Help Contents


Quick Start Guide

Portfolio Lists





Bubble Charts

Bar Charts

Ranking Charts

Efficient Frontier Charts

Sensitivity Charts


Using the Prioritization Engine in Optsee®

The Optsee® Prioritization Engine automatically ranks your projects based on a SMART Score calculated from the actual project data and weights that you assign to that data. The higher the weight, the more influence that attribute has on the Overall SMART Score.

See also SMART Prioritization Using Simulations on how to use Monte Carlo simulations to visualize and manage portfolio and project risk.

For example, if "Profit" is assigned a weight of "1000" in an Attribute form and "%Probability of Success" is assigned a weight of "500," then "Profit" is going to have twice the influence on the total SMART Score as "%Probability of Success." If two projects have identical "Profit" values, than the project with the higher "%Probability of Success" value will have the higher Overall SMART Score.

Optsee® uses a rigorous scoring system developed by researchers from Harvard, MIT, and the University of Southern California called SMART (Simple Multi-Attribute Rating Technique). The SMART system in Optsee® uses weighted normalized SMART Score value curves to compare projects. For example, in the value curve for "Profit" illustrated below, the relative "worst" outcome is the lowest profit and the "best" outcome" is the highest profit. Projects with the highest profits would contribute 100 value points and projects with the lowest would contribute 0 value points to the Overall SMART Score. Projects with Profit between the "Best" and "Worst" would have SMART Scores calculated along the straight line between 0 and 100.

The Total Value Score is the sum of all the individual value scores multiplied by their normalized weights.

You can adjust your individual value score curves by using the slider on the Attribute form. Sliding it to the right creates an "Increasing Rate of Return" compared to the straight line "neutral" curve such that better projects have a proportionally higher rate of return than worse projects. For example, if the attribute were "profit," the increasing the rate of return would create a bias for higher profit projects in the "profit" value score.

Similarly, sliding it to the right creates a "Decreasing Rate of Return" compared to the straight line "neutral" curve such that better projects have a proportionally lower rate of return than worse projects. Adjusting for a decreasing rate of return is useful when small increases in the "better" projects are not a valuable as small increases in "worse" projects.

You can select 4 different kinds of curves: Linear, S-Type (Logistic), Step, and Custom. All these curves are adjustable.

Essentially, here is how the Overall SMART Score for each individual project is calculated:

1. The individual value for each project"s attribute value is calculated according to selected value curve. These are the individual unweighted SMART Score values.

2. The individual unweighted SMART Score values are adjusted in proportion to their relative weights or importance that you have assigned. These are the weighted SMART Score values.

3. The weighted SMART Score values are summed together to yield the Overall SMART Score value for the project.

Click here to read a more detailed description of how the Overall SMART Score for each individual project is calculated

You can toggle between the actual attribute values and the individual weighted SMART Score values by clicking the [Calc.] and [Actual] button on the Portfolio Data Form.

1. Click [Calc.] to display the "normalized weighted SMART Score values" in the attribute columns. Note that the values in the attribute columns sum to the number in the "Value Score" column:

You can view these values in your bubble charts by clicking on the "Use Weighted Value" check boxes in the new bubble chart form. See "Creating a Bubble Chart" for more details.

2. Click [Actual] to toggle back to display the actual values are displayed in the attribute columns.

Therefore, using the prioritization engine gives you an objective relative ranking of your projects based on the importance that you assign to the attributes by the weights. We recommend experimenting with different weighting models using a small group of projects to find a set that ranks those project the way you would. Then you can have more confidence in using those weights on a larger portfolio of projects.

Two important factors to keep in mind:

• Don"t weight two or more Attributes that basically describe the same thing, like "Profit" and "ROI" - in other words, don’t double-count! Only weight the attribute that best represents the value and will be the most influential.

• Don"t weigh too many attributes or you will dilute the effectiveness of the model. Four to seven weighted attributes will give you the best results.

Next: Using the Prioritization Bar Charts