A couple of weeks ago Gemba Academy filmed my Toyota Kata class and some shop floor work with a live audience at one of their customer’s sites. One of the participants asked a really good question. Upon reflection, I think I can answer it better here than I did “live,” so I’m going to take a do-over.
The team had analyzed their current condition, had established a pretty good target condition, and was working through obstacles.
One of the obstacles was around the fact that the written procedures had not kept up with the way the work was really being performed. This is actually pretty common in industry. The people doing the work know how to do it, and get it done in ways that are better than what is in the documents.
Nevertheless, they needed to update those procedures. If they did not, then new people, or workers that might be rotated into the area temporarily for some reason, would struggle to perform the work in the best way.
This obstacle was not in the way of reaching their target condition process. However they knew the target process would not be stable in the face of people rotation or turnover. The question was along the lines of:
Is it an obstacle if we can’t sustain the target condition unless we address it?
Answer: It Depends
At the risk of bringing up some really old U.S. political humor, “It depends on what your definition of ‘target condition’ is.”
Here is what I am thinking now.
The first step is to get the target process, as they defined it, to work at all. To do this, I would work to control variables, including trying hard to avoid rotating people through there while I am getting it dialed in.
Once we have established that the target process can work with experienced people, then the next target condition might well be to get this process anchored well enough that it will sustain over time without tons of intervention.
Maybe my next target condition is to be able to sustain the target process no matter who is doing it (assuming they have the basic qualification to do that kind of work).
One of the obstacles in the way of that target condition could well be “Our documentation is obsolete.”
Most documentation I have encountered in any industry is actually pretty poor. So this represents an opportunity to experiment your way into developing process documentation that (1) can actually be followed as written and (2) might even be useful for training someone. I’ve never seen that work without a process of iterative trials.
So in this case, I would say “Get it to work the way you intend it to first.” Make that your target condition. THEN start looking at what erodes it if new people step in. In this case, especially, that is going to involve much more than simply updating documentation. How can you set up the work area so that anyone knows what must be done next? What do you need to teach? What do you need to communicate?
I’m Still Thinking About This
Finally, I think this is one of those real-world cases where there isn’t a hard right or wrong answer. There wouldn’t be any harm in updating the process documentation early – except that I expect they will have to do it over once they learn more.
And – not all “obstacles” are actually problems to solve. Sometimes (though less often than we think), there is just something that has to be done that we already know how to do – we just haven’t done it. In those cases, just do it and move on – EXCEPT: Make sure you predict the result of your “just do it,” and CHECK to make sure it worked the way you thought it would. I’ll lay even money it doesn’t, but you won’t know unless you construct it as an experiment. “Just do it-s” usually turn in to “Oh… that didn’t work quite like we thought it would.”
Just make sure you are deliberately learning rather than doing things by rote.