MANAGEMENT METRICS
UNDERSTANDING AVERAGE
Average is the one number most likely to be wrong ... reflecting nothing
Peter Burgess COMMENTARY
Average is the one number most likely to be wrong ... reflecting nothing
It is commonplace for people to say that 'everyone is different', but when it comes to the data that are used to make big decisions almost everywhere in the eco-system that drives everything in the world we live in, we use 'averages' and 'aggregations' to measure progress and performance and make decisions. his makes no sense. It is insane!!!!!!
I picked up on this issue when I was a student about 60 years ago, and have used this insight all my working life. In fact, I have defined management from time to time as the art of 'moving the average.
Fish Market Prices
I recall doing some consulting work in Kuwait in the early 1980s. There was one fishing company that was substantially more profitable than most of the others, yet they were operating on the same fishing resources and selling in the same market. How could that be. I obtained data about daily market prices from the fish market and was given access to daily sales info from this company. When I got to look at the movement of prices during the day in the fish market there were days when prices went up at the end of the day, and other days when the prices went down at the end of the day. The company with an unusually good profit performance stopped selling on days when the prices went down at the end of the day but continued selling if the prices were going up. It was th only company at the time doing this, and the reason was that at the time it was the only company that had cold stores and freezer capacity to store fish with minimum degradation until better market conditions prevailed.
Production Costing
At one point in my career I was both the CFO and in charge of manufacturing ... a production facility employing about 1,300 workers. This included a non-ferrous metal foundry producing thousands of different castings of various shapes and weights. When I first came to the company the foundry has a really simple (and useless) system of costing information for its castings ... simply put it was total cost of foundry divided by total weight of castings produced equals cost per pound of a casting, and the cost of the casting was therefore this cost times the weight of the casting. This did not seem to make sense to me. For years the company's design engineers had been reducing the weight of the castings, thinking they were reducing the cost of the product (and therefore improving profits ... but in reality the design engineers were making the product less robust and less valuable for the customer and they were increasing the real cost of production because the complex light weight castings were much more difficult to produce in the foundry and scrap rates were very high compared to bigger simpler castings. The average cost number that had been used for years ... actually decades ... did a huge disservice to the company.
Peter Burgess
|