I think it is more succinct and pointed to say that statistical process control (the essential core of Sig Sigma) is only useful when you have a process amenable to statistical control, e.g. one with a relatively small number of well-defined variables and enough data to fit to a statistical distribution or trend. It doesn’t work well with problems that have a large number of variables or a paucity of consistent data, a fact that is readily apparent when you look at the results critically. Unfortunately, that critical analysis is glossed over in most applications of Six Sigma projects (at least the ones I’ve seen) or not done at all because the basic problem isn’t measurable and the participants are not versed in statistics. Most projects that do succeed in a measurable improvement typically do so not because of the methodology but often despite it, and because the participants already know what the solution needs to be (“Stop doing useless, wasteful shit like sending hourly progress reports on the change in height of growing corn,”) and have some impetus, however improperly formulated, to fix it.
The same can be true of a broad array of the type of frameworks and certifications that msmith537 inveighs about above. ISO 9001 certification–than banner of any company that seriously wants to be a supplier to top tier OEMs in any capacity–is a well-intended effort to make sure that companies have well-documented processes and maintain good records on how work was performed so that root causes of problems can be easily identified. But in the attempt to cover every possible industry and type of work, the ISO 9001 requirements don’t actually require a company to generate useful processes and documentation, or even consistently apply the ones they may have; it just mandates that they have documentation of some kind. The result is that a bunch of processes are created for processes that no functional adult should need or would refer to while the critical skilled work which is difficult to train for and document thoroughly is often ignored in a single-minded pursuit of getting certified.
Are you an executive-level manager who wants to actually improve your product or service? Stop pushing “business initiative” magicks down from the top without any real insight as to how they are going to be implemented at a detail level, and provide useful resources and training from the bottom up. Let the workers and line managers tell you what improvements need to be made and how they can be done, and provide the resources you have to make it so. Explicit and quantitative process improvement methods are only useful when being applied by people who understand the problem at hand and can measure the results, and as a secondary benefit, also provide a sense of ownership of the problem instead of just turning the crank on someone else’s pet methodology.
Stranger