Tuesday, March 27, 2007

Six Sigma - Sustained Benefits or Just Another FAD

posted by ShyK at 00:36

Six Sigma got some negative publicity after a CNN republication of a Fortune magazine article by Betsy John concluded Look Out & not in and quoted Charles Holland the CEO of Knoxville, Tenn. based consultancy firm Qualpro to debunk Six Sigma.

Around the same time Karen Johnson from Wall Street Journal (paid content) wrote an article attributed The Home Depot’s stock price decline (and subequent Bob Nardelli exit) to Six Sigma.

Scott Adams then authored a Dilbert strip where Dilbert recommends PHB to try some fad which was not debunked - and this is probably what gave the conclusions wider publicity than anything else.

According to Wikipedia, the gist of the Charles Holland article is that Six Sigma is effective at what it is intended to do, but that it is "narrowly designed to fix an existing process" and does not help in "coming up with new products or disruptive technologies." The above quoted article doesn’t say that, but in another interview Dave Cochran of Qualpro says that Six Sigma tends to have the following disadvantages

  • The process improvement becomes the responsibility of a Six Sigma expert (who is probably trained on it for 4 weeks) and not the line staff responsible for process
  • The approach is to breakup the whole process and look at a part there-of, so a local optimization may not necessarily cause global optimization
  • Takes immense investment which could be utilized elsewhere

Charles Holland's views are available in the Chief Executive Journal but his views have been widely challenged - though most of the criticism centers around the fact that Charles Holland recommends the MVT methodology from Qualpro. This article primarily suggests that

  • Off the 58 companies that Qualpro announced, 53 trailed S&P 500 index over a 5 yr (or less) period
  • Post 2001 - While General Electric the Poster Child of Six Sigma has trailed the S&P500 by 30%, Ford and The Home Depot have trailed it by 30% & 60% respectiveliy
  • Its important to test any of the process changes recommended as Qualpro experience indicates that 25 percent of proposed changes help, 22 percent hurt and 53 percent make no difference to a process at all

Critics have questioned the data used to arrive at the conclusions. These center around

  • Measure Used – Stock price (alone) not being a true reflection of success
  • Credibility of the sample – why only 58 of Fortune 200 companies, why these 58 and the difference between announcing & implementing Six Sigma initiative
  • Selective data – GE Implemented Six Sigma in 1996 and saw a sustained stock price increase till 1999 and apparently even a 3:1 bonus in 1999 while the Charles Holland article (probably) refers to post 2000 data
  • The fact the Charles Holland is using the data to sell Qualpro's MVT methodology

To put this in perspective - an independent set of Analysts (paid content) analyzed some 40 Six Sigma project announcements across 6 firms and concluded that “in the long-run, stock performance of Six Sigma companies did not significantly outperform S&P 500.”

On the other hand, an article in i-Six Sigma magazine (paid content) claims that Six Sigma has saved Fortune 500 companies $427 Bn since 1987. This article suggests
  • Of the top 500 public companies in the United States, 53 percent have deployed Six Sigma to some degree
  • Those Fortune 500 companies with the largest revenues are more likely to have a Six Sigma initiative. Eighty-two percent of the top 100 companies use the methodology, whereas only 27 percent of the bottom 100 companies use it.
  • The 47 percent of Fortune 500 companies that have not yet embraced Six Sigma have $500 billion that could potentially be put to their bottom lines.

So, is Six Sigma just another case of "Lies, Damn Lies & Statistics" or is it another great methodology that has become a buzzword being touted by PHB's as a panacea to all ills


Link to Six Sigma - Sustained Benefits or Just Another FAD Post a comment 0 comments
Links to this post:
` `