In the previous article (see How Many KPIs Should My Contract Have? – Part 1) we examined the benefits and costs of using performance measures. In this article we are going to examine how to balance the benefits versus the costs of using them .
On method is to use the science of human factors engineering to determine the optimal number of performance measures. This field examines the cognitive workload on humans describing the optimal amount of activities, including complexity of each activity, allowing humans work at their best. For example, in the defence and aerospace sector, much of this research and the associated “standards” drive how the various instruments and controls are positioned in aircraft, helicopters and spacecraft.
The famous 1956 paper titled, The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information, by George A. Miller observed that through a variety of studies our short-term memory had a capacity of about “seven plus-or-minus two” chunks. These chunks represented a concept or thought as opposed to single value. However, other research found that this study was sometimes overly optimistic and perhaps the real value was a low as four.
Accordingly, it is possible to apply this thinking to our problem of how many performance measures we should use.
Firstly, if it makes sense for your contract, I recommend you divide (or chuck) the performance measures into different tiers reflecting their utility, potentially using a 3 tiered approach based on Strategic Performance Measures (SPMs) / Enterprise Performance Measures (EPMs), Key Performance Indicators (KPIs) and System Health Indicators (SHIs). You can learn more about each of these tiers, including their role and responsibility, from previous articles (see When is a KPI not a KPI?).
Secondly, limit the number of performance measures for each of the tiers, especially the SPMs and KPIs which will get the most attention, to between 3 and 5, or perhaps up to 7. Where the performance measures are simple, intuitive and easily understood, including objective, quantitative or common, industry standard performance measures, it is possible to use the higher band (e.g. between 5 – 7). However, where the performance measures for more complex and harder to understand performance measures including subjective, qualitative or bespoke performance measures, the lower band is recommended (e.g. between 3 and 5).
Finally, in setting the total number of performance measures avoid combining multiple performance measures into a single measure in an attempt to minimize your total number of performance measures. For example, KPI1 is made up of KPI1.1 and KPI 1.2 and KPI 1.3, etc.
So when considering how many performance measures you should have in your contract keep in mind the words of the American actress Julie Newmar:
“More is not necessarily better. Better is better.”
It is better to deliberately and consciously consider what is the optimal number of performance measures for your contract, including how they aggregated rather than simply applying a “magic number”. But which performance should you choose? That is a question for another time.
Great next part. Good to see Millers being used. As time is usually not critical higher numbers is theoretically practical but maybe not useful to senior managers. Prioritising the top 5 in your groupings along Trickers line as used by AICD is handy and can be useful especially in comparing activities in an an enterprise!