Selecting Performance Measures

Many organizations are struggling to implement useful metrics. The common lament is that measurement is complicated and time-consuming, people cheat when it comes to reporting performance results, and often, the most valuable metrics require special projects to be undertaken to enable data collection. Moreover, when one considers the effort involved in collecting data, analyzing performance, creating reports for management and finally holding review meetings to discuss the results, the cost of maintaining a measure can easily outweigh the benefit of whatever insights it yields. Fortunately, this problem can be largely avoided by linking measurement to strategy. This subtle shift in focus can make the valuable difference between knowing whether you are doing things right versus doing the right things.

For example, when a large manufacturing company realized that it needed to improve its success rate rolling out new products to the marketplace, it became apparent that a significant behavioral change would be required by the sales, marketing and engineering divisions - each of which would have to work with the other two to make sure that customers would be willing to pay for the new features being added. In essence, a new strategic objective, “incorporate the voice of the customer into new product development (NPD),” was necessary before any performance measure could be selected. Ultimately, the metric chosen by the NPD process.” This metric forced the organization not only to be more disciplined about its NPD process, but also to force cooperation among key stakeholders in that process. Before long, the company noticed that its new products were being accepted by customers with fewer complaints, warranty claims decreased and market share increased as the company became proficient at matching product releases to customer preferences.

This example illustrates that the most valuable metrics start with a clear understanding of the strategy, focus on a change in behavior to address the key issue preventing achievement of the strategy and encourage cross-functional cooperation. Moreover, establishing good performance measures must start at the top of the organization, because it is unlikely that any of the divisions in the previous example could have instigated a fundamental improvement in the NPD process without the support of higher-level executives promoting the necessary interaction across functional lines.

While the process of selecting performance measures often involves more art than science, there is a pattern to the approach that many organizations have applied successfully.

Step 1: Identify Your Strategic Objectives

Many organizations have found the strategy map component of the balanced scorecard framework an effective tool for articulating strategy and organizing objectives into a logical cause-and-effect format. Most maps consist of 15 to 20 objectives, each of which describe a desired strategic outcome or a driver of those outcomes and, therefore, set the context for measurement.

Step 2: Define Your Strategic Objectives

Figure 1 shows two examples of an objective description for an IT organization that address these questions. Some organizations have found it helpful to brainstorm a list of critical success factors (CSFs) as a way of surfacing the ideas behind their objectives. For example, an objective such as “maintain a safe work environment” may consist of several CSFs, such as educate the workforce on the proper use of safety equipment, provide appropriate protective equipment to employees and monitor incident reports to identify potential hazards that could cause a serious accident.

In either case, whether you write detailed sentences or bullet-point CSFs, the process forces you to think of the “value drivers” behind a given objective.

Step 3: Brainstorm Potential Measures

Step 4: Select Measures

Ultimately, choosing which metric is best requires a set of guiding questions or evaluation criteria to consider when discussing the merits of each potential measure. The following questions can be useful during this streamlining process:

• Is the measure quantifiable and repeatable?
• Will the measure be useful for decision-making?
• Can reliable and accurate performance data be gathered frequently?
• Does the measure focus on the issue preventing us from achieving the objective?
• Will the measure drive desired changes in behavior?

Some organizations find that many of the driver measures listed are useful for assessing performance, and decide to combine them into an index. In this case, they track them as one summary analytic. Typically this involves converting data values into a common scale and weighting each measure according to its significance as a component of the analytic.¹

Selecting good performance measures is not as difficult as you may think. By following these guidelines, you can increase your focus on a manageable and high-impact set of performance metrics that will quickly become integral to your organization’s success.

Reference:

1. Mark Graham Brown. Beyond the Balanced Scorecard: Improving Business Intelligence with Analytics. University Park, IL: Productivity Press, 2007.