Project Measurement

Steve Neuendorf

• Publisher: Berrett-Koehler Publishers
• Year: 2001
• ISBN:
  9781567262209
• Edition: 1
• Topic: Negocios y empresa
• Subtopic: Inversiones y valores

Learn how to create a measurement-oriented program for your unique organization. With new insights on measurement as a communications tool, this book offers step-by-step guidance on how to use specific quantitative actions-and-results assessments to ensure the success of key initiatives. You'll master the new skills needed to incorporate stakeholder requirements into the structure and operation of the measurement program. You'll also learn how to implement a measurement program and develop understanding of the measurement life cycle, including how to keep your tactics and processes evolving to stay in step with your organization and new challenges.

CHAPTER 1

Why Measure?

There are some excellent reasons for not measuring. No one can say that you missed your goal; it is just another opinion. Cost savings are substantial, since measurement can be costly. It may be that you feel you are the best at what you do and have everything under control. Measurement would only confirm that, so why bother?

The truth is that everyone measures; everyone has some sort of quantification for what they do. In business, measurement is forced to be somewhat formal because of accounting requirements. For public companies, certain other standards are imposed that prescribe measurements and measurement discipline and formality. Throughout the organization, everyone has at least an informal mental picture of the quantitative nature of what is necessary to do his or her job and manage risks. A detailed report is unnecessary for the material people to know how many pallets are on the dock or for the production people to know the size of various queues or the inventory of key supplies and materials. In fact, it is an accepted generalization that during a relatively steady state, even the most comprehensive measurement program has little effect on current or future events.

However, the opportunity to effect change or the ability to respond efficiently to change is lost if there is little or no measurement formality or discipline. If a key person leaves, the successor must start from scratch to figure out what his or her predecessor knew and needed to know to perform that job. If there is a significant change in a product or the demand for a product, the adjustments needed are made by trial and error, which can be costly. Usually, any ad-hoc need for information is met with inappropriately high cost in gathering the information and a high risk that the information will be inaccurate or unverifiable.

Thus, chances are that even if an individual is convinced that he or she is not using measurement, the organization likely is measuring, quantifying, and supporting decisions in an informal manner. The change is not in going from not measuring to measuring, but in going from a lack of formality to more formality and from a lack of communication to a much higher level of communication regarding quantitative information.

SOME MYTHS AND FACTS ABOUT MEASUREMENT

When someone does not want to do something, it does not really matter why it isn’t done as much as it matters that it isn’t done. Since myths usually contradict or exaggerate facts, perpetuating a generally accepted myth can often be an effective means to prevent measurement from moving forward.

Myth: Measurement Is Expensive

Fact: Measurement is a cost. In a business, the concepts of expensive and cheap do not apply. Outlays have a cost—the amount of cash or other resources committed. Benefits are always associated with costs. Cost-benefit analyses are used to support decisions about which outlays should be made. For example, successful software engineering process measurement programs require an outlay of 1 percent to 3 percent of the total cost of software engineering. Many industry leaders invest more than 5 percent of total costs in measurement. Typical benefits realized by these companies as an indirect result are continuously increased product and service quality, as well as annual improvements in productivity of 10 percent to 15 percent.

Myth: Measurements Can Be Used against the Persons Being Measured

Fact: The myth is true, in the same sense that matches can be used to incinerate civilization. It is possible to say that they can be, but only once. After that, what is done is no longer measurement, but is more akin to creative writing and risk management. Dr. W. Edwards Deming observed that the system dictates its own performance; the performance of people working in the system is largely a product of the system. In a system fraught with fear, one compensates by producing measures that mitigate the results of what the people in the system fear. One of Deming’s famous 14 points is to drive out fear.¹ Measurements are used correctly to assess the performance of the system and to identify system improvements.

Myth: Measuring Can Improve Processes

Fact: Processes can be improved only by changing the processes. Measurement helps identify beneficial changes that must be implemented for improvements to occur. However, processes can be changed independent of measurement. The advantage of measurement is leveraging the change investment. Good measurement identifies those changes with the highest return. For example, in software engineering, if processes are routinely changed by investing in all the newest tools and technologies, an annual improvement of about 3 percent in productivity can be realized. Most of what is acquired ends up as “shelfware” and considerable effort is spent trying out these new tools that are never used. By using measurements to identify the best improvements and focusing training on implementing those improvements, the same investment can consistently yield 10 percent to 15 percent productivity improvements.

DEMONSTRATING LEADERSHIP AND ARTICULATING STRATEGY

Another opportunity for a comprehensive corporate measurement effort is to demonstrate leadership. It is widely accepted that good leaders are those who inspire a clear vision of where the organization is headed. If the vision can be expressed quantitatively, where you are in relation to the vision and your progress toward the vision can also be expressed quantitatively. With direct feedback concerning progress, it is much easier to continue on a successful course or to change to an alternate course.

ENABLING MANAGEMENT

Empowered is one of those overused and abused words, but it is hard to talk about measurement and management without embracing the notion of empowerment. By empowering someone, power that was previously at a higher organizational level is now reassigned and redefines a person’s role, giving him or her the responsibility for making a decision, authority to make that decision, and accountability for the results of that decision. In truth, most people are uncomfortable making decisions, and sociologists have noted that this discomfort is really at the root of our hierarchical social and economic system.

Measurement and quantitative information are essential parts of decision support systems (DSS). They are intended to increase the decision maker’s confidence level by providing a better understanding of alternative options for a decision and for forecasting the consequences of any particular decision option.

Another consequence of empowering those at lower levels of an organization is that they will be within the larger population of the organization and will naturally have more communication channels to deal with. Decision makers at lower levels need to provide more people with certain information and, in turn, receive information from more people than was required when the decision was made at a higher level of the organization. Less of this information is summarized (“the bottom line”) than is customary when conveying information to higher organizational levels.

EFFECTING PROCESS PROFICIENCY AND IMPROVEMENT

As noted in the Preface, the maturing process of going from unconsciousness to consciousness implies that measurement is not a goal. Rather, it is a means of achieving the goal of competence, however that might be expressed. This relationship is also important to consider when the organization is looking to cut costs. If cutting costs merely meant spending less money, it would not matter what was cut. However, what it really means is that costs need to be cut by something less than the impact of those cost cuts on revenues.

A good analogy would be to liken cost cutting to a surgical operation. If the goal is to cut out waste and the first thing identified as waste is the lighting in the operating room, there cannot be much hope for successful surgery. In fact, not only is the patient (process) in grave danger because the operation will be performed in darkness, but there is also a risk to the surgical team (unintended staff loss). While this analogy sounds ridiculous, it is the equivalent of organizations embarking on cost cutting by eliminating or substantially reducing measurement and other activities aimed at improving processes, which almost always results in faster, better, and cheaper performance.

ALIGNING THE FACETS

Serious improvement involves much more than improving processes. Organizations are multifaceted organisms (as the name implies) and many factors must come together for success. Similar to the marathon runner who is in excellent condition except for a foot blister, although everything else is perfect, that one weak area may result in an unsuccessful outcome. Organizational facets, too, can be clearly identified. For example, consider an organization with the major facets of leadership, strategy, project management, the management of project management, process, and environment. It is easy to see how, for example, the best processes will not excel if project management is poor, or how a poor strategy will not succeed even if all other facets are at their best. Usually, these relationships are the overlooked correct answers to questions arising from the lack of results after significant investments in one or more facets. A fortune can be spent on process; however, if the environment in which that process is expected to perform is not aligned properly, the process will not perform as expected.

An effective measurement program identifies all necessary facets for success and works toward evaluating the alignment of these facets. This point is significant in establishing expectations for improvement activity, both in estimating the investment to achieve success and in understanding the results from such investment.

Measurement is important, but deceptively simple-looking; a misguided venture into measurement can easily create more problems than it solves. Misadventures in measurement often make subsequent endeavors that much more difficult to undertake. “We tried that here and it did not work.” A good illustration is in a classic W.C. Fields movie where he entices a “mark” into a game of cards. The soon-to-be-victim says, “A game of chance?” In his famous stage whisper, Fields says “Not the way I play it.” Examples abound for things that were “tried here and didn’t work.” Total quality management (TQM), benchmarking, business process reengineering (BPR), and numerous other approaches would fit this improvement opportunity characterization where the appropriate response is “not the way we played it.”

NOTE

1. W.E. Deming, Out of the Crisis (Cambridge, MA: MIT Center for Advanced Engineering Study, 1986).

CHAPTER 2

What Is Measurement?

Understanding measurement is the key to acceptance and cooperation for an effective outcome. Measurement has a structure, with each of its elements working together to form a system. Why some things are done and why others are not done is not always apparent to everyone. Effective use is encouraged, while misuse is obvious and discouraged. Most often, problems arise with measurement programs not because measurement is not beneficial, but because abuse causes harm to the organization and the people who make up the organization.

WHAT ARE MEASURES?

Measures are analogous to the hierarchy of data:

Data are the basis of a measurement program. In general, measures have some:

• Quantitative elements (a number—e.g., 42, 127)
• Qualitative elements (e.g., high, heavier, warmer)
• Unit-defined elements (e.g., degrees Fahrenheit, inches, pounds)
• Unit-less elements (e.g., a weighting factor).

To be rigorous, consideration should be directed toward several properties of measures in evolving effective measurement. The key ones are precision, accuracy, and validity:

• Precision is the degree to which a measure exactly expresses the property being measured; e.g., inches are a more precise measure of length than are feet.
• Accuracy is the property of a measure that describes the extent to which the measure value represents the property value being measured. The measurement method often determines accuracy. “Stepping-off” a distance is generally less accurate than using a measuring tape; however, the way the measure is taken also affects its accuracy.
• Validity is an eval...