🚀 go-pugleaf

John, Dan, Quality colleagues:

Metrics, a.k.a., everything can be quantified.  Wrong!  How many times must
this be relearned?

The conflict in Southeast Asia was quantified to death (literally).  Body
counts, truck "kills," whatever.  Explained by the
quantifiers/nonquantifiers (those who use both) to the head quantifier
(only) as not doable (quantify only).  Done anyhow.  Acknowledged by the
head quantifier/number cruncher thirty-five years later as: "Some mistakes
were made."

How come we cannot quantify Warren Buffett's transactions and model them?
The data on his transactions are relatively easy to come by.  Almost
certainly he uses tools that quantify.  But that ain't all.  If it were, he
could put everything on autopilot and go on sabbatical.

A few colleagues have been quite flattering in suggesting that correcting
some treatment errors in traditional financial calculations points to a new
model for corporate finance.  Very flattering, but unfortunately wrong!  My
view does little more than correct handling and logic flaws and some poor
assumptions.  See . . .
http://www.texas.net/~rmartin/nr7aa003.html or
http://members.tripod.com/~Ray_Martin/DCF/B-MFAQs.html
. . . for details.

After (or before) we quantify something, we have to make subjective
judgments about what it will (or did) mean.  Those that make good judgments
succeed, ceteris paribus.  The others fail.  All with the same numbers.
Quantification, metrics if you favor a decaying buzz word, can be a powerful
tool.  But we are a long way from putting the world on autopilot.

Ray Martin
E-mail: RMartin@texas.net
Home Page Menu: http://www.texas.net/~rmartin/RMenu.html


-----Original Message-----
From: John Zavacki <jzavacki@greenapple.com>
To: Multiple recipients of list QUALITY <QUALITY@PUCC.BITNET>
Date: Friday, August 14, 1998 9:50 PM
Subject: Re: Supplier metrics


>In discussing subjective criteria, Dan says:
>> In evaluating suppliers, one often wants to measure
>> subjective criteria.
>> this is done with a survey instrument of some type. Some of the
>> possible things you may ask about are, "Is the supplier cooperative?"
>> or "When he material hits the receiving dock is it packed well?"
>> or "Is the shipping documentation easy to find?".
>>
>It seems as if we have an issue of the semantics of metrics on our hands.
>Cooperation can be measured by things like response time, effectiveness of
>corrective action, numbers of changes made without prejudice, etc.  If
there
>is an operational definition for proper pack, or a standard for pack, that,
>too can be quantified: number of times pack incorrect, etc.  If the packing
>spec entails shipping documentation, that too can be quantified: proper
>location for shipping documentation, etc..  None of these are subjective.
A
>good control plan would cover them all.  Subjective measures are opinions.
>They have nothing but other opinions to benchmark their value.  You might
>still want to ignore the notion of subjectivity (it smacks of the same
>mental models which insist on 'employee performance ratings' and the like)
>and just evaluate suppliers on things you can actually measure.
>
>One subjective measure is the TQM notion of "customer delight".  Can you
>measure it? Can you recreate it? Of course you want to do it, but it's kind
>of like falling in love, you know it when you feel it, but customer delight
>and falling in love don't last, they both turn into loyalty and constancy.
>Those you can measure.
>
>Peace,
>John F. Zavacki
>jzavacki@greenapple.com <mailto:jzavacki@greenapple.com>
>