value. In Figure 2 the vertical axis on the right shows P-results. The light curve shows how P-proposals affects P-results when QPS = 3 (a realistic value). Increasing P-proposals raises P-results, which increases a portfolio's value.

Figure 2: How the quality of proposals affect project selection and portfolios
Figure 2: How the quality of proposals affect project selection and portfolios

Project Selection

Two aspects of PPM affect the quality of your project selection, QPS. The first aspect is the quality of your project prioritization. The quality of your prioritization is affected by uncertainty, the complexity of your proposals and your evaluation technique. Uncertainty and complexity cause evaluation errors. They make some Good proposals look like Bad ones, and vice versa. These errors decrease the quality of prioritization. As uncertainty and complexity increase, the quality of your prioritization decreases. (We will consider evaluation techniques later in this paper.)

While your prioritization is imperfect, it is still useful. The higher a proposal's evaluation, the more likely the proposal is a Good one. The lower a proposal's evaluation, the more likely the proposal is a Bad one. This quality of prioritization causes a second aspect of PPM to affect the quality of your project selection. Fortunately, this aspect is something you control completely. It is the number of proposals that you select.

To see how selecting more or fewer proposals affects the quality of project selection, consider what happens when you fund down a ranking. Suppose you select cautiously by selecting only the proposals with the highest evaluations. Your portfolio will be small, but the selected proposals are the most likely ones to be Good proposals. Your portfolio will have a high P-results, implying that QPS is high as well. If you select more proposals, these proposals will have poorer evaluations. They are less likely to be Good ones. As a result, P-results will decrease, which implies a lower value of QPS. The relationship is clear:

Selecting more proposals reduces the quality of your project selection.

Because of this relationship, you can adjust the quality of your project selection by raising or lowering cutoff values, hurdle rates or the sizes of strategic buckets. (This relationship occurs even when project interactions exist, so long as they are not too numerous.)

Figure 3 shows the impact of both effects on QPS. Consider the lower curve. If you select all proposals P-proposals = P-results. This situation implies QPS = 1 (see Bayes' law). As you select fewer proposals QPS increases. Now consider both curves. The higher curve represents better prioritization. For all levels of selection, except funding all proposals, it produces a higher value of QPS.

Figure 3: How prioritization & the number of selected proposals affect the
Figure 3: How prioritization & the number of selected proposals affect the quality of project selection (QPS)

Notice how prioritization and selection interact. Suppose that achieving your goal for P-results requires a QPS = 2. If your prioritization is poor, you must select cautiously. If your prioritization is good, you can achieve your goal while selecting more proposals. By showing how funding down a ranking affects QPS, and thus P-results, Figure 3 fulfills a need of PPM. It is the first metric that evaluates project prioritization and shows its impact on the portfolio.

The Bayes' Law PPM Model  The Bayes' Law PPM Model

Home | Issacons | PM Glossary | Papers & Books | Max's Musings
Guest Articles | Contact Info | Search My Site | Site Map | Top of Page