Myths of KPIs
Recently I have been reminded once again of how KPIs are misused and abused. Just like six centuries ago, when many thought the world was flat, mankind was blind to the realities that are there to seen on closer observation. We are blindly applying old thinking on how we measure, monitor and improve performance.
I am going to post a series about the myths of performance measures and KPIs. I would be interested in your feedback.
Myth: Most Measures Lead To Better Performance
Every performance measure can have a dark side — a negative consequence or unintended action that leads to inferior performance. Well over half the measures in an organisation may be encouraging unintended negative behaviour. It is imperative that before a measure is used, the measure is:
- Discussed with the relevant staff: “If we measure this, what will you do?”
- Piloted before it is rolled out.
- Abandoned if its dark side creates too much adverse performance. I expand on this dark side in the section on Unintended Consequences.
As Dean Spitzer says “People will do what management inspects, not necessarily what management expects.” How performance measures can go wrong can be illustrated by this example:
A city train service that had an on-time measure with some draconian penalties targeted at the train drivers. The drivers who were behind schedule learned simply to stop at the top end of each station, triggering the green light at the other end of the platform, and then continue the journey without the delay of letting passengers on or off. After a few stations, a driver was back on time, but the customers, both on the train and on the platform, were not so happy.
Management needed to realise that late trains are not caused by train drivers, just as late planes are not caused by pilots.
Lesson: Management should have been focusing on the controllable events that led to late trains, such as the timeliness of investigating signal faults reported by drivers or preventive maintenance on critical equipment that is running behind schedule.
I will continue to post other myths of performance measures in the following weeks. If you would like to keep up with my current research and speaking tours follow me on twitter @davidparmenter.
For more course details, articles and also access to my more in-depth blog posts visit www.davidparmenter.com.
Helping companies overcome 💻 web dev and 📈 marketing challenges
7yDavid, nicely written! I like an approach from lean six sigma. If there is a mistake - train coming late - search for a process that failed and don't concentrate on the issue itself. Issue is simple - late train. But the process behind it is difficult - including other traffic on the rails, number of people getting on/off, weather, etc etc. The danger of KPI is it moves are attention to the results away from the processes. Then, we tend to fix the most obvious or the latest part of the processes, and we can cause many unintended consequences.
Member of the Board of Directors at PRACE (Partnership for Advanced Computing in Europe)
7yAn other classical exemple is H index for academics, with the urge of publish or perish, whatever the quality being the study, as long as this is sexy enough to "sale". This explain also why so few (so useful) paper are published on what went wrong: this will not grow your H index !
Policy Lead, Trust & Safety at Google
8yAbsolutely
Leading the Upper North Island Sales Teams (Strategic, Aquisition and Expansion)
8yHi David. I always find your articles thought provoking. Over the years I have watched many businesses struggle particularly with how best to measure and reward utilisation and productivity in order to encourage the desired behaviour from their teams. So the topic really resonated with me. I am looking forward to reading your future articles in this series.