People use statistics as a drunk uses a lamppost – for support rather than illumination. – E. Houseman (1903)
Our online activity has meant a ubiquitous lens is shone on our lives where the accessibility of data leaves leaders believing every metric can be measured, compared and leveraged, contributing to a curious new world of seemingly ‘crystal clarity’ that was seldom thought possible.
The problem, however, arises when data collection occurs without the insight required to take contextual influences into account. Leading the way for deeply consequential misinterpretation and misjudgement to arise.
This ubiquitous lens, unless harnessed correctly, stands to confuse as much as it will clarify.
So for me, context is the cornerstone of moving from data to insight in a human way, which will clarify, rather than confuse.
Nowhere is this more important than in the HR function, where the source of data; humans, operates in a fashion that led author Phillip Lieberman to label us as the “unpredictable species.”
This is best illustrated by highlighting a data source that is predictable, like books in a bookstore.
For the bookstore, doing inventory is easy. The metrics needed are things like how many books are on the shelves and how many have been bought.
Once a book is bought, there are no external factors influencing its behaviour not to be removed from that shelf.
The book doesn’t decide that its commute is too long, or that its wife no longer wants it to work at its company, or that its pay packet isn’t satisfactory.
Thus the likelihood of it leaving the bookstore, (unless I am seriously overlooking something) for any reason other than it being bought or stolen is highly unlikely. This is because the external factors surrounding it are predictable.
Humans on the other hand are a different story. Aristotle wrote that we are “rational animals” pursuing knowledge for its own sake. We live by art and reasoning he said.
And whilst I’m sure many HR professionals would love to see their employee population be as predictable as books on bookshelves, myself included, the multiple external factors that make humankind human are unlikely to change soon.
Which makes producing blanket metrics and chasing large numbers such a dangerous game, because in doing so, the individual is naturally rejected. It’s like saying “we have increased our likes on Facebook which means that we are doing better as a company.”
All that we get from this is a top line number. What do you actually mean by this? What does this equate to? If these questions aren’t followed up, then where does the return on investment come from? Is this useless?
HR systems for the most part aren’t built in ways that enable contextual data to be factored in.
This problem becomes even more salient when you consider that HR systems for the most part aren’t built in ways that enable contextual data to be factored in.
Nor will HR systems allow data to be exported for further analysis, leaving you stuck analyzing the top line, leaving retroactive approaches to remain.
The right tools for the right job
Until contextual factors can be included within HR systems and data analysis, I personally believe HR isn’t critically equipped to have a discussion around data at all.
Instead of building a data set from the ground up specifically focussing on important questions we want answers to alongside the external factors that influence them, we tend to take a retroactive approach by attempting to find data to support new questions.
The same applies for the big data phenomenon (which by the way, unless you are Facebook or some other such platform are unlikely to truly have a big data set).
Gartner’s definition of big data is “high-volume, high-velocity and/or high-variety information assets that demand cost-effective, innovative forms of information processing that enable enhanced insight, decision making, and process automation.”
Playing with data sets of this nature without taking into account the contextual factors that directly influence decision-making processes as they happen will ultimately lead to skewed results and misinterpretation.
So my advice to those individuals taking the data approach is simple: ensure you keep the human in human resources.
I am not ruling out this approach whatsoever, however ensuring HR professionals take a step back is key, especially if we are to glean the results required that are free of bias.