Once again why benchmarking does not help with employee survey work

Benchmarking employee survey data leads to the wrong conversations -thus the wrong actions and zero results in most cases.

I was just watching a short video that Mark Effron posted on the problems with  benchmarking.  He joins a distinguished group of scholars and authors who have been discussing why benchmarking does not work for the last few years.  The two people who come to mind are Mark Huselid and Jac Fitz-enz.

It's great to hear someone else explain why benchmarking makes no sense.  Despite the logical case against it, the process continues to be used.  Let me add my two cents to the discussion (ok, it may be more than two):

* Benchmarking is a going backwards exercise.  The data you are comparing yourself to were collected in the past.  It's like comparing your stock price today to the average stock price of your competition from last year.  Who would do that?

* Benchmarking focuses your managers on the wrong stuff.  Benchmarking is a deadly denial barrier; it allows conversations to be diverted to why a score is higher or lower than data that are irrelevant.  Time is taken talking about things that really don't matter when you could be focused on gleaming insights from the data about what is happening today in your own organization.

* If you are going to benchmark, why use just one data set?  When we do compensation studies, we always use multiple data sets because we know there are problems with the data.

* When you collect data matters.  At eePulse we have been doing surveys (short energy pulses) as frequently as weekly since 1996.  We know for a fact that when you collect your data impacts your results.  Trends are better than point-in-time data, but when stuck with point-in-time data at least be clear about what the time period of collection does to your results.  Benchmarking does not take this fact into consideration.

Mark Effron's short video on benchmarking