Why Measure Performance?

Leading organizations such as Sears, Analog Devices, FMC, and Brown and Root1 are discovering that there are two key reasons to invest in performance measurement programs:

1. Strategic Alignment. Performance measurement programs can be used clearly to communicate and reinforce messages about what is important in the organization, and to engage people in aligning to objectives.
2. Strategic Learning. Performance measurement programs can promote learning as managers sift through performance data and discover pockets of excellence or weakness where more information about what works and what doesn’t can be found and reapplied. Learning also occurs as performance data is analyzed to discover trends and relationships between measures (e.g. How does employee satisfaction affect customer satisfaction?)

Preliminary studies are demonstrating that organizations with effective performance measurement systems are more than twice as successful as similar firms2.

Why Read This Paper?

Performance measurement has historically made a less-than-spectacular contribution to the practice of management. We recall the "man as machine" era with its time and motion studies and the failure to recognize the potential of individual initiative and creativity. We have seen sub-optimization in manufacturing caused by a focus on efficiency metrics for one machine, but resulting in increased inventory levels and reduced production overall. Cost accounting, a traditional performance measurement method, has driven poor business decisions by allocating an overhead burden of as much as 1000% to each direct labour hour. In summary, we have seen that performance measurement programs can be equally effective in promoting undesirable behaviour as desirable behaviour.

Despite past failures, we argue that performance measurement remains the key tool which senior leaders can use to align their organizations and to create strategic learning. Furthermore, we argue that these two results are integral to achieving high performance in your organization. In this paper, we will show how effective performance measurement creates these results.

Implementation of an effective performance measurement program is almost always difficult, precisely because it is so effective at creating the kind of change that organizations typically resist. We describe a case study of a performance measurement system implementation, and later identify organizational locks that block implementation and suggest strategies for unlocking these blocks.

Technology plays a special role in performance measurement, both as an organizational lock, and as an immensely powerful enabler. In this paper we will describe the attributes of an information system which can be effectively used to create alignment and strategic learning in your organization.

Finally, we provide a self-assessment tool that you can use to judge the effectiveness of the performance measurement program in your organization. We hope that the ideas presented in this document will contribute to higher performance in your organization. While a myriad of actions, behaviours, skills and programs are required to unleash the productive potential of each individual in your organization, we believe that effective performance measurement is the thread that can tie all of these activities together into a dynamic and effective strategy.

Strategic Alignment and Learning

Strategic Alignment
"What gets measured gets done."

All of strategy, from vision, to mission, to objectives and goals can be condensed to this one sentence. This is the power and the risk of performance measurement. If you establish a measure, and faithfully attend to it, communicating widely your expectations and results, your organization will begin to align to it. If the wrong measure is selected, your organization will align in the wrong direction. If an accurately selected set of measures that reflect a balanced view of overall performance is implemented, people will align with the strategic intent of the organization. (The concept of a balanced measurement system originated with Kaplan and Norton and the Balanced Scorecard)

To understand the process of alignment, place yourself in the shoes of Mary, a first level manager. As senior leadership parades out a new performance measure - say customer satisfaction - Mary thinks, "Hey that’s a great idea. Customer satisfaction is important. Too bad it will never happen around here". Many programs to achieve strategic alignment end right here, and alignment does not occur.

About a month later, Mary is at a division meeting where her director reports the customer satisfaction results for the division and compares them with results from other divisions. Mary thinks, "Wow, this is great. We do care about customer satisfaction. I sure hope George over there in client service gets his people in line so that our numbers improve." Here we have achieved the first level of alignment. Through regular reinforcement of a measure, Mary has received and internalized the message that this is an important objective for the organization.

The second level of alignment occurs when Mary discovers that her actions can contribute to customer satisfaction - not just the actions of her colleagues. Mary might find that calls to customer service can be reduced if the documents sent out by her staff are written in plain language. At this second level of alignment, Mary is actually acting in alignment with the strategic intent communicated by the performance measure.

Now consider a third scenario, about six months later. Mary is at a regular one-to-one meeting with her director. The director says, "Our organization’s customer satisfaction ratings are at 4.5 and our senior leadership have set an objective to achieve a 6 overall by the end of the year. I have committed our division to reducing turnaround time on customer correspondence from 45 days to 10 days as our contribution to that objective. I would like you to work with your staff to identify the actions you will commit to in order to help us achieve a turnaround time of 10 days."

When Mary commits to specific actions and targets which contribute to overall organizational performance measures, we have the third level of strategic alignment. Mary has received the message that the measure is important. She has begun identifying actions she can take to contribute to improving results on this measure. Now she has engaged her staff in setting and committing to specific actions and targets that will help ensure that the organization achieves the desired results.

Strategic Learning

Some performance measurement programs fail because the wrong measures are chosen. Others fail because the correct measures are chosen, but the environment changes and the measures do not. Still others fail because senior leaders simply become bored with the measures, and stop paying attention to and reinforcing them. In all of these cases, the core value of performance measurement has been overlooked:

Performance Data + Insight = Strategic Learning

Recall Mary, our first-level manager at a division meeting where customer satisfaction scores from across the organization are being shared. As she reviews the data, she notes that another division has achieved better results than her own division. After the meeting she calls a friend in that division and asks how those scores were achieved. During the conversation, Mary learns that the high performing division has implemented a toll-free interactive voice response system where clients can call to get information about their files without having to wait for an available client service representative. Clients are responding positively to this approach, and their overall satisfaction has improved as a result. Mary recognizes that a similar approach could work with her clients and begins to research and sell the idea in her division. She has achieved "differential learning" - the first level of strategic learning. The performance measurement system has allowed her to tunnel through the organization and uncover knowledge from another division that can be reapplied in her own.

A second level of strategic learning can be achieved when an organization is monitoring a balanced set of objectives. Suppose that Mary’s organization is monitoring both employee satisfaction, and customer satisfaction. An analyst performs causal pathway analysis on the data, and discovers that there is a cause and effect relationship. That is, when employee satisfaction improves, customer satisfaction improves. What once may have been a "soft" intuitive notion, is now empirically validated and provides justification for investment in programs to achieve employee satisfaction.

At the second level of strategic learning, organizations are evolving their performance measurement systems, and thus their strategies as they uncover cause and effect in their own organization and discover the levers required to achieve their desired results. However, learning at this level can still lead to failure, if senior leaders are not learning about the interaction between their organization and the external environment.

Suppose Mary’s division is a government agency processing emergency shelter requests for subsidized housing. As the organization achieves alignment and strategic learning internally, it improves its customer satisfaction scores dramatically, and becomes a model agency promoted across the public service as an example of a high performance organization. However, a number of disturbing trends are occurring in the external environment. The number of requests for subsidized housing is increasing. The incidence of violent crime is increasing, and unemployment is up. Analysis may show that improved customer satisfaction actually increases demand for emergency shelter.

At this stage, many organizations ignore the external environment and await direction from on high as things continue to get worse. Others may actually attempt to reduce the emphasis on customer satisfaction in hopes of reducing demand for subsidized housing. If anyone is still analyzing the data, they may find that this strategy is also correlated with undesirable results in the external environment. Paradoxically, both increasing and decreasing customer satisfaction may create negative results externally. The organization is locked into a self-reinforcing cycle with the environment in which it operates.

Here is the opportunity for strategic breakthroughs - the third level of strategic learning. (Peter Senge refers to this as double-loop learning 9.) As senior leaders confront the paradox of the self-reinforcing cycle, and investigate the interaction of the organization with its environment, they are seeking to discover what the organization is missing. Using focus groups, scientific research, analysis of existing data and other learning tools, new strategies may emerge. Perhaps what is required is complete replacement of the existing strategy - all the existing measures need to be changed or refocused. Or perhaps a new complementary strategy may emerge either within the organization or in partnership with another.

Suppose that life skills or addictions emerged as root causes of the negative results in the environment. Mary’s organization begins working in partnership with other social service agencies to combine requests for emergency shelter with applications for life skills training or addiction counseling programs. The performance measurement system is revised to incorporate these additional outcome measures, and as the organization aligns with the new measures, new learning occurs, leading to new breakthroughs, new failures and new possibilities.

Performance Continuum

Strategic alignment and learning do not operate in isolation. As organizations progress to higher levels on these two axis, they begin to exhibit characteristics of high performance organizations. As higher levels of learning are achieved, the organization becomes more responsive and agile. As higher levels of alignment are achieved, the organization becomes more focused and engaged. Together, high levels of alignment and learning contribute to the empowerment of individuals in a cohesive learning community.

Performance Measurement Implementation Case Study

We have seen the effect of a performance measurement program on Mary, a manager in a government agency dealing with emergency shelter requests. Carol is the senior manager in this organization who was responsible for implementing the performance measurement program. Carol’s experience illustrates some of the difficulties associated with implementing an effective performance measurement program.

Situation Assessment

As a senior manager for several years, Carol had always been frustrated by the lack of information she had about what was happening in the organization. In her own division, she had implemented various performance measures over the years, and soon came to be appreciated by her superiors for the fast and accurate information she could provide about operations in her division. This new assignment, however, was a bit more challenging. Carol had been asked to develop a performance measurement program for the entire organization.

In her own division, it had been a relatively simple matter to convince the five managers who reported directly to her of the value of performance data. After only a few months, her team had developed the skills to collect, analyze and communicate performance data in a way that created alignment and learning in her organization. Across the organization, however, there were more than a hundred managers who had little or no experience with effective performance measurement programs. In fact, many of these people had previous experience with ineffective performance measurement programs. She suspected that her new assignment was viewed negatively in many parts of the organization.

Carol knew that the perception of performance measurement was a reflection of problems with existing measures. To gain the trust of managers, she had to develop a very clear picture about what was going wrong and acknowledge these problems. She had to ensure that managers knew that she understood their concerns and would address them in the development of performance measurement program for the organization.

To accomplish this, Carol conducted a situation assessment. The assessment consisted of interviews with numerous managers at all levels throughout the organization, analysis of existing reports and performance measures, and analysis of existing strategy documents. Her findings were as follows:

• performance data was not trusted for accuracy
• performance data was not available on a timely basis
• performance data was held secret from managers
• performance measures focused on financial issues and activity counts only and did not accurately reflect the work of departments
• performance measures were not integrated nor analyzed properly
• existing strategic plans and statements did not outline specific or measurable objectives

Carol presented her findings to management teams across the organization, using specific examples and stories where possible. By relating stories of inaccurate data, misuse of data and irrelevant data, she identified with manager’s personal feelings of distrust and let them know that she would strive to avoid these difficulties in the new program. Carol followed up with a memo to all managers summarizing her findings and reinforcing the intent to avoid these problems in the new performance measurement program.

Selection of Measures

The selection of appropriate measures for a performance measurement program is the key to its success. Carol knew this, and also understood that the exercise of selecting measures would engage managers in a very difficult and contentious process. From her own experience, and her discussions with the president, Carol had developed a fairly good idea of what the measures would be, but she also understood the importance of checking her assumptions and involving others in the process.

Carol established a cross-functional team representing a diagonal slice through the organization. The team included people at all levels from vice-presidents through to administrators, social workers and engineers. Although Carol limited the team to 12 people, she ensured that it was representative of all divisions and all management levels.

The agenda for the team’s first two-day meeting included a discussion of guidelines for selecting performance measures. In this session, participants had the chance to ensure that their concerns and reservations were noted and would be considered in the selection of measures, along with the concerns expressed in the situation assessment.

Prior to the meeting, Carol had enlisted analysts from the Information Systems Department and had commissioned a two week study to develop process models of all major business processes in the organization. These analysts presented their findings at the kick-off meeting giving team participants a process-oriented view of the organization to complement the functional view they were accustomed to.

The remainder of the two-day meeting was spent going over each function and each process in the organization, brainstorming measures that would give a balanced view of its overall performance. The brainstorming process for each function and process identified measures from each of four perspectives: financial, client, internal business process, and organizational development and learning. The use of these four perspectives helped ensure that a balanced view of the activity was being developed.

At the end of the team’s first meeting, all of the measures identified in the brainstorming exercise were documented. The team then took this list back to their departments and discussed the list as widely as possible to determine if anything had been missed, and to consider which measures were most appropriate.

The second meeting of the team, about two weeks later, was focused on identifying the "top ten" measures that would form the performance measurement program. Measures were selected based on the guidelines discussed in the initial meeting, on the availability of data, and with a strategic view as to where the organization needed to go in terms of performance.

Carol summarized the results of this meeting and presented the measures to the senior leadership team. With one or two changes, the measures were adopted and implemented.

Data Collection

Collecting data for the selected measures turned out to be challenging. Some data was available from existing systems but was not trustworthy. Other data was collected on incompatible systems and still other data was not being collected at all.

For measures where the existing data was inaccurate, Carol worked with Information Systems and data entry staff to identify root causes. She found that inaccuracies were caused by certain fields not being filled in, by data entry errors, and by backlogs of data entry. To address these issues, Carol designed reports that would highlight inaccuracies and missing data, and instituted regular data-entry audits. Where the same data was available from two systems, she asked for reports showing discrepancies. She distributed these reports to managers on a regular basis, and encouraged them to correct the problems. Carol worked with Information Systems to develop a process for getting data from incompatible systems into appropriate reports and data formats that could be used in the performance measurement program.

Data that was not being collected included measures such as employee satisfaction, client satisfaction, and leadership effectiveness. For external data, Carol initiated surveys of clients and other external stakeholders that would be performed on a regular basis. For internal data, Carol worked with Information Systems to develop automated survey systems on the internal Intranet. These survey systems were flexible so that new surveys and survey questions could be set up at any time. The surveys could be automatically administered by email or a web site (if confidentiality was important), and results tabulated into the appropriate format.

Communication of Measures and Results

After the measures had been identified, Carol had sent out a memo to all staff describing the performance measurement program, the selected measures, and the implementation process. She also attended every major management meeting over the next month to discuss the program, identify concerns and answer questions.

As part of the implementation, Carol had worked with Information Systems to develop an internal web site where all staff would have the ability to review current and historical performance data down to the individual work group level. This data was available immediately upon collection. As each new set of numbers was collected, Carol briefed the senior leadership team, discussed the results, and worked with them to determine what further analysis was required. She also prepared a message for the president that would be distributed to all staff highlighting important trends and drawing attention to the measures that needed improvement. Carol also attended other management meetings on request, and provided detailed analysis of the data where appropriate.

Performance Measurement Review

After a few months, Carol initiated a brief review of the performance measurement program. She interviewed several managers, and conducted a quick email survey of all managers. Her analysis confirmed her suspicions. The measures were appropriate with one or two exceptions, and the process was going well. However, some areas of the organization were benefiting more from the program than others. Although everyone was getting the same data, not everyone was realizing the benefits of strategic alignment and learning. Carol realized that the issue was the way in which the data was being used.

Carol created a training program for managers introducing them to the concepts of continuous improvement, statistical process control, and appropriate use of performance data. The program included role plays and other group processes in which managers could experience appropriate communication of performance results, and case studies showing how strategic learning could come from appropriate analysis of performance data. This one-day program was made available to all managers as a result of the performance measurement review.

Other smaller issues arose during the performance measurement review. New problems with data collection and survey administration were identified, and Carol worked to respond to these problems. She realized that the effectiveness of the new performance measurement program was linked not only to the implementation process, but also to ongoing responsiveness to problems and changes in the organization. She committed to reviewing the effectiveness of the performance measurement program on a regular basis.

Organizational Locks - Keys to Unlock Them!

There are many reasons why performance measurement efforts fail, and most of these are embedded in the culture and recurring behaviours of managers. We call these organizational locks, because they prevent performance measurement efforts from opening the door to organizational transformation. Here are four organizational locks we have identified, along with the keys to unlock them.

Lock: Managers are too busy fighting fires to worry about performance measurement.

Key: We once worked with a manager who was busy from early in the morning until late at night running from one meeting to another, writing documents and communicating on the telephone. We asked what his top priority was, and how much time he was allocating to it. The answer was an information system that would enact the entire mandate of his department, and he spent less than an hour each week on it.

As Stephen Covey teaches10, it is essential to set aside time for planned activities if you ever wish to reduce the number of unplanned activities you are forced to participate in. If you have begun to understand the power of performance measurement, and see how it may eventually reduce the fires you fight each day, we hope that you will make it a priority and allocate just one hour each week to it. As you begin to focus on it, your subordinates will quickly come to view it as a priority as well.

Lock: Managers hold on to performance data and refuse to communicate it to subordinates.

Key: Collect performance data centrally, then expose it to all employees via intranet and email. With current technology there is no reason not to broadcast this information and reap the benefits of strategic alignment and learning instantaneously. However, if your managers are likely to engage in this lock, you have a larger problem with organizational culture.

Lock: Managers use performance data to blame individuals or pass the buck rather than to engage in strategic learning.

Key: Effective use of performance data does require a set of management skills that may be lacking in organizations that have not previously been exposed to statistical process control, total quality or other analytical approaches. Management training may be necessary either as a refresher or from the ground-up on your organization’s performance measurement program and the skills and ground rules for using the data.

Performance measurement has the potential to drastically alter the prevailing patterns of organizational communication and thus will typically be met with resistance. Some of this resistance is simply a matter of a lack of knowledge. Training on how to use statistics and performance measures can help. However, resistance can also be rooted in the feelings of fear, insecurity and cynicism experienced by individuals throughout the organization. These attitudes can only be influenced by a strong and vocal executive champion who consistently reinforces the purpose of the system and directs the attention of executives and managers away from unproductive and punitive behaviours.

Lock: Managers resist the development of performance measures on the grounds that their work is too specialized or unpredictable to be fairly measured by a statistic.

Key: This lock is both true and false. On the one hand, almost all work in an organization is too complex to be summarized in a single statistic. If it were not, we would probably not require human employees to perform it. On the other hand, there are aspects of almost every activity that can be measured and improved. We suggest two responses to this lock. First, we recommend the implementation of a monthly management letter to add qualitative data to the performance measurement system. Second, we recommend that a balanced set of measures be developed - not with the view of summarizing all that is done, but as a way of providing a useful aggregate approximation of what the organization is trying to achieve.

Performance Measurement Technology

There has been a great deal of emphasis on performance measurement technology over the years with various buzz words attached. Decision Support Systems (DSS) gave way to Executive Information Systems (EIS) and later Business Intelligence Systems (BIS) and new acronyms arise with unpleasant regularity. Still, few or none of the systems proposed or implemented to date address all of the requirements of an effective performance measurement program. Here are some requirements for a system to meet this challenge.

The system must:

• provide all staff with access to performance data in a format that is easy to learn and use (data must be available down to the work group level of granularity)
• provide various views of the data to support interactive learning
• support cascading of performance measures and targets down through the organization including wizards to assist managers in developing balanced measures and targets
• allow every manager to "sign-up" for specific actions to achieve targets and track completion
• support agenda setting for regular management meetings as insights occur in the strategic learning process throughout the organization
• provide advanced statistical analysis tools to establish correlation or cause and effect relationships between measures
• provide functionality for automated survey administration (e.g. employee satisfaction, 360 degree leadership reviews, etc.)
• allow data to be input manually or incorporated automatically from external sources

Performance Measurement Self-Assessment

We have discussed the benefits and outcomes of an effective performance measurement program, and briefly mentioned the potential risks of ineffective performance measurement. In this section, we present the attributes of an effective program in the form of a self-assessment that you can use to evaluate your own organization. To use the self-assessment, simply answer yes or no to each of the following questions. If you answer any of the questions with a no, you may want to consider developing an action plan to address this deficiency in your performance management program.

1. Does every person in your organization know what your performance measures are and see the actual results regularly?
2. Do your performance measures reflect a balanced view of organizational objectives?
3. Do your performance measures provide a fair and consistent view of contributions from everyone in your organization?
4. Is your performance data believed and trusted for accuracy?
5. Does your performance measurement program encourage each individual or team in your organization to share ownership for achieving improvements.
6. Have you established targets for each of your performance measures?
7. Has each manager in your organization "signed-up" for specific actions to achieve your performance targets?
8. Is your performance data being analyzed to discover internal correlation between measures?
9. Is your performance data being analyzed to discover correlation between internal performance measures and external environment indicators?
10. Are your performance measures regularly reviewed and modified to remove irrelevant measures and add measures for new initiatives?