Another in the occasional series where I have a mild but sustained moan about things in our world that should know better. This one has been building for a while but was prompted by an email this week as “one of the most important HR metrics you need to track.

Calling a test score a measure of effectiveness

Someone, somewhere, decided that dividing a post-training score by a pre-training score and multiplying by 100 counts as measuring training effectiveness. It does not. It measures whether people scored higher on a test after doing training than they did before. That is a different question. It is also, frankly, not a very interesting one.

Presenting recall as performance

The formula returns a number, and that number is real…but only within that system. It is a number about assessment performance, not job performance. Those are not the same thing. A well-designed assessment that genuinely stretches people might produce a low score. A trivially easy post-test will produce a high one. The metric cannot tell the difference.

Offering the wrong advice when the number is low

“If training effectiveness is low, check whether training is aligned with employee needs.” Yes. Absolutely. Also: that is something you should have done before you designed the training. Using a post-hoc score to trigger diagnostic thinking that should have happened at the start is not continuous improvement.

Dressing up convenience as insight

A formula that lives inside HR software is easy to sell. It is a number on a dashboard that gives the appearance of measurement without requiring any of the hard work that measurement actually involves: defining what good looks like before you start, identifying the performance problem, tracking outcomes over time, and involving managers. None of that is in the formula, but that is why these formulas are so attractive. They turn measurement into something tidy, immediate and dashboard-friendly, but they do not turn it into evidence.

And I know I am not exempt

I have spent years arguing that L&D needs to take measurement seriously. Every time someone sells a shortcut and calls it a metric, that argument gets harder to make. So I am not really having a go at one company’s blog post. I am having a go at the whole comfortable ecosystem that makes easy numbers feel like evidence.

It is not evidence. It is arithmetic.

What HR metric annoys you most? Tell me below. Or don’t, I don’t have a KPI for comments.