Performance Measurement and Metrics

Measurement is important. Our decisions are based on facts and data and decisions have consequences – for managers and the business.

Managers know this, yet for all its importance, organizations often do a poor job of measuring performance. The wrong things are measured, data is misinterpreted or misrepresented. At times performance measurement is simply misused and abused.

Performance measurement  has at most, two legitimate purposes:

• Provide evidence and improve understanding as to how, or how well, systems and/or processes are functioning.
• Provide the type of information required for people to take action and improve performance.

Everything else is window dressing or exercises in evidence corruption – including those performance measurement systems designed to allocate rewards and punishment. You can have measurement that drives performance improvement across the business or you can have measurement that rewards and punishes at the expense of the business – but you can’t have both! (see Deming)

Converge Performance Measurement

Converge builds performance measurement systems designed to yield evidence and understanding while supporting action that improves performance.

A good performance measurement system must do more than just look good (although looking good helps). It must accurately and reliably convey performance information so that people can take effective action.

That’s what Converge builds. Performance measurement that looks good in the executive suite and drives improvement across the business.

This requires more than simply defining what to measure or producing colorful graphics. It requires knowledge of data gathering and analytical techniques as well as the proper methods of interpretation and presentation of results.

Characteristics of Good Performance Measurement Systems

Performance measurement systems that look good and drive improvement have all of the following characteristics.

• Measure system and process performance. System and process information is information that can be used as a basis of action to improve performance. Everything else is at best, waste. The value of information is in the actions it promotes.
• Balance  outcome, output and in-process measures. Exclusive focus on results or outcomes is an exercise in management through wish fulfillment. To be useful, information must show causality, mechanisms, provide explanation and reveal systemic structure.
• Focus on the critical few as opposed to the trivial many. There are far too many causes affectioning performance for us to measure all of them. We need to focus on what is important, the critical few causes we can manage or manage around.
• Are explicitly linked to the purposes and objectives of the organization. Performance measurement needs to link back to the aims, purpose and objectives of the business. Everything else is nice to know information that will be safely ignored.  
• Analyze and present data in context. Performance measurement always has two dimensions: proximity to target and variation. Without both, you don’t have performance measurement. This requires the use of proper data analysis and presentation tools such as the control chart (pictured right).
• Appreciates the difference between enumerative and analytic aims. Enumerative data is used to describe or inspect conditions. Analytic data is used to explain or predict performance. Almost all performance measurement requires analytic data, most performance measurement systems produce enumerative data. The mismatch is fatal to sound, evidenced-based decision making.
• Guard the integrity of the data against corruption in presentation. Sources of evidence corruption in presentations are: effects without causes, cherry-picking, overreaching, chart junk, junk charts and the rage to conclude (see Beautiful Evidence by Edward Tufte).