Why improving the average sometimes does not improve anything

How many times were you already presented with a survey that explained an improvement through an average figure that has gone up? Such a change usually sounds positive doesn’t it?


When reading survey results many corporations immediately look at the improvements and aggravations section. They want to know where they got better and where they might have fallen behind. And often little attention is given to those average figures that have not changed (thinking everything remains as it already was). But there is a flaw.

Obviously, the average figure is the result of those numbers below and above that average. But we don’t know if it was triggered by similar numbers or opposite extremes. And this is often overlooked unfortunately.


Imagine you have two cooking plates. One is very cold (-20°C) and the other one is hot (+60°C) . Assume now you put your left hand onto the cold plate and the right hand onto the hot one. On average you should feel ok. But in reality it would feel very uncomfortable. We could increase the temperature of the cold plate to -10°C for example. On average you should feel even better now. But only once you bring both extremes to a normal temperature (in this case let’s say +20°C) you have really improved the situation for your hands. And the average temperature would have even not changed compared to the initial situation.


The example above might sound funny and a bit far-fetched. But it illustrates how an average number could be misleading. And this is true whether the average has increased, decreased or remained unchanged.


Next time somebody in your company presents a survey result think “cooking plates and hands”. And while it might make you smile, it could help bring up those questions which reach beyond the executive summary. Best case they lead to the real pain points of your focus group. And who would not want to better know where to really improve for our people?


Chris Frey @chrisfrey.com