The danger in the mean

Author: Stephen James Founder, SilverJet Insight

Heard about the man with his head in the oven and his feet in the freezer? On average, he was comfortable.

I hope you enjoyed that one, it is a light hearted introduction into the dangers of making assumptions based on averages.

You aren’t average, are you?

Are you a male around 5ft 8, 5ft 10 or a female around 5ft 2 1/2 to 5ft 4?

If so, you are in the ballpark of average, height wise that is.

If not, I bet most of you aren’t too far off.

The point to this, height is just one aspect, you aren’t an average person just based on your height, you are a multifaceted being that excels in some areas (and the opposite in others). Being treated as average in all aspects of life leaves you feeling as though you are misunderstood.

Apply that sort of thinking to your customer base. Are you misunderstanding them?

Without intending to directly to quote Mr McVeigh (my old maths teacher), there are 3 types of average that can be used and each have their own best use cases.

The Mean

What most people tend to understand as the average, add the numbers together and divide by the number of numbers, e.g 1 2 3 2 1 1 2 3 4, the mean is 2.1.

The Median

Arrange the numbers in order and find the middle number, e.g. 1 1 1 2 2 2 3 3 4, the median is 2.

The Mode

Count how many times a number occurs, the one that occurs the most often is the mode, e.g 1 2 3 2 1 1 2 3 4, the mode is 1


You can see that by applying a different definition of average, you get a different outcome. Here ends the maths lesson.

But, take that methodology and apply it to your customer base. Get hold of a list of all your customers across the last 12 months with a total spend figure next to them. Work out the Mean as an average.

Now apply the Median methodology. What this will give you is a number that 50% of your customers spend less than.

Has a grenade just gone off in your head?

Drawing on a story from my own experience, a piece of customer research came back that said a brand had a 30% share of wallet, nods of approval all round and congratulations on doing a great job. Fast forward a couple of days and while the accepted belief was that the average customer spent around £4000 the data showed that 50% of customers spent less than £800 in a year.

The grenade moment was not just that the difference between mean and median was so great, but that an £800 spend wouldn’t even cover 1 projects worth of materials, never mind the 30% share of wallet across their market that they thought they had. You can’t buy a basic kitchen and appliances for £800, so half of the customers that walked through the door didn’t buy what the brand sold.

This is the sort of data and insight that can lead to a marketing communications strategy that targets customer behaviours at a more granular level, increasing effectiveness and reducing wastage.

Monitoring the change in a metric across periods of time is a valid methodology but taking it in isolation can remove its effectiveness.

Scratch a little deeper into your data and build a view of your customer base, monitor their key metrics and look for the real signs that will challenge you and your business.