In the portfolio management space, we seem to live on a continuum. On one side, we segment the portfolio and treat the different portfolios as separate entities. On the other side, we combine all of our work and prioritize against the overall organizational goals.
In my last post, I presented a framework for creating at least three different work portfolios within an organization:
- Mandatory – work required to keep our executives out of prison…generally considered a solid reason for any project.
- Discretionary – work that someone thought would be a good idea that may/may not be evaluated within a specific silo and/or against the needs of the overall organization.
- Baseline – work that may be rationalized against assets to ensure it supports multiple silos within the organization. In this case, the asset is prioritized, and the work inherits the prioritization at the direction of the asset management team. (Note that I know of at least one organization where “baseline” work is referred to as “mandatory,” which may add to some confusion.)
In this post, I’d like to present a framework for understanding when we want to simply lump everything into a single portfolio – or break it out into multiple silos. One of the concepts from the Gartner IT PPM Summit a couple of weeks ago comes to mind, i.e. the concept of the bimodal IT department (or, as we jokingly have been referring to it, “bipolar IT”). Per Gartner, in the next several years, up to 50% of IT spend will be directly controlled by the business. In short, IT will be split between two operating models:
- Mode 1 – the “traditional” approach of planned, risk averse projects. Projects are planned and approved, and executed to support the needs of the business, as interpreted by IT.
- Mode 2 – the accelerated approach of shrinking the request to fulfillment lifecycle and getting projects done in the fastest way possible to directly respond to the needs of the business.
The reason I bring this up is because it’s absolutely applicable to the portfolio discussion. The reality is that segmenting our portfolio allows us to do two things:
- Accelerate the project approval lifecycle. I no longer have to review the benefits of the project against all of the other projects – which in turn allows us to avoid an inevitable formal cadence where projects are suggested each year or quarter and then approved as part of structured review process. (Try enforcing something like that in an Exploration & Production company and you’ll see how quickly the business will revolt). When the portfolio is segmented, we don’t need all of the approvals and formal review schedules.
- Implement a learning strategy. As Henry Mintzberg writes in his many books on the topic, strategy is a learning process, a series of continuous experiments where ideas are tried out, feedback is gained, and that is fed back into the strategy of the organization. In this case, segmenting the portfolio allows us to build that feedback loop into the project sensing and prioritization mechanism. If all projects within an organization are prioritized against all other projects, that may impair the learning process – as each step towards understanding the strategy is hampered by an interdependence on projects approving all other strategies.
Hence, my initial stab at capturing this thought process in a visual model yields something like this (which I am sure will evolve over time).
A quick definition of terms:
- Learning Strategy –the ability to work on a strategy and prioritization mechanism that has not been perfectly articulated, i.e. to build an accelerated feedback loop into the strategy to assess how well we’re hitting our goals – and in fact whether or not we’re working towards the right goals. (Note that I considered swapping this out for market volatility, i.e. how quickly the organization needs to adapt to changing market conditions – but figured that was essentially saying the same thing. Think the stability of large utility IT vs. the short term timelines of exploration IT - and then throw in a sprinkling of how utilities are attempting to accommodate new technologies such as smart grids and solar generation.)
- Defined Strategy – a strategy that is well articulated and measured. For example, this year, our goal may be cost cutting overall. Hence, we will charter projects targeting established mechanisms of cutting costs (consolidation, rationalization, divestment, etc.)
- Speed – the need to shorten the request to fulfillment value chain and generate value in the form of deliverables in the absolute shortest time possible. Often, this results in inefficiencies as the organizations sprints to keep up with market demand – but those efficiencies are tolerated. The alternative, after all, would be losing out to the competition. Segmenting the portfolio enhances speed as it essentially pre-authorizes the funding decision….as long as the project falls within the purview of the specific portfolio segment, the portfolio owners can make their own decisions.
- Efficiency – the need to make the best use of a limited set of enterprise constraints. When efficiency reigns supreme, it’s important to slow down the approval cycle and validate all projects in the context of all other projects.
Hence, when I try to plot the conversations we’re having in the energy sector this year (depressed prices) against the conversations last year (higher prices), I end up with something like this picture.
The conclusion then is that while there is an appropriate portfolio structure for each market condition – that structure needs to shift when the market does. If we lock ourselves into a specific portfolio model and ignore external change, we won’t be able to perform our roles as effective investment advisor to the organization.