Projects to increase productivity and efficiency in your business have diminishing returns in cases where you can never eliminate that inefficiency entirely. For example, if you achieve a Six Sigma improvement of a 70 percent decrease in assembly defects, a follow-up project that further reduces defects by another 70 percent actually eliminated far fewer defects.

Put another way, say you have 100 defects per month in your organization. Improving this by 70 percent leaves you with just 30 defects next month, so a second 70 percent improvement gets you to 9 defects in month three.

Then say you are Six Sigma masters and eke out a third 70 percent improvement — now you’re dropped to just 3 defects. As you can see, each improvement is smaller and smaller, even though the efforts to achieve each subsequent improvement were almost certainly greater and greater.

At what point do you say “enough”? This has a lot to do with your organization’s goals. (It also has something to do with ethics and the law — there should be no end to improvement efforts to get employee fatalities to zero or prescription drug dispensations to 100 percent perfect.)

For example, most organizations would invest a lot in improving a 60 percent customer satisfaction rating, but be perfectly happy to maintain a 98 percent average without embarking on an improvement project. The potential gains in increasing one additional satisfaction percentile aren’t worth the resources involved in executing a project, particularly relative to an organization’s other, more pressing needs.