« People Persons | Main | The Right Data in the Right Place at the Right Time »
Thursday
Oct112012

Variation vs "The Number"

We never cease to be amazed in our work with the consistency with which people overlook variation.   It must be ingrained in our society.  It may even be embedded in our genetic code.  People want to be given, and only want to deal with, one number i.e the point estimate.  They do not want to be bothered with having to comprehend ranges or even worse standard deviations. 

This belief or habit is no more prevalent in business than in budgeting and forecast. 

We hear senior executives support this line of thinking all the time.  Many times we have been in meetings where the most senior person in the room says something akin to “just give me the number.”  Often this is more of a command that is barked out in frustration.  When “the number” is communicated, the commitment of the team to hit that number is assumed by the leader and hence by the entire team.  It doesn’t matter weather the standard deviation associated with "the number" may be half of “the number” and thus making the “the number” kind of fuzzy. 

Why are we still in this dealing with this in 2012?   Quality management has been a fixture in business since the 1980s.  Many companies have embraced Six Sigma.  Statistics courses are now required in most business and engineering programs.  Students are even taking statistics courses in high school. People are exposed to the concept of variation more so than ever before.  Modern ERP systems are data rich and have sophisticated analytic methods that could really help communicate the variation in play in almost every aspect of the business.  

We simply do not routinely report variation nor do we use it in most of our decision making.  

We just want to know “the number.”  We just defend or react to "the number."

Back in the 1980s, the days of the quality awakening in this country, the value and importance of control charts, statistical process control charts, was preached to us.  Control charts teach us about the inherent variation in every single thing we do.  We learned about common cause variation and special cause variation.  We were taught the difference between common cause and special cause variation.  We learned to tactically ignore the common cause variation and react to the special cause variation.  We learned that if the level of common cause variation was unacceptable, reducing this kind of variation often required a structural improvement in the process. 

Back in the 1980s, we were given a future view in which control charts were going to be generated for every important measure or indicator in the business.  Control charts were going to be fixture in how we ran our businesses.  It mattered not if it were a manufacturing or white collar measure.  A control chart would help us track the measure and determine if changes in performance levels were significant or not. 

It still sounds logical today!  There is still a need for this today.  Why did it not take hold?  Why hasn’t this become an institutional practice?   With the ERP systems of today, it should be easier than ever to generate control charts of any type, for any functional measure, whenever we want.  We just do not seem to want it... at all.  

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>