# Tuesday, June 17, 2008

A long time ago, C++ MVP Bruno van Dooren started a new job. He blogged about why the procedures they follow are important:

We make medicine. Ours is injected directly into the bloodstream of patients who are already weakened by their condition. It goes without saying that there is absolutely no room for making mistakes. Any possibility for error has to be eliminated.

Most of us don't have stakes quite that high, but we work in environments where mistakes should not happen. And yet, they do. And more importantly, every once in a while a cascade of mistakes (I typed something in error, then I didn't run the tests because it was such a small change, then someone else just copied without checking etc etc) means that something bad actually happens "in production". Our procedures are designed to prevent this - the bad cascade happens when people don't follow procedures.

I've found that everyone sets procedures aside some of the time. Heck, if you have something wrong in production at 7pm Friday night, you don't leave it like that because you can't raise a second person to sign your release form, you fix it, and you do the release form Monday. And some procedures seem to be always ignored, which is a sign (when I have my I Own The Company Hat on) that they should be changed. But what interests me is that some people almost always ignore procedures, even good and helpful procedures that everyone else follows. Why would that happen?

I don't have a lot of data on this. People don't like to sit down for an open and honest chat with their boss about why they don't do things the approved way. In fact, most folks who never follow procedure go to some trouble to make it look as though they do follow it. What's more, once I become aware that someone routinely ignores a procedure and hasn't come forward to suggest it should be changed, our opportunities for conversation tend to dwindle, since they're usually busy with a new employer :-). But it seems to come down to one of three things:

  • The procedure takes a little extra time. The person is selfish or lazy or both, and skips that time. The pain the procedure saves is not their own, so they would rather have the time to either do other work and look very productive, or relax at work and have more free time. Sometimes this is a cue to me to reduce the time that the procedure adds, but not always. If you had a developer who took the read only flag off all source files so as not to be bothered with the time it takes Visual Studio to check those files out on edit, would that be OK?
  • The person doesn't know or understand the procedure, or doesn't believe that management will change the procedure if they come forward and suggest a better way. This is why we all suffer through endless "presenting the new procedure" meetings. How else can you tell people? How else can you make people believe? Thing is, I don't know anyone who was actually convinced by a "presenting the new procedure" meeting. I do better by following it myself (tough, because for many tasks in the firm I only get near them in those emergency "I'll do the forms tomorrow" situations) and by paying strict attention to whether others are following. I also need to be more transparent about "presenting the new procedure that was changed because X convinced me this new way is quicker but still prevents errors".
  • The person has discovered a way to do the tasks that is quicker or easier, and still prevents errors, (or maybe that prevents more errors, or whatever) but actively doesn't want the rest of the firm to know it. A sort of "I'm all right Jack" approach that makes that person more productive and perhaps more promotable, but does the firm as a whole no service. I often find out about this only long after the person has "left to pursue other opportunities" since hoarding information and increasing your personal success only are not paths to promotion with me, and this behaviour tends to show up consistently in many things the person does.

Of the three, the hoarding one is the hardest to deal with as a manager. Believe me, you will get more gold stars for having the idea that made your whole department faster than you will for being the fastest in your department. Even if, after your new procedure is implemented, you are somehow second fastest, I will know whose brain is saving/making me money.


Tuesday, June 17, 2008 7:53:23 AM (Eastern Daylight Time, UTC-04:00)  #