Why portfolio management is NOT (just) project selection

FMLOGOPAGEPortfolio management and project selection

Every now and then, I find myself in discussions about “portfolio management” where the implicit definition of portfolio management turns out to be “project selection”. Apparently, portfolio management is often interpreted as a one-off decision (upfront) whether a project must be run or not.

When looking at the essence of portfolio management, I prefer the following definition:

innovation portfolio management: the decision-making process to align all innovation activities with the organization’s strategic goals for optimum value creation subject to constraints on resources and robustness.

From this definition, it follows that at least 3 flaws in the simplest interpretation of project selection:

  1. It suggests that, once selected, projects are run to completion (I am tempted to add “No matter what happens..”‘ although that often is the case). Or, in a somewhat more mature organization, projects may run until they start failing. In a well-run portfolio management process, projects run until their resources can be reallocated to improve the portfolio value.
  2. It suggests that portfolios can be build on a project by project basis; however, since projects are almost always connected in various ways, this is not true. These connections include sharing budgets and resources as well as influencing each others benefits (by cannibalization of product sales for example). Even without these dependencies, it suggests that at some point in time (the portfolio meeting) everything that needs to be known about all project candidates is known. This is never the case in reality; attempting to run a process this way means that project candidates that are not yet mature enough are completely ignored.
  3. More importantly, and possibly reinforced by portfolio theory in economics or mathematics (where portfolio management is seen as an optimization problem), it suggests that a portfolio is composed by switching projects “on” and “off”. In reality, portfolio management is about maximizing the value as much by modifying project parameters (most often start dates and resources allocated) as by just selecting them in or out of the portfolio. Of course, proper ownership of these changes by those responsible for them is needed.

So portfolio management is not just saying yes or no (all or nothing) to projects. At the very least it includes what-if analysis per project (variants as we call them) and preferably a portfolio analysis reveals options in which projects can be improved to contribute more value to the portfolio.

Portfolio management and project prioritization

Even worse, sometimes portfolio management is seen as project prioritization, and then a subsequent process is required where the prioritized list of projects is confronted with budget constraints. It has been shown extensively that a portfolio that is created by first prioritizing and then cutting off the project list when budget is depleted leads to sub-optimal portfolios in most real-life cases. All of the above arguments apply to this approach as well, and in addition the following false premise is the foundation of this approach:

Fallacy: if we try hard enough we can concoct a “project attractiveness” formula that calculates a value for each project so that a higher value for a project always means it will contribute more to the portfolio value.

No matter how hard you try to puts weights on strategic contribution, risk, financial value, and cost, this will never properly address the balance of the portfolio or the strength of the projects’ synergies (whether positive or negative). In addition, it is really tough to get consensus on such a formula: it needs to cover so many aspects of attractiveness that it looses its transparency to the typical decision maker.

See my earlier blog posts here for more complications on project prioritization and here to see how to shift portfolio management from a project-by-project review to a more meaningful portfolio-level alternatives decision-making process. Or have a look at the basic implementation of portfolio scenario composition:

8 thoughts on “Why portfolio management is NOT (just) project selection

  1. @Jac.
    Hi Jac,
    Questions for you.
    1. Just out of curiosity, how projects Max-2 and Max-3 on your bubble chart are very attractive and at the same time they have low NPV?
    2. What if there are FTE resources in addition to costs & budget? Your balancing process seems more complicated.
    3. Do you call your process – portfolio optimization?
    Hi Jack,
    Several comments.
    1. With all due respect, it seems similar definitions of portfolio management (maximize value, align resources with strategy and risk, etc.), sound like common sense and were well known for a long time before December 2007. The problem is – how practically derive “right” portfolio? We discussed this topic at many conferences.
    Your definition of portfolio management includes portfolio risk management. Just out of curiosity, how did you incorporate risk in portfolio planning, especially for interdependent projects? How did you account for risk mitigation strategies in portfolio planning?

    2. About “one click” optimization models. Serious portfolio analysis assumes several aspects.
    a. Building adequate model including multiple project profiles, multiple types of resources, strategy, and business rules including projects interdependence, etc.
    b. Providing comprehensive data analysis.
    c. Running multiple “what-if” scenarios to validate the model. BTW, each scenario corresponds to optimal or unfeasible portfolio.

    3. FYI, our portfolio planning methodology is based on combined usage of portfolio optimization (portfolio selection) and portfolio simulation models (risk assessment). Both models incorporate projects interdependence. Simulation model is used to validate portfolio selection and make portfolio risk assessment. It also incorporates multiple risk mitigation strategies (companies do it, why we cannot simulate it?). Happy to discuss details.


    1. Hi Vladimir,
      Thanks for your comments. In response to your questions:
      1. our Attractiveness score does not include but complements the financial value indicator. This means Max2 and Max3 value is high as indicated by the Attractiveness, but does not (yet) show up in the NPV dimension.
      2. FLIGHTMAP fully supports working with resource constraints (in addition to budget constraints); it is just not shown in the example.
      3. We call the process portfolio management, because it is not a one-time optimization or analysis exercise.

      Great to get so much valuable feedback from all of you!

      1. Hi Jac,
        Thank you for response.
        Let me respond.
        1. Attractiveness & NPV. In my opinion, the only projects which could fall into this category are early stage ones. (Please, give another example). There is well known bucketing mechanism (Early stage bucket vs. late stage bucket in simplest case) which could be easily implemented using real optimization tool for portfolio selection.
        2. Your portfolio management process is very similar to the process optimization model uses. The only difference that you will end up with feasible portfolio, and I will end up with optimal one. It means that I will find solution with max portfolio value. BTW, I forgot to mention, that even solutions are the same in both cases, (it is possible), time to generate optimal solution could be in several seconds vs. much more time in your case even for medium portfolio (e.g. 50-60 projects).
        3. One-time optimization does not make sense. There is always what-if analysis. My recent optimization project included 84 portfolio scenarios (each was optimal). It also included several unfeasible scenarios.
        Best regards,

  2. I would first like to appreciate the insight you have so eloquently delivered in this blog. Newbies like me really take this level of information and run with it (don´t blame us, we’re trying!). Often new PMs are confused and do not know the diference between managing a project, program or portfolio. Please le me know if I can dive a little deeper into your knowledge pool, as we have a graduate school project on portfolio management that I can certainly use your level of expertise. Again, great post Jac and great comment Jack!

    1. Hi Jose,
      Thanks for your kind words; always good to get feedback that someone appreciates my blogging. Contact me via Bicore to see how I can further help with your grad school project.

  3. Jac, Excellent review of the issues with the perception of what Portfolio Management is and what it should be. I don’t think I have ever seen it explained better. I have worked in government technology selection, military R&D portfolios, Pharmaceutical R&D portfolios, and Agriculture R&D portfolios and many of the same issues come up. * Prioritizing being equated to optimizing. * Optimizing seen as a push-button solution (this perception often comes from Ops Res experts or applied mathematicians, not from Mgmt), * Ignoring dependency between projects (risk, cost, value, timing), * Lack of project alternatives (without alternatives for a project, mgmt is forced to live with the all or nothing approach).

    Here is my company’s definition of Portfolio Management (I like your addition of Innovation Portfolio Management which we imply when we say R&D Portfolio Mgmt).

    R&D Portfolio Management: A systematic process, driven by proactive decision-making, which utilizes development project assessments and alignment with strategic goals to achieve a development portfolio with maximized value and manageable risks. (Definition developed and used by Kromite LLC since December, 2007)

    Of all of these, I would say the most commonly ignored part is the Project vs Portfolio mentality. Portfolio Objectives and constraints are the important objectives and constraints for the company. The Project Teams need to make sure that the dependencies between their project and others are identified, that resource impacts on the speed, risk, and final deliverables are assessed, and that uncertainties are brought out, not hidden. Without these three issues being addressed, Portfolio Management is handicapped – adding much less value to the company than it could.

Please share your thoughts on this blog

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s