Wrong usage of metrics could be devastating

credit: www.canddi.com
TL;DR

In this blog post I will compare wrong usage of metric in economy and software testing.

Croatian economy is on the wrong path. One of the main reasons is very low level of economy reforms. We are still in chains of economy from socialism era. Croatian forces that are against change enjoy using wrong metrics in their favor. One of my favorite is:

"It was much better economy situation during the socialism. For example, city of Sisak alone had 30000 jobs."

Using that metric, they want to stress out that unemployment rate in Sisak was very low and that was good for economy.

Software tester does not need to be economy guru, but he/she has at disposal HICCUPPS consistency heuristics. Let's take comparable product (Cuba) consistency heuristics. In 2014 Cuba unemployment rate was 3.3%. Something fishy?

The problem was with efficiency of those 30000 jobs in Sisak.  There were jobs without any purpose, many of workers did not provide any additional value. Let me clear that that was the system problem, not the workers.

In software testing we have comparable wrong metric situation: number of test cases created and executed. If you need to explain your product owner why counting test cases is this wrong metric, you can use this blog post as a claim to prove that this is wrong and dangerous metric.

Labels: