Is measurement crucial to success?

Jack Welch former CEO of General Electric has been quoted on management thousands of times, “If you cannot measure it, you cannot manage it,” is among the more famous ideas.  Is that true?  Does measurement always convey meaning?

There is little doubt that most aspects of business require measurement to maintain order and discipline, to allocate scarce resources and to report to governmental agencies and the public.  There are times though when measurement of this type is counter productive, even harmful.

The flaw is threefold.

1. Not every important thing can be objectively measured in the short run
2. Too much precision adds little value to most situations,
3. Many people think the inverse must be true..

Measuring things before their time is likely to mislead rather than inform.  It is impossible to decide what value to assign to a one third finished research project.  Attempting to assess the value of impossible to value things is a failing option.  But everyone does it.  For taxes, for financial reporting, for internal management.  Smart managers know enough to pay little attention to this kind of information.  Regulators and most investors do not.  You are usually better off knowing nothing than knowing something that is a guess.  A 6-year-old child is one third done becoming an adult.  If you believe that then you probably believe inventory values in a financial statement or the value of a long term derivative.

Precision adds nothing once you have exceeded the meaning level of the information.  The irrational transcendental number Pi has been calculated to millions of decimals.  49 decimals of Pi are enough to calculate the circumference of the observable universe give or take the diameter of a single hydrogen atom.  If that calculation is your purpose, further precision adds nothing to the meaning of your answer.

Precision can cause confusion.  Years ago water pollution was measured in parts per million.  Now parts per quadrillion are not uncommon.  Does refining the problem by making the filter a million times finer add much.  People take bigger numbers more seriously, so 100 parts per quadrillion sounds more ominous than a ten thousandth of a part per million.  Sometimes precision helps but usually the idea of the size and frequency of a problem is enough to act intelligently.

The inverse is false.  Things that are measurable are not necessarily manageable. In fact measuring sometimes incentivizes people to pay attention and act on what they see. Only because it is measured. Have a reason to measure.

Measurement is meant to give you a clue to meaning.  Measurement is valuable when you compare to standards.  Measurement is valuable if something is changing from one state to another.  Profit to loss for example.  Measurement is not valuable unless you know the limits to how it was found and the purpose for its collection.

It is an old story.  Forests are made up of trees, and knowing everything about a tree will not tell you much about a forest. Knowing too much about trees can keep you from even noticing forests.  Not seeing the general case hurts you.

Detail is expensive.  When it exceeds your problem limit, detail can be a trap.