No One Root Cause

Everybody Loves Analysis

When I first got involved with performance technology, I couldn’t help but notice the amount of emphasis placed on “Analysis.” The implication was that most consultants didn’t do enough (or any, or not the right kind) and that, as a result, often conducted training that was off-target or even unnecessary and, ultimately, an ineffective waste of time and money for all concerned. It might make someone feel like they did something but it wasn’t anything that should have been done.

As I looked at the various writings about this issue, it seems that there was almost an implied ideal model of the eccentric genius who wandered around the workplace poking at things, asking bemused questions, and then coming out with a brilliant but simple solution that everyone had overlooked but, once implemented, resulted in vastly improved performance. (If you are old enough, think “Columbo.” If not, maybe think “House.”)

The problem with this model might be that I just don’t know how to do it right. I admit to having asked questions trying to figure something out and then stumbling on something that others with more know-how have overlooked. It works like this though. It starts with people initially dismissing or ignoring your question because you aren’t the technical expert (and maybe at least partly because you didn’t ask the question clearly) but if you keep after that nagging suspicion and eventually find something, you are rewarded with satisfaction.  You might not get credit though, as the experts may intentionally or unintentionally lose track of where the thing started.  But, once you do this and it works, you actually risk becoming more of a menace to yourself and others because “hey, it worked once” and so you might learn that, no matter what anyone says, never give up…and sometimes, you might not actually have a point.

What is the Root Cause?

But I digress. The issue here is the assumption that there is one root cause to every performance situation. In practice, we often find that is really not the case. We tend to find a bunch of causes, all of which are important contributors and each of which needs a fix to get the improvement we want. But there is rarely a single root issue or one simple thing to fix that solves everything.

Or, in a way there might ultimately be a root cause but it is not manageable. For example, let’s say you have a cable company that uses technicians to install and service their systems.  And, let’s say they don’t define work processes or standards, don’t provide tools, don’t train the technicians, don’t perform quality assurance, and don’t pay particularly well. Let’s say those service people demonstrate poor performance – they often make mistakes, take too long, and irritate customers.  Lot’s of things to fix. If you want one root cause, it would eventually end up being either poor management, out of control business growth (resulting in management not being able to keep up), or greed. I would say any of those “root causes” are way out of reach for a solution.

But before you get to faulty human nature, you could stop earlier and decide to figure out some work processes and document them. You could provide training and reference tools for use on-the-job. You could measure performance and provide feedback and rewards for those that perform above standard and provide corrective feedback, coaching, and eventually discipline for those that don’t. There is a better sequence to undertaking these interventions – for example, defining the work and building the support before you start firing poor performers would be smarter – but there is no one easy thing to fix. You could try incentive and reward the top performers, but rewarding performance without providing the measures (or creating measures but then not using them) would be addressing only part of the problem and would almost certainly result in a failure. In fact, this performance framework (first defined by Tom Gilbert) works for every job and you can use it to find things to fix that will yield results reliably.

At the same time, if you identify the “biggest problem” by prioritizing you might make some progress on a specific issue but the first step would be to get the current process under control before going into focused improvement.

Key principles: Control before improvement. But, if you can’t get control, build the framework of systems, tools, and practices that will support performance.