Six Sigma and the Perils of Process

We had to read about Six Sigma process management last week for class. Six Sigma is a set of practices that allow companies to improve their processes towards satisfying customer needs, which is a laudable goal. The basic idea is that you have to first Define your goals, find ways to Measure your performance relative to those goals (which has its own problems), Analyze your measurement results, and then Improve and Control your process to narrow the gap between goals and results. Six Sigma calls this the DMAIC model. And that’s where I start having issues (note: this is based on reading 70 pages of a single Six Sigma book so I am extrapolating wildly).

As far as I can tell, Six Sigma is designed to make mid-level project managers feel more important and knowledgeable than they are. Rather than say “First you have to figure out what needs to get done, and then make sure that you are doing what you said you would”, Six Sigma introduces a massive infrastructure with indecipherable jargon like DMAIC and Process POA and SIPOC diagrams and QFD Matrices, terms which have precise definitions to make those who have not studied Six Sigma look ignorant.

People who sign up for the Six Sigma cult are rewarded for their increased study with status symbols leading up to the “black belt”, giving them the title of a martial arts sensei for their command of the minutiae of management jargon. By using this jargon to separate their discussions from the rest of the organization, project managers can feel that they have mastered a discipline, putting them at the same level of achievement as trained software engineers or quantitative analysts. Since this jargon is used to refer to what should be common sense (identifying goals and ensuring movement towards those goals), Six Sigma has the feel of the more bogus aspects of critical theory, where jargon can be used to dress up basic concepts to the point where nobody can determine whether a paper is legitimate.

Our lecturer made the good point in class that the prevalence of Six Sigma does mean that it provides a common framework that one can use to communicate between organizations. Aligning language is difficult, and perhaps Six Sigma has to use artificial jargon to ensure that none of its terms could be confused with terms we might think we know. This would serve a valuable purpose, but the sense I get from Six Sigma is that it’s considered to be an end in itself, rather than a means to more effective communication.

One of the perils of introducing process is that it overrides the initiative and decision making capacity of the people who have the most relevant information. I have been suspicious of process ever since an experience where an organization applied its process so restrictively that we never answered the questions that the process was originally designed to ask. We followed every single detailed step of the process, but we didn’t get any of the benefits that the process was supposed to deliver because we couldn’t adapt it to our particular situation. Six Sigma has the feel of a methodology where that could easily happen. By letting people displace responsibility onto the process (“I did what Six Sigma said I should do”), it distracts organizations from focusing on the core function of delivering value to customers.

This distrust of process may be a result of my never having worked at a large organization. Larger organizations require a certain level of process and infrastructure to function, much like animals need a skeleton to grow past a certain size. And I have seen the disadvantages of not having any process in place at some of the smaller organizations I have experienced. Perhaps if I had worked at an organization where process enhanced productivity and initiative, I might be more amenable to believing in its usefulness.

Process is a powerful, but dangerous, weapon. When used skillfully and appropriately, processes can enable organizations to be more productive and react more intelligently by ensuring that knowledge is distributed to where it is needed. But when process is an end in itself, or is misused in a task for which it is not designed, it can choke an organization and prevent people from achieving what needs to get done. Six Sigma might be great if an organization could use the parts that it found useful to help align language and objectives. However, Six Sigma could be a disaster if if is imposed in a top-down fashion where the whole infrastructure is implemented regardless of whether it is appropriate or not.

8 thoughts on “Six Sigma and the Perils of Process

  1. From what I’ve heard about Six Sigma (admittedly, not a lot) it also sounds like it’s really designed for manufacturers. There seem to be a lot of deep assumptions built into it that will help you create more or better widgets, but that don’t necessarily help if your organization is doing something other than making things.

  2. I think Six Sigma was originally developed for manufacturing, but it’s surprisingly being used in services these days as well. Many of the people in my class who work in finance have to use it, as it’s used to optimize customer transactions and things like that.

  3. Funny… Six Sigma came up during my dinner with Raj on Friday. He told me that they were implementing Six Sigma in his research group, which is a horrible misapplication of the process–he agrees that it should largely be applied to manufacturing.

    However, he did note that some of his PhD research was bastardized and incorporated into Six Sigma–for instance, see this paper.

    His specific argument for implementing this process isn’t that it is really revolutionary: all of these steps, taken individually or in aggregate, are “duh” levels of obviousness: measure what the quality of your output is, and figure out sensitivies of various inputs to optimize where you want to concentrate your improvements.

    However, although the steps are obvious, they don’t necessarily get implemented. The reason is, at least in part, that large organizations fragment responsibility among teams, so “common sense” steps like these can fall through the cracks, or be shuffled off as “somebody else’s problem.” Therefore, by creating a process that Management buys into and supports (which is unfortunately filled with jargon), it actually happens.

  4. I wrote an essay many years ago about this type of thinking.

    Recently, I thought of the real reason why it doesn’t work as well as it should: it fails to accomodate the lunatic fringe trend of the bell curve. Six Sigma is all about FORCING processes to conform to the peak of the bell curve, which is centered around an idea or process which has already been established. Life doesn’t work that way. In Nature, the lunatic fringe, the deformity, the offshoots are what make change (whether for good or bad), usually a little bit at a time. The result is that, over time, the bell curve of a species or behavior shifts to become centered over that ‘different’ focal point. Six Sigma leaves no place for the freak, the random event, or diversity to spread the events of daily life, and thus, new ideas which would possibly improve the future customer base (black swans). It is designed to avoid changes which aren’t already designed into the system. In political and philosophical decision making, this is the equivalent of making all decisions based upon urban living, with no consideration for where the resources come from that make the city possible in the first place, and thus, being completely surprised whenever something surprising happens outside of the structured environment.
    Sensitive dependence on dependency conditions.

  5. This is by far the most accurate description of six Sigma I have ever read, and I have done quite a bit of research into the six Sigma process. I commend you for having so much insight without having worked for a large company that uses the six Sigma model. My wife and I have both work for major corporations, in fact I work for a Fortune 100 company. Both of our companies have adopted the cult like following of six Sigma practices, and have had major detrimental effects (although invisible to them). You have accurately described the precise ramifications that go unnoticed by those who implement this management tool, and yet those of us doing the work see on a daily basis. I am a tech in the telecommunications industry and my wife is a nurse. The policies and procedures that have come out of the six Sigma process have ultimately had the exact opposite effect of the goal, however, the project leaders are getting the data they need and therefore believe it’s working. Neither of us are allowed to take care of our customers or our patients in the way we know is best. Rather, we take care of six Sigma.

Leave a Reply

Your email address will not be published.