Compromise is a Failure in Continuous Improvement

There’s a general feeling that compromise is a good thing.

In most cases, in the real world where relationships is the ‘product’, it is. Compromising on how you pick movies, or who does what chores, or selecting meals is pretty important in relationships.

But at work, compromise is generally a failure. That’s because relationships are a means to an end. Good relationships are important at work, but you don’t judge a company’s performance by how well the employees get along. You judge it by its profitability. Good relationships certainly correlate strongly to profit, but banks won’t lend money to an unprofitable, but cooperative company.

That’s because work is process driven. That means personal preference, the driver for most compromise, should not be much of a factor. What should dictate decisions is results. Hard numbers. Stopwatches and calculators.

Let’s say that two people doing a process are trying to decide where to put a particular tool. One person wants it in location ‘A’, and the other person wants it in location ‘B’. Do you compromise and split the difference and put it right in the middle where nobody wanted it? Of course not.

Do you take turns choosing where to put tools? No. You measure.

You get objective data to support which location is better. If you have a process set up, and it is followed properly, one of the tool locations will be objectively better. It might only save a few seconds over the other location, but one spot will be better.

(NOTE: If everyone isn’t following the same process, you’ve got bigger problems than where to put a tool!)

Now, there can be some debate when it comes to which metrics are most important. Let’s say a decision has the effect of making quality and lead time work in opposite directions—i.e. you’ll get better quality at the expense of slower service, and you have to choose between two competing improvement ideas. One highlights quality. The other makes delivery better.

Even in this situation, you can come up with a pecking order. I teach that hitting your quality metrics is most important. Then hitting your delivery metrics. Then your cost metric. Safety is non-negotiable. If you are missing on your quality metric, go after that. If you are hitting on quality, but your delivery is only so-so, that should drive your choice, as long as it doesn’t make you miss on quality.

So where will bad compromises come into play most often? The most common time we see compromise in the improvement world is when there are competing ideas for what to change. I’m sure anyone who has been around continuous improvement for a while can think of a debate in a kaizen room where the merits of two ideas were discussed ad nauseum.

In those situations, we often, eventually, decide to use a vote or something similar to make a decision. Or we agree to flip a coin. Basically, we compromise. Instead of measuring, we try to make people feel better about the decision-making process.

The truth is, most of these decisions can be made by simply selecting some judgement criteria, and then measuring the results (or making reasonable estimates if cost is prohibitive about trying out both ideas).

Go to the shop floor and have a person stand holding the tool in both locations before you spend money moving benches around. Mock up the workstation out of cardboard in a parking lot and test out two layouts. Get creative on how you gather facts, but get the actual facts.

OK, I’ll admit that this article is a bit tongue in cheek. I’m not really saying to stop any compromising at all. I’m just saying that you shouldn’t debate things that can be objectively measured. In most cases, we compromise because we don’t gather data first.

In that tool example above, if the discussion was , “location ‘A’ is 4 seconds faster than location ‘B’ on a 82 second process”, would we even need to talk about compromise? Is there anyone who would even advocate for the slower location?

Get the data and make good, informed decisions.