A good way to begin understanding what portfolio management is (and is not) may be to define the term portfolio. In a business context, we can look to the mutual fund industry to explain the term's origins. Morgan Stanley's Dictionary of Financial Terms offers the following explanation:

If you own more than one security, you have an investment portfolio. You build the portfolio by buying additional stocks, bonds, mutual funds, or other investments. Your goal is to increase the portfolio's value by selecting investments that you believe will go up in price.

According to modern portfolio theory, you can reduce your investment risk by creating a diversified portfolio that includes enough different types, or classes, of securities so that at least some of them may produce strong returns in any economic climate.

Note that this explanation contains a number of important ideas:

- A portfolio contains many investment vehicles.

- Owning a portfolio involves making choices -- that is, deciding what additional stocks, bonds, or other financial instruments to buy; when to buy; what and when to sell; and so forth. Making such decisions is a form of management.

- The management of a portfolio is goal-driven. For an investment portfolio, the specific goal is to increase the value.

- Managing a portfolio involves inherent risks.

Over time, other industry sectors have adapted and applied these ideas to other types of "investments," including the following:

Application portfolio management: This refers to the practice of managing an entire group or major subset of software applications within a portfolio. Organizations regard these applications as investments because they require development (or acquisition) costs and incur continuing maintenance costs. Also, organizations must constantly make financial decisions about new and existing software applications, including whether to invest in modifying them, whether to buy additional applications, and when to "sell" -- that is, retire -- an obsolete software application.

Product portfolio management: Businesses group major products that they develop and sell into (logical) portfolios, organized by major line-of-business or business segment. Such portfolios require ongoing management decisions about what new products to develop (to diversify investments and investment risk) and what existing products to transform or retire (i.e., spin off or divest).

Project or initiative portfolio management: An initiative, in the simplest sense, is a body of work with:

- A specific (and limited) collection of needed results or work products.

- A group of people who are responsible for executing the initiative and use resources, such as funding.

- A defined beginning and end.

Managers can group a number of initiatives into a portfolio that supports a business segment, product, or product line. These efforts are goal-driven; that is, they support major goals and/or components of the enterprise's business strategy. Managers must continually choose among competing initiatives (i.e., manage the organization's investments), selecting those that best support and enable diverse business goals (i.e., they diversify investment risk). They must also manage their investments by providing continuing oversight and decision-making about which initiatives to undertake, which to continue, and which to reject or discontinue.

The article will focus primarily on initiative portfolio management. Let's begin by looking at how most businesses decide what initiatives to fund and pursue.

Enterprise business strategy: The starting point

For many years, businesses have defined and used processes to create and qualify goals they must achieve in order to prosper within a specific time period. These goals might include provisions to:

- Improve revenue overall or within a specific business segment.

- Create, refine, or retire products.

- Improve the current cost structure or mix.

- Gain additional market share.

Definitions of business goals may also specify achievement metrics, such as: "improve overall revenue by five percent within the next two years " or "improve the cost structure via a five percent reduction in labor costs within the next year. "

Managers typically record these goals and metrics, along with their evolving understanding of them, in an enterprise business strategy document. This document also spells out results the business must achieve in order to prosper.

As an enterprise constructs, discusses, and refines a business strategy, its focus shifts from answering the question: "What goals and related specifics must we achieve?" to answering the follow-on question: "What mechanisms will we create and implement to reach these goals?"

One such mechanism is an initiative. As we noted above, an initiative is a mechanism businesses use to both frame and drive efforts toward achieving stated goals (or goal components) in the enterprise business strategy -- which should always serve as a starting point and continuing source of reference.

Along the path to business success

So far, we have defined a logical, simple progression of efforts to achieve success as a business entity:

- Identify business goals.

- Qualify and quantify the business goals.

- Define initiatives to enable and achieve the business goals.

However, when businesses begin to carry out these efforts, they must conduct additional activities, which we will discuss below.

Choosing among competing initiatives

Initiatives may arise from various sources within an organization or enterprise, including businesses or business segments that intend to contribute directly to the enterprise business strategy. The stimulus for an initiative may be either internal or external; it might be a client request to change an existing product or a new regulation imposed by an industry body. In addition, initiatives may be responses to changing conditions or needs within the organization, such as the introduction of new technologies or the re-organization of a business unit.

Enterprises need a process to help them make sense of these competing demands, relate them to enterprise business strategy goals and goal components, and match them with the finite resources -- of all types -- that are available.

In one organization I know of, the CIO learned that a competing organization's IT group was on a path to improve software delivery capability. That organization's CIO was implementing process improvement by means of the (then current) Software CMM model (SWCMM), and had set a twenty-four-=month (typical industry performance) deadline for achieving the next level.

Determined not to be beaten, the first CIO called in his direct reports and issued a mandate: His IT organization would achieve Level 2 capability in twelve months. To achieve this goal, managers would have to either defer or slow down high-priority software projects. This did represent a choice among competing priorities, but the options were based solely upon the CIO's perception of importance and worked poorly for the organization. The organization's major software projects -- which were aligned with important components of the enterprise business strategy -- lost resources, had to perform additional work, and were subject to many additional capability reviews. In the end, although the IT organization did achieve the next capability maturity level in twelve months, it lost credibility with internal customers because it failed to carry out urgently needed projects on schedule, and as planned.

Assessing initiatives

To understand which new initiatives to pursue -- and which existing initiatives to continue pursuing -- enterprises must conduct systematic assessments, using the set of goals, goal components, and achievement metrics in the enterprise business strategy as major reference points.

From a basic (and vastly overly simplistic) perspective, initiatives that contribute to a specific goal, goal component, or achievement metric are "good"; all others are "less than good." Of course, in practice, making the judgment is typically not that simple.

To review both potential new initiatives and (periodically) each existing initiative effectively, the enterprise needs to compare the initiative against an agreed-upon set of measures that quantify, in multiple dimensions, its "value" contribution. In other words, this composite metric measures the initiative's potential support for a goal or goal component in the enterprise business strategy.

Checking for links to the business strategy

An enterprise business strategy addresses several business areas and their associated goals. So the strategy is really a collection of related items, each supported or enabled by goals, goal components, mechanisms, proposed changes and transformations, and multiple initiatives. Given this complexity, it can be challenging for the organization to devise an effective review process to continuously assess and confirm that both proposed and existing initiatives are firmly attached to specific (and evolving) enterprise business strategy goals or goal components.

For example, one organization I know of was required (as a matter of business practice) to modify its core set of application software systems each year to accommodate its customers' new and changing business requirements. Achieving this goal was crucial to the company's continuing business success. Some of its competitors were already doing the same thing, more successfully. Accordingly, the organization began giving initiatives linked to this goal top priority in terms of funding and resources -- in other words, classifying them as "good" initiatives. Not surprisingly, soon more than half of all approved initiatives appeared to be linked to this business goal. However, upon close scrutiny, many of them were actually aimed at filling other needs. Those who proposed these initiatives simply realized that linking them to the annual customer change requirements process was (almost) a blank check for approval, funding, and resources. Even worse, some initiatives the company actually needed to support customer changes were denied because their proposals did not sufficiently emphasize the link.

Apportioning resources

Although an organization should do everything in its power to create and execute an effective enterprise business strategy, it must also recognize that its resources -- staff, outside consultants and contractors, infrastructure, capital, and human expertise -- are finite. Competing needs within the business will reduce the share of resources available to achieve strategic goals; so it is important to quantify their relative importance and define benchmarks for apportioning these resources.

In addition, organizations need a dynamic measurement mechanism to provide -- at multiple intervals and points -- a snapshot of the resources each initiative is consuming. To be fair, these measures should consider the initiatives' linkages to various goals and goals components, which in turn should be ranked by priority.

Providing control and oversight

Suppose an organization could identify a single group of enabling mechanisms and initiatives at the time it refines and completes its enterprise business strategy. And further suppose the organization then proceeded to execute that group of mechanisms and initiatives successfully, without any changes. In such a case, the organization could use a simple progress checklist to exercise control over the mechanisms and initiatives.

Unfortunately, things do not work this way in real organizations. As we noted above, at any point in time, some initiatives are in progress, proposals for new initiatives are being considered, and the business environment is subject to constant changes, which have varying degrees of impact upon the organization and its initiatives. In this complex, fluid environment, providing an appropriate measure of oversight and control can be a major challenge.

One organization I know of fully embraced the idea of oversight and control -- perhaps too enthusiastically. Every proposed new initiative required approval by multiple organizational entities, and the process involved multiple review-and-approval "events." Eventually, the Activity of preparing business cases and justification models became an end in itself for the teams proposing these initiatives; they devoted weeks of effort to it, and their presentations consumed hours of managers' time. Unfortunately, that left little time for the managers (and teams) to investigate what the proposed initiative was really about, and approval was typically granted to initiatives with the most creative, colorful presentations and financial models. Later, as the discovery process uncovered the unexpected, this led to a disruptive, expensive churn of contents and direction.

Basic concepts and components for portfolio management

Now that we understand some of the basic dynamics and inherent challenges organizations face in executing a business strategy via supporting initiatives, let's look at some basic concepts and components of portfolio management practices.

The portfolio

First, we can now introduce a definition of portfolio that relates more directly to the context of our preceding discussion. In the IBM view, a portfolio is:

... one of a number of mechanisms, constructed to actualize significant elements in the Enterprise Business Strategy.

It contains a selected, approved, and continuously evolving, collection of Initiatives which are aligned with the organizing element of the Portfolio, and, which contribute to the achievement of goals or goal components identified in the Enterprise Business Strategy.

The basis for constructing a portfolio should reflect the enterprise's particular needs. For example, you might choose to build a portfolio around initiatives for a specific product, business segment, or separate business unit within a multinational organization.

The portfolio structure

As we noted earlier, a portfolio structure identifies and contains a number of portfolios. This structure, like the portfolios within it, should align with significant planning and results boundaries, and with business components. If you have a product-oriented portfolio structure, for example, then you would have a separate portfolio for each major product or product group. Each portfolio would contain all the initiatives that help that particular product or product group contribute to the success of the enterprise business strategy.

The portfolio manager

This is a new role for organizations that embrace a portfolio management approach. A portfolio manager is responsible for continuing oversight of the contents within a portfolio. If you have several portfolios within your portfolio structure, then you will likely need a portfolio manager for each one. The exact range of responsibilities (and authority) will vary from one organization to another, 1 but the basics are as follows:

- One portfolio manager oversees one portfolio.

- The portfolio manager provides day-to-day oversight.

- The portfolio manager periodically reviews the performance of, and conformance to expectations for, initiatives within the portfolio.

- The portfolio manager ensures that data is collected and analyzed about each of the initiatives in the portfolio.

- The portfolio manager enables periodic decision making about the future direction of individual initiatives.

Portfolio reviews and decision making

As initiatives are executed, the organization should conduct periodic reviews of actual (versus planned) performance and conformance to original expectations.

Typically, organization managers specify the frequency and contents for these periodic reviews, and individual portfolio managers oversee their planning and execution. The reviews should be multi-dimensional, including both tactical elements (e.g., adherence to plan, budget, and resource allocation) and strategic elements (e.g., support for business strategy goals and delivery of expected organizational benefits).

A significant aspect of oversight is setting multiple decision points for each initiative, so that managers can periodically evaluate data and decide whether to continue the work. These "continue/change/discontinue" decisions should be driven by an understanding (developed via the periodic reviews) of a given initiative's continuing value, expected benefits, and strategic contribution. Making these decisions at multiple points in the initiative's lifecycle helps to ensure that managers will continually examine and assess changing internal and external circumstances, needs, and performance.

Governance

Implementing portfolio management practices in an organization is a transformation effort that typically involves developing new capabilities to address new work efforts, defining (and filling) new roles to identify portfolios (collections of work to be done), and delineating boundaries among work efforts and collections.

Implementing portfolio management also requires creating a structure to provide planning, continuing direction, and oversight and control for all portfolios and the initiatives they encompass. That is where the notion of governance comes into play. The IBM view of governance is:

An abstract, collective term that defines and contains a framework for organization, exercise of control and oversight, and decision-making authority, and within which actions and activities are legitimately and properly executed; together with the definition of the functions, the roles, and the responsibilities of those who exercise this oversight and decision-making.

Portfolio management governance involves multiple dimensions, including:

- Defining and maintaining an enterprise business strategy.

- Defining and maintaining a portfolio structure containing all of the organization's initiatives (programs, projects, etc.).

- Reviewing and approving business cases that propose the creation of new initiatives.

- Providing oversight, control, and decision-making for all ongoing initiatives.

- Ownership of portfolios and their contents.

Each of these dimensions requires an owner -- either an individual or a collective -- to develop and approve plans, continuously adjust direction, and exercise control through periodic assessment and review of conformance to expectations.

A good governance structure decomposes both the types of work and the authority to plan and oversee work. It defines individual and collective roles, and links them to an authority scheme. Policies that are collectively developed and agreed upon provide a framework for the exercise of governance.

The complexities of governance structures extend well beyond the scope of this article. Many organizations turn to experts for help in this area because it is so critical to the success of any business transformation effort that encompasses portfolio management. For now, suffice it to say that it is worth investing time and effort to create a sound and flexible governance structure before you attempt to implement portfolio management practices.

Portfolio management essentials

Every practical discipline is based on a collection of fundamental concepts that people have identified and proven (and sometimes refined or discarded) through continuous application. These concepts are useful until they become obsolete, supplanted by newer and more effective ideas.

For example, in Roman times, engineers discovered that if the upstream supports of a bridge were shaped to offer little resistance to the current of a stream or river, they would last longer. They applied this principle all across the Roman Empire. Then, in the Middle Ages, engineers discovered that such supports would last even longer if their downstream side was also shaped to offer little resistance to the current. So that became the new standard for bridge construction.

Portfolio management, like bridge-building, is a discipline, and a number of authors and practitioners have documented fundamental ideas about its exercise. Recently, based on our experiences with clients who have implemented portfolio management practices and on our research into the discipline, we have started to shape an IBM view of fundamental ideas around portfolio management. We are beginning to express this view as a collection of "essentials" that are, in turn, grouped around a small collection of portfolio management themes.

For example, one of these themes is initiative value contribution. It suggests that the value of an initiative (i.e., a program or project) should be estimated and approved in order to start work, and then assessed periodically on the basis of the initiative's contribution to the goals and goal components in the enterprise business strategy. These assessments determine (in part) whether the initiative warrants continued support.

This theme encompasses the notion that initiative value changes over time. When an initiative is in the proposal stage, it is possible to quantify an anticipated value contribution. On this basis (in part) the proposed initiative becomes an approved initiative. But what about an initiative that is a large program effort, with a two-year duration? It is highly unlikely that the program's expected value will remain static during the entire two-year period, so continuous value monitoring is necessary. From this, we can derive an essential statement:

Initiative value changes and requires continuous monitoring over the life of the initiative.

In a future article, I will provide a more extensive discussion of portfolio management essentials, as seen from an IBM viewpoint.

Conclusion

As a project-based enterprise, IBM has extensive experience in applying portfolio management techniques and tools for its company business. In addition, IBM is a significant player in the portfolio management marketplace, offering the IBM Rational Portfolio Manager along with related services.

Recently, based on client and internal experiences, a survey of ideas, and research and development efforts, IBM has developed a model of the types and nature of specific work efforts that fall within the boundary of portfolio management. IBM is also conducting a variety of efforts aimed at identifying a body of methods and practices that define, enable, and integrate the work of portfolio management.

This is the first, in a series of articles that will deal with various aspects of the practice of portfolio management, as well as the enabling body of definitions, models, and methods that guide the execution of effective work effort in that space. For more information go to The Rational Edge .

Notes

1 Portfolio management experts have differing views about the scope of responsibility for the portfolio manager's role.

About the author

Michael F. Hanford is the chief methodologist for the IBM SUMMIT Ascendant methodologies and a member of the IBM Rational commercial methods content team. He has also worked as a methodology author, a manager for large consulting engagements, and a leader of enterprise process assessment and transformation efforts for IBM Rational and PriceWaterhouseCoopers Consulting (PWCC). Prior to joining PWCC, he was director of software engineering practices for Fidelity Investments Systems Company

Michael F. Hanford, Chief Methodologist, SUMMIT Ascendant Methodologies