Does Six Sigma Suck?

by F.

My former employer nominally adopted Six Sigma while I was there, in an attempt to copy GE. Since I wasn’t in an engineering group, I don’t know the impact on product development, but in the service groups (such as mine) it seemed to be ignored totally.

Partly, this was because measuring service outputs is hard if not impossible, unless you are billing by the hour. This is a good thing if you are a service provider (HR, business development, legal) because it means you are less accountable than, say, salespeople with a quota. As a service provider, there is nothing to account, so you can game the system. Largely, you will be rewarded based on the perception of your value, which is highly manipulable (if you have the right skills). The most important skill is lying, followed by bragging to communicate your lies. For a clear overview of this problem, see Chapter 2 of Fukuyama, especially the section entitled “Principles, Agents, and Incentives.”

I always wondered whether Six Sigma was a hoax or not, or whether it would apply beyond manufacturing. There’s some interesting evidence that it is a hoax, described in yesterday’s WSJ. The evidence comes from a company (QualPro) flogging a competing methodology, so they are not totally credible. But still:

Of the 58 companies reviewed in the QualPro report, 52 underperformed the S&P 500 index from the time they launched their Six Sigma programs through Dec. 5, 2006. Other underperformers include Lockheed Martin Corp., Ford Motor Co. and Xerox Corp.

Ouch. Faced with this data, the adopters pull out their best defenses, none of which seem very good to me:

“Six Sigma is not the end all be all,” said Robert Ferris, a spokesman for Honeywell. “It is simply a set of process tools. We would never suggest that a company’s performance is solely linked to the adoption of these tools.”

“Quality programs are not designed to be measured by a company’s stock price,” says Jeffrey Pfeffer, professor of organizational behavior at Stanford University’s business school and a Six Sigma advocate. Improvements generated by Six Sigma, he adds, “may or may not be reflected in the stock price.”

More data would be nice, but when 90% of adopters underperform the S&P 500 while using the method, you’ve got to wonder (assuming an unbiased sample—it’s of course possible that only screwed up companies adopted Six Sigma. Those firing on all cylinders hardly need to reengineer their processes, one could argue. But these companies don’t sound very screwed up). In any event, it seems that skepticism about six sigma is warranted.