A few weeks back I blogged about what I am now calling “The Perfect Metric Fallacy“. If you missed it, here’s the definition

The Perfect Metric Fallacy: the belief that if one simply finds the most relevant or accurate set of numbers possible, all bias will be removed, all stress will be negated, and the answer to complicated problems will become simple, clear and completely uncontroversial.”  

As I was writing that post, I realized that there was an element I wasn’t paying enough attention to. I thought about adding it in, but upon further consideration, I realized that it was big enough that it deserved it’s own post. I’m calling it “The White Collar Paradox”. Here’s my definition:

The White Collar Paradox: Requiring that numbers and statistics be used to guide all decisions due to their ability to quantify truth and overcome bias, while simultaneously only giving attention to those numbers created to cater to ones social class, spot in the workplace hierarchy, education level, or general sense of superiority.

Now of course I don’t mean to pick on just white collar folks here, though almost all offenders are white collar somehow. This could just as easily have been called the “executive paradox” or the “PhD paradox” or lots of other things. I want to be clear who this is aimed at because  plenty of white collar workers have been on the receiving end of this phenomena as well, in the form of their boss writing checks to expensive consulting firms just to have those folks tell them the same stuff their employees did only on prettier paper and using more buzzwords. Essentially, anyone who prioritizes numbers that make sense to them out of their own sense of ego despite having the education to know better is a potential perpetrator of this fallacy.

Now of course wanting to understand the problem is not a bad thing, and quite frequently busy people do not have the time to sort through endless data points. Showing your work gets you lots of credit in class, but in front of the C-suite it loses everyone’s attention in less than 10 seconds (ask me how I know this). There is a value in learning how to get your message to match the interests of your audience. However, if the audience really wants to understand the problem, sometimes they will have to get a little uncomfortable. Sometimes the problem is arising precisely because they overlooked something that’s not very understandable to them, and preferring explanations that cater to what you already know is just using numbers to pad the walls of your echo chamber.

A couple other variations I’ve seen:

1. The novel metric preference As in “my predecessor didn’t use this metric, therefore it has value”.
2. The trendy metric  “Prestigious institution X has promoted this metric, therefore we also need this metric”
3. The “tell me what I want to hear” metric Otherwise known as the drunk with a lamp post…using data for support, not illumination.
4. The emperor has no clothes metric The one that is totally unintelligible but stated with confidence and no one questions it

That last one is the easiest to compensate for. For every data set I run, I always run it by someone actually involved in the work. The number of data problems that can be spotted by almost any employee if you show them your numbers and say “hey, does this match what you see every day?” is enormous. Even if there’s no problems with your data, those employees can almost always tell you where your balance metrics should be, though normally that comes in the form of “you’re missing the point!” (again, ask me how I know this).

For anyone who runs workplace metrics, I think it’s important to note that every person in the organization is going to see the numbers differently and that’s incredibly valuable. Just like high level execs specialize in forming long term visions that day to day workers might not see, those day to day workers specialize in details the higher ups miss. Getting numbers that are reality checked by both groups isn’t easy, but your data integrity will improve dramatically and the decisions you can make will ultimately improve.