Manage your project portfolio your way

Related Links

Box Scores

Say you're an agency competing for millions of dollars in project funding. How do you start building a proposal? How are your projects going to affect your agency and the federal government as a whole?

Answering such questions can be daunting, but with the right portfolio management software it's not impossible.

Portfolio management software is similar to database software in that it only works if you have an objective. Unfortunately, portfolio software is also like database software in that it's difficult to use, and understanding how to maximize results takes time.

To determine how to get the most from portfolio management software, we tested two packages that reflect different approaches'Computas Metis and Expert Choice'and used them in the GCN Lab. Not surprisingly, we found one size does not fit all. Agencies should weigh the nature of their projects against the strengths of a software system.

Metis: Intuitive interface

One goal we had in the GCN Lab was to improve the logistics of our review process. The lab takes in up to 360 products a year for review. Needless to say, the cycle can create problems.

As soon as we started using Computas' Metis 3.4, we noticed gaps in our product submission format, and we were better able to understand past mistakes. Because Metis integrates enterprise architecture features with its portfolio management capabilities, it was able to recognize that the problems were a result of our inability to cope with last-minute changes.

The Computas philosophy on portfolio management is to unify operations and goals in a common software interface. It uses templates to create both the EA models and their subfolders, which are the portfolios and projects an agency is working on.

Using the product's generic enterprise model template, we mapped the GCN Lab infrastructure, which included personnel, network and testing resources.

With all the information in place, we noticed there was no way to disseminate information changes to the entire lab staff after a product's arrival. So we implemented a form that centralizes data for every product from its submission to when it goes back to the vendor. So far it has reduced errors by 98 percent.

The downside to the Metis philosophy is that you have to learn a new language to create models. In addition, the system requires users to enter lots of very detailed information for a model to be of much use. Gathering all the information took four weeks'and we're a small operation.

But at the end of those four weeks, we had a model that included everything from finance and budget information to network security protocols, computer infrastructure and personnel lists. And because the Metis modeling template language is highly graphical, the learning curve is low. Metis models look like Microsoft Visio diagrams. You can drag and drop items, and add virtual buttons that switch you between objects, files, folders or models.

We also liked the Metis system's ability to draw detailed relationships between objects and departments. These links enable perhaps the most useful feature of Metis, the relationship matrix, which shows in chart format how two items influence or relate to each other.

Expert Choice: Getting straight

The simple-to-use Metis interface is the software's most competitive feature'and the most noticeable difference between it and Expert Choice.

Expert Choice is a decision support platform that includes an application called Enterprise Portfolio Analysis, which takes a more straightforward approach to portfolio management than Metis.

Expert Choice was developed to make portfolio decisions by analyzing only what's necessary to achieve your goals. There are no bells or whistles, nor is the software as graphical as Metis.
Expert Choice operates more like a database program, but it is as effective as it is bland.

If you need funding for a project, or if you're dealing with numerical data, the Expert Choice system is the better way to achieve your goals.

The Portfolio Analysis program was developed using a decision-making methodology called Analytic Hierarchy Process, which uses the collective knowledge of a group to reach a decision.
For example, you can create a pros and cons list, which Portfolio Analysis turns into data that will help you arrive at decisions.

The Expert Choice approach is about quantifying data and modeling that information into a risk-versus-benefit solution. Once your data has been entered, processed and quantified, the software can present 'what if' scenarios that allow you to envision the results of your choices.

In helping with our product submission process, Expert Choice took a lot less time to learn than Metis and proved equally effective. Expert Choice helped us see what was wrong with our process as easily as Metis, and it was better at showing the risks of making other decisions.

First you decide whether you want to build your model from the bottom up using a pros and cons list, or from the top down using objects and relationships, as you would with Metis. Because the Portfolio Analysis tool isn't as graphical as Metis, the top-down approach is more difficult to use.

Then you prioritize goals and evaluate alternatives. Expert Choice offers several options. For example, if you have a lot of statistical information, alternatives can be shown in a graph. If you have little numeric data, use a descriptive rating system with words such as 'good,' 'bad' and 'worst.'

Suitability to task

The Expert Choice approach makes the most sense for agencies that rely on numerical analysis to make decisions. But in organizations such as our lab, where statistical data is scarce, it isn't as effective as Metis.

Computas wields a broadsword where Expert Choice employs a scalpel. Ideally, an agency would use both to manage the vastly different problems associated with most agency goals.

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above