Putting Portfolio Best Practices to the Test: The Verdict

Portfolio best practices

Portfolio best practices

I was recently involved in a panel that reviewed a portfolio best practices guidebook (get it here).  The panel covered a wide range of industries, including executives from Henkel, Appvion, Medtronic and Aerojet Rocketdyne.  My read of the debate is that the guidebook represents a good set of practices to aspire to, most companies do some elements of the process, and that few companies do it all well.  The official report should be out in a month or two.

A few highlights of the debate from my perspective:

The biggest takeaway was that it is critical to build your portfolio process from a quality perspective.  Portfolio processes are fundamentally about conflict resolution, in particular saying “no” to perfectly good ideas in order to fund better ones. Implement  your process, both analytically and organizationally, to deliver this result as effectively as possible.  Just comparing a bunch of projects on a graph or having a portfolio committee won’t cut it.  Ask yourself fundamental questions like: What do executives need to make decisions?  How do you compare projects?  How do you treat uncertainty?

Like all quality processes, portfolio management is a journey. Start with a process that solves a real business problem. Then improve it each cycle, always working towards a process that improves the return on the portfolio, results in aligned decisions as easily as possible, and mobilizes the organization to execute on priorities.

One of the most important, and least common, elements of a portfolio process is a calibration step. Many companies conduct evaluations of their projects and then just parade them in front of a committee of the great and the good to set priorities in a sort of justification process. Companies need to put most of the effort into ensuring the credibility and comparability of data. One way of doing this is to have an explicit calibration step, in which peers and experts review project evaluation inputs (things like market size, share, probability of success, etc.) and discuss before decisions are made.

Another critical element is the treatment of uncertainty. Many companies just throw some assumptions into a business case and calculate NPV, which results in a nearly useless measure of value.  If there is uncertainty in your portfolio, better to treat it explicitly, by using ranges and conducting sensitivity analysis. The tornado diagram tool can be very useful.  Learn more about a tornado diagram here.

Finally, processes and tools co-evolve. If you are running your portfolio comparisons on excel, you will severely limit your process. On the other hand, complex IT systems to address portfolios often impose unrealistic process requirements. Best is to use tools like portfolio navigator that are designed to implement best practice processes like the one in the guidebook.