Performance measurement with the right metric is the cornerstone of all improvements efforts.
If you don’t or can’t measure where you are now versus where you were before, then how do you know if you have improved?
Just as importantly, however, is knowing if your measure actually reflects your current position or the issue that you are trying to highlight or control.
I don’t mean knowing whether the metric you use is the correct one but whether the value actually reflects the issue.
One of the things that leads to problems in this area is that sometimes people mistake precision for accuracy. Let me explain.
First of all, in the way that I have used the term, a metric is a unit of measurement that quantifies a characteristic.
For example, some inventory metrics are stock turns and inventory value. These metrics enable us to quantify the inventory characteristics and, by comparing measures, any change in value.
Precision refers to the number of significant places to which a value is measured, sometimes called exactness. It is also used to reflect the consistency with which something can be measured.
Accuracy relates to the measurement being true and correct or free from error.
So why do these definitions matter?
Sometimes people will select the right metric but not be so careful about the input data or assumptions they use in the measurement. However, because they use a spreadsheet to make the calculation they will be able to calculate a value to 10 decimal places (or more). This precision gives the illusion of accuracy.
You will probably see this a lot when people are presenting percentages. I am sure that you will have encountered reports with percentages shown to two decimal places (e.g. 85.63 %).
Here is an inventory example.
Many companies measure their slow moving inventory and I sometimes see reports that, for example, 23.47% of inventory is slow moving.
The use of two decimal places gives the impression of accuracy. But in terms of decision making it probably doesn’t matter whether the value is 22%, 23% or 24%. The idea is that almost a quarter of inventory is slow moving.
In this case I would ask if the value truly reflects the slow moving stock or merely reflects a convenient way to measure stock.
Many companies measure slow moving inventory as stock where they hold more than 12 months supply. But this doesn’t reflect that the stock is slow moving, it reflects that it is over stocked.
This difference matters because the action that you would take differs for slow moving stock (i.e. the stock doesn’t move much) against overstocking where the stock might move daily but have been over ordered.
In this case the metric is OK – measuring slow moving stock – but the way it is measured is wrong and this means that a precise value does not accurately reflect the problem.
Precision and accuracy can be confusing issues but understanding the difference, and using that understanding to challenge the conclusions you reach from the data, can make a big difference to your improvement program and subsequent results.
Click here for information on our Practitioner Playbook.
Author: Phillip Slater