2 min read

Feedback at the Supermarket

A while ago, I was at the self-checkout machine paying for my groceries when the machine ate my penny. Strange. Oh well, it's only a penny.

A couple of weeks later, it happened again. Same grocery store. Same machine. "I should tell someone that there's something wrong with this machine," I thought. So I did.

The employee was very responsive. But I suspect she thought I cared about the penny. I did not care about the penny. I cared about the fact that this machine was working improperly – going on at least two weeks now – and someone should fix it.

Did it get fixed? Three or four weeks later, it ate my dime. It's entirely possible that someone was purposefully pilfering pennies, but let's presume the grocery store did not actually want to steal my money (at least, not illegally). This is a feedback problem.

It's a feedback problem that is ubiquitous in large organizations. A "low-level" employee – or a customer or client– notices something is wrong. Or even that something could be improved with minimal cost. But that message never gets passed to someone who is in a position to decide to do something about the problem, much less someone who can actually fix it. So problems persist, week after week after week.

You can think of this kind of this problem mathematically. Here's one proposed way of breaking it down:

  • The likelihood of noticing a problem. How many pennies had this machine nicked without anyone noticing? The notice-ability of a problem is probably only weakly correlated with the importance of the problem. In this case, this is an less important problem that is also hard to notice. But if you visited my house, you'd notice many problems that a reasonable home-owner would ignore because they just aren't that important.
  • The likelihood of speaking up about the problem. This is probably a function of the nature of the problem and the person who notices the problem. Problems that do not inhibit core goals – like with the penny – don't get reported. But if the machine had prevented me from checking out, I would have reported it the first time. A more vocal person reports more problems. I'm probably somewhere in the middle.
  • The time it takes to communicate the problem to someone with power to solve it. I'm guessing this is a function of the formal organizational structure, the organizational culture, and the individual reporter. What channels exist for reporting problems (or "on-the-ground" information more generally), up the chain of command? What latitude do employees have to operate on their own as problems come up? How open are managers and other decision-makers to feedback?
  • The time it takes for the problem solver to act. This seems mostly a function of the nature of the problem (e.g., the effort, time, and cost it would take to resolve the problem, the relative importance of resolving the problem compared to other problems, etc.), the organizational infrastructure (including both personnel, but all available resources and they way they're put together), and other burdens that the problem solver has.

I'll leave the task of formalizing the equation as an exercise for the reader.

Of course, this is all making some pretty heavy assumptions: that there aren't communication failures, that the problem is relatively well-defined and easy to recognize, that solving the problem is relatively straightforward (and won't, for instance, create more problems), among others.

But from my limited experience with various organizations, even the basic feedback problem hasn't been solved in practice.

Depending on the nature of the organization, this can be extremely dangerous. Pennies are one thing. Plumbing problems are another. And infrastructure collapse something else.