Project Management Institute

Creating breakthrough thinking with metrics

This article is copyrighted material and has been reproduced with the permission of Project Management Institute, Inc. Unauthorized reproduction of this material is strictly prohibited.


Organizations need to identify, create, and deliver outstanding results. The most radically-improved results are breakthroughs. Breakthroughs are an entirely new generation of a product or process, and are “breakthrough in the sense that their core concepts and technologies break new ground for the organization” (Wheelwright, 1992, p50). Experts often contrast breakthrough innovation from incremental innovation due to its radical nature, and characterize it as a “discontinuity” in the trajectory of improvement (Leifer, 2000, p19).

In studying organizations that are among the vanguard in innovation, Robert Tucker asks and answers an important question: “What motivates managers and leaders across the organization to embrace innovation broadly and give it ongoing priority? Metrics.” (Tucker, 2002; p6)

This paper describes an approach for developing effective metrics, with particular emphasis on breakthroughs. Metrics are a catalyst for creating improvements, both incremental and breakthrough. Metrics are not a “data dump,” nor a fear-inducing, command-and-control tool.

This paper rests on a proposition that metrics are a message intended to spur actions by others. Robert Eccles and his co-authors argue that one of the essences of management is the skillful use of rhetoric. They describe it in this way: “To view management from a rhetorical perspective is to recognize that the way people talk about the world has everything to do with the way the world is ultimately understood and acted in, and the concept of revolution and change depends to a great extent on how the world is framed by our language” (Eccles, 1992, p29, italics added). In other words, language matters; people's understanding of their world is based on language. The job of managers is to use rhetoric to shape the way people understand their situation, and motivate people to action. Language is an important element of revolution and change.


A Structured Process for Creating Metrics

On the surface, identifying metrics is an interesting and attractive concept. However, it requires sustained efforts to get results. It is not a magic bullet.

The following paragraphs describe a structured process for systematically creating improvements through metrics. The process facilitates individuals and organizations selecting appropriate metrics and supporting the chosen metrics. Participants are individuals and small groups.

Step 1) Select the Audience

Metrics are a type of message that communicates information to an audience, who will then act on the received information. First, it makes sense to determine the audience for the metric. Typically, participants brainstorm a list of potential audiences, selecting a smaller number of people for further consideration. Because multiple stakeholders are always present, there will be multiple evaluations of performance.

Examples of audiences would include the individual, their boss, their customer, their user, their senior management team, etc. It is advisable to identify a diverse set of stakeholders. Do not limit the audience to a chain of command like the boss, the boss’ boss, the boss’ boss’ boss, etc. It is better to identify audiences that include your external customer, a regulatory agency, and so forth. Experience shows that sometimes the data is available elsewhere in the organization, or suggests a valuable improvement opportunity.

I also ask people to explain underlying context such as market conditions, organizational changes, and so forth to assure that they are thinking through the implications of their choices. This examination often reveals many unchallenged assumptions that offer the potential to get unstuck.

Step 2) Clarify Performance

In Step 2, participants decide where they want to focus their efforts. Step 2 could be performed in parallel with Step 1, but requires much more perspective and thoughtfulness.


Experience shows that it is hard for most people to prioritize. Each individual in an organization has many pressures on them, and have a hard time ignoring urgent but unimportant requests. Further, each audience is likely to focus on different elements of performance. The goal is to answer the question, “What is important to each of the audiences?”

There are several dimensions to performance:

  • The area of performance. There are many areas where organizations measure performance. For example, firms can measure process performance, project performance, portfolio performance, enterprise performance, individual performance, etc.
  • The level of performance. I am using the word “level” to refer to the amount of detail/abstraction as well as the scope of the innovation, that is, incremental, substantial, or breakthrough. Experience shows that the chosen metric must match the situation. Exhibit 1 illustrates these levels.
There are many levels of performance

Exhibit 1 - There are many levels of performance

A few comments on the difficulties of organizational performance measurement are in order. Many people struggle with quantifying performance. Why do they struggle? The majority of people in organization focus on activities – the process of doing – rather than the outcomes of their activities. This seems to be especially prevalent in organizations that are driven by rules and procedures – that is, bureaucratic organizations. Paradoxically, these activity-driven people often are the first to claim that they are “results driven.” Asking the question, “What is performance?” turns out to be very difficult for many people. What people do (the activity) is different than what they accomplish (the result). People need to achieve outcomes, so Step 2 forces people to answer this question:

What do you need to be able to accomplish?

Experience shows that the pursuit of breakthroughs requires conscious choice. Managers need to decide whether they want to follow one of the following three strategies:

  • Strategy 1- Extrapolation of past performance (incremental). Incremental, continuous improvement efforts seldom lead to breakthroughs.
  • Strategy 2- Jump to a new level by identifying a function that should be provided but is not (breakthrough)
  • Strategy 3- Jump to a new level by “raising the bar” to a level of performance that hasn't been seriously contemplated (breakthrough)

Participants ultimately need to resolve and refine the dimensions of metrics considered, because having too many things to measure causes a loss of focus.

Step 3) Define Performance Outcome Metrics

Looking broadly across industries and applications, I discovered that organizations use hundreds of measures. Here are a few observations on measurement:

  • Organizations don't lack for ideas about what to measure
  • Organizations do not lack data. Automation of systems increases the amount of data that is available to the manager.
  • The presence of measurement – in itself - does not signify good management
  • Many people tend to (erroneously) think that the only good measures are those that are quantitative

In every organization, there are some people who would rather look at an unimportant particle and measure it precisely, than imprecisely measure an important indicator. There is a tradeoff here: very precise measures are typically “backward looking” and tell the organization where it has been. They don't induce action. Instead, predictive metrics give managers information so they and set and achieve goals. The rhetorical perspective presented earlier leads to this definition: a metric is a measure that has meaning. Metrics are to information as measurements are to data. Pragmatically, a measure is no good unless it fosters some action by members of the organization. The characteristics of a good metric are:

  • It measures something important
  • It has relevance to people
  • It measures something that is directly controllable and is resistant to gaming
  • It incents the organization in the proper direction
  • It is a member of a very small, lean family of measures

Many organizations select their metrics from benchmarks, because benchmarks can give managers a quick indicator of the things that they should be focusing on. My research on metrics shows, “Most thought leaders suggest that you should view benchmarks – especially those external to your organization - with some skepticism. No one is against setting an aggressive stretch goal for performance; just make sure you understand the data that underlies the benchmark” (Githens 2003b).

Step 4) Identify Enablers

Enablers are skills, practices, and capability that cause an outcome (result). As you answer this question, you are identifying enablers: “What things cause your goal to happen?” To create change through metrics, managers treat the following axiomatically:


You cannot measure what you cannot describe. You cannot describe what you do not understand, nor can you manage what you do not understand.

Experience shows that some people find modeling their organization and selecting enablers is straightforward and easy, and others find it more difficult. I believe that organizations with effective measurement systems have people with some inherent ability to think abstractly and systematically, and have sufficient time to “model” their organization's behavior and performance.

In addition to individual's inherent ability, it is also important to have tools for modeling performance and improvement. For example, the tool of benchmarking can help to establish and validate a model. However, benchmarking is a rigorous endeavor that requires commitment. An important rule is to define/understand the organization's own process first. Start by answering these questions: What aspect of the process should we benchmark? Whom should we benchmark? The real value of benchmarks and benchmarking are in how they help a firm with a functioning process promote inquiry and self learning.

As managers develop this deepened understanding of enablers, they find it easier to identify predictive metrics and focus attention on them.

Step 5) Develop the Scorecard

The scorecard is the presentation of metric to the intended audience. Because scorecards are visible, I often find that scorecards are the thing to which people gravitate. In this step, participants address some practical questions:

  • Where does the data come from? Is it important data? Is it believable data?
  • Is the metrics presentation format understandable and usable by the audience?

Like benchmarks, people make the mistake of oversimplifying and picking methods that are inappropriate to their situation. Regardless, scorecards provide a place to get started. As people get progressively more comfortable with the scorecard's reporting format they can then start to probe deeper into the underlying causes.

This step is where the difference between predictive and reactive metrics is very apparent. It makes sense for a manager to try to identify predictive metrics and focus attention on them. Hopefully the organization has selected some predictive metrics and these predictive metrics should appear in the scorecard.

Step 6) Develop Influencing Strategies

Information has to be relevant to the individual; this relevance creates motivation for understanding and action. (Githens, 2003a) In this step, participants develop answers to questions like the following:

  • What methods will managers use to communicate expectations and metrics?
  • What will the manager do if they see indicators of poor performance?
  • How can managers get alignment?
  • What resources need to be provided?
  • What incentives can be provided?
  • What performance barriers need to be removed?
  • How susceptible is the metric to gaming?

Breakthroughs often require a change to the paradigm where people reframe the issues and boundaries. Breakthroughs seldom come easily, especially in large, stable, complex organizations. The claim that “people resist change” is an oversimplification. The extent that individuals and groups support change depends on a number of factors, including their perception of resources availability, perception of and attitude towards risk, and experience, among other things.

Not surprisingly, leadership influences the organization's ability to innovate. Robert Tucker (Tucker, 2002, p2) observes, “Because no one is in charge of innovation, or of setting stretch goals, the focus is on risk-adverse, incremental improvements.” Many managers avoid ambiguity and focus solely on their own department's issues. People generally want to have consistent measures over time so that they can examine trends. This has proven to be helpful when examining data for complex phenomena like stock market trends or global warming. However, as Eccles points out (Eccles, 1992, p156), when managers focus on stability of measures over time, they tend to prefer to “monitoring” data rather than make a decision and take action, they “game” the measures, and ignore the underlying (often-complex) causal factors in favor of “the numbers.”

Step 7) Create Team and Organizational Learning

I define culture as the “shared learning of a group of people.” Learning is the engine for organizational improvement; it aligns behaviors on the right things, assuring that the organization achieves the desired performance. The final stage of the process is to facilitate the decision makers to identify strategies for organizational learning to adopt the metric and the behaviors. This is necessary because

  • Removing barriers is essential to creating improvement
  • Breakthroughs require some un-learning and reframing
  • Incremental improvements need to be locked into the day-to-day behavior

Inertia is the characteristic of organizations that could be termed “keep doing what you've been doing.” Organizations must overcome inertia if they are to innovate. Team learning helps to create sustainable improvements to address the underlying forces of inertia.

Breakthrough thinking is a capacity for generating breakthroughs. James Adams considers thinking as a skill, “The time honored method of improving one's skill is to be continually conscious of one's performance and seek to improve it – usually according to an ideal or standard of what is desirable” (Adams, 1974, p1). Adams believes that recognizing conceptual blocks to creative thinking is an essential skill. He defines conceptual blocks as, “Mental walls which block the problem solver from correctly perceiving a problem or conceiving a solution” (Adams, 1974 p11). Exhibit 2 describes five types of blocks that should be recognized as organizations go about creating change. It includes some metrics development implications.

Blocks with Implications for Metrics and Breakthrough

Exhibit 2: Blocks with Implications for Metrics and Breakthrough

Conclusion – Three Essential Elements

Metrics help managers improve how they think about performance. As Exhibit 3 illustrates, metrics, inquiry, and strategy interact to characterize superior organizational performance.

There are many interactions in creating superior performance

Exhibit 3 - There are many interactions in creating superior performance

Good metrics help managers focus. Returning to Eccles’ perspective, “A rhetorical view of management, it must be stressed, does not absurdly deny the existence of facts. It merely asserts that whatever these facts are, their importance and meaning are only established through language” (Eccles, 1992 page 290).

Considering metrics gets the organization to ask better questions about what is important. I often ask people to describe how they got to their present set of metrics. “Did your organization stumble into them, or were they generated as part of a thoughtful process?” The answer suggests that organizations “muddle through” with an unstructured problem-solving strategy, and then pick the first good answer that comes along, without considering the organization holistically. The alternative approach described here is more likely to generate effective metrics, because metrics help people focus on the right things.

These better questions lead to strategic insights. Conversely, weak thinking leads to oversimplification and little ability to achieve strategic impact. The metrics approach described in this paper helps people ask and answer appropriate questions:

  • Who is the audience?
  • What is important to individuals and groups? What does someone need to accomplish?
  • How do you choose to measure it?
  • How can you achieve alignment and support, and thus achieve your goals?

Metrics help to drive good strategy that secures competitive advantage. In his study of innovative companies, Robert Tucker (Tucker 2002, p51) notes, “Executives agree that the way you measure innovation progress (or lack thereof) determines the type of innovation you get, and the degree of magnitude as well. Metrics tend to determine whether you company's focus will be on incremental innovation or on breakthroughs.”

Tom Kendrick of Hewlett Packard provides a thought which provides some validation for the Exhibit 3 model, “Effective metrics have these three characteristics: they are part of something bigger (focus), support the kind of behavior you are trying to foster (strategy), and facilitate good business decisions making, helping you get answers to critical questions (inquiry)” (Management Roundtable, 2001, comments in parenthesis added).


Adams, J (1974) Conceptual Blockbusting Stanford, CA: Stanford Alumni Association

Eccles, R, Nohria, N., & Berkekey, J. (1992) Beyond the Hype: Rediscovering the Essence of Management, Boston: Harvard Business School

Githens G (2003a, January) Relevant Metrics Improve NPD Performance Visions Magazine 27(1), p.26-27

Githens, G (2003b, October) “Benchmarks and Benchmarking: A Skeptical Look,” Visions Magazine27(4) October (scheduled for publication)

Leifer, R., McDermott, C., O’Connor, G., Peters, L., Rice M., & Veryzer, R. (2000) Radical Innovation: How Mature Companies Can Outsmart Upstarts, Boston: Harvard Business School Press

Management Roundtable (2001) Product Development Metrics Handbook: What Every Manager Needs To Know About Measuring Product Development, Waltham, MA: Management Roundtable

Tucker, R (2002) Driving Growth Through Innovation, San Francisco: Berrett Kohler Publishers

Wheelwright S., & Clark K., (1992) Revolutionizing Product Development: Quantum Leaps in Speed, Efficiency, and Quality, New York: The Free Press

This material has been reproduced with the permission of the copyright owner. Unauthorized reproduction of this material is strictly prohibited. For permission to reproduce this material, please contact PMI or any listed author.

Proceedings of PMI® Global Congress 2003 – North America
Baltimore, Maryland, USA • 20-23 September 2003



Related Content