The belief that “if you cannot measure it, you cannot manage it", popularized in such books as the Balanced Scorecard, is wrong.
A central decision of managers is to strike the right balance between the measurable and the unmeasurable.
We need measurement, but more importantly, we need good measurement.
The over-reliance on measurement often instills a false sense of confidence in the knowledge of performance.
by Phil Green, president, Greenbridge Management Inc.
The belief that “if you cannot measure it, you cannot manage it,” popularized in such books as the Balanced Scorecard, is wrong. It has misinformed and led many people astray.
This belief has spawned measurements and performance indicators for every facet of businesses with little or no regard for the meaning of the numbers or their trustworthiness, or even whether they are necessary.
People manage most aspects of their lives without recourse to measurement, whether it is their child’s struggles at school, or their relationship with their spouses or their boss. Yet measurement is one of the hallmarks of our advanced civilization.
Peter Drucker wrote that a central decision of managers is to strike the right balance between the measurable and the unmeasurable. It is a mistake to rely too much on quantification and overlook vital, but unmeasurable, information.
Some of the things that have the biggest effect on businesses—such as the sudden launch of better competing products—are not measurable until it is too late to act. One of the most important skills of the successful manager is to judge character and inspire action, tasks psychologists have attempted to quantify but whose efforts pale beside intuition and the wisdom of experience.
The over-reliance on measurement and on information technology to generate it produces realms of data that often instill a false sense of confidence in the knowledge of performance. Concreteness and profound knowledge about the business can only come if there is profound knowledge about the measurement process itself, and the reliability of the resulting information.
We need measurement, but more importantly we need good measurement. We measure things everyday without even thinking about it, relying on the measurements implicitly.
To inform rather than misinform, management must ensure the trustworthiness of measurements. This means understanding the assumptions behind them, and asking “which measurements can be trusted, and which should be shun?”
One has only to read the daily financial papers to realize that many executives are struggling with measuring the most basic measure of performance, their company’s income. Headlines scream out “restatement of earnings” on an almost daily basis.
The Washington Post recently reported that there were 270 restatements of earnings before the Sarbanes-Oxley Act was enacted in 2001, and 1,200 in 2005.
Managers face the same issues with more mundane measurements.
How reliable are the inventory data in a supply chain, when different nodes in the chain have varying methods for defining, counting, estimating and tracking it? At one factory, the key measures of process performance were subject to various certifications.
They were so variable, however, that they accounted for most of the variation observed in the process. The use of these numbers to manage the process lowered productivity.
A CEO told me recently he was concerned that his company’s environmental compliance performance indicator did not highlight the most serious concerns. I have rarely not seen safety statistics that are misreported or misinterpreted.
Fortunately improving the reliability of measurements usually produces high returns.
Fortunately improving the reliability of measurements usually produces high returns. By helping clients improve their measurement processes I have helped them achieve such stunning gains as:
Million dollar increases in manufacturing productivity when my client started measuring the right things reliably
A 50% reduction in inventory when my client started measuring the right parts of their distribution process