Principle 6: Variability (Item 3)
All systems and processes exhibit variability, which impacts on
predictability and performance.
The method you choose to present data and information can help understand
the variation in that data and help you understand your processes.
Tabulated data is useless in showing variation and providing an understanding
of what is happening. Ditto for all the financial presentations we have
seen.
Bar charts have little value and 3-D bar charts should be banned as
they mask the variation.
We strongly recommend the use of Control Charts
where ever possible.
When we talk about variation, we are not talking about financial variance
– the difference between actual and budgeted expenses and revenue.
The terms `variation' and `financial variance' are quite different.
The common reporting procedures about financial variance contribute
nothing to the understanding of the causes of variation and so contribute
nothing to control of the processes that cause it. The common practice
of month-to-month comparisons of financial variance can be destructive
of good business outcomes. It does not contribute to understanding the
causes of variation, it causes focus on minimizing the variance (so
that you does not have to report), causes unnecessary knee jerk reactions
and unnecessary work to explain non-issues and does not recognize the
existence of variation. It is the equivalent of asking, "what caused
that particular wobble?" in the wobbling pen example.
Many people in companies (including executives and managers) point
to last month's results (which is finally better that the trend or average)
and say "see it is improving!" This is very bad management.
It is used to fend off criticism and prevents any serious attempt at
systematic improvement. It will only stop when CEO and other executives
refuse to accept the excuse.
Month-to-month comparison and financial variance are bad measurements.
They should be replaced with reporting tools such as run or control
charts.
Are the numbers 111 and 112 different from each other? This question
is at the heart of all discussion about variation.
Some would argue that "Of course they are different. How could
they not be different? It is a stupid question! If they were the same,
there would not be the need for different numbers."
To some extent the answer to the question "are the numbers different",
depends on whether or not you want them to be. In some of the examples
that follow, you may be saying, "of course they are different".
In others, you may be saying, "of course they are the same".
The answer also depends on the scale. Scale tells you something about
the size of the difference. What if they are 111 miles and 112 miles
and you are walking. What if they are $1.11 and $1.12? Not a big difference
if it was the price of your lunch. But what if that was the unit price
and you had to buy 100,000 of them? What if they are $1.11 billion and
$1.12 billion (a difference of $10 million) and it is your budget overrun?
What if they are estimates of the distance to a star that is either
111 or 112 light years away? What if 111 seconds is the time of the
first place winner in an Olympic event and 112 seconds is the time for
the person who came last? What if they are the diameter of a part and
111 mm is in specifications and 112 is not?
What if we had chosen 199 and 200? Is $199 different from $200? Retailers
know the numbers are the same, but our mind tells us $200 is much more.
Is 74.9 cents per litre different from 75 cents per litre?
What about 6 and 7? And they are earthquake readings on the Richter
scale! Now they represent very different indications of survival and
the extent of the disaster.
These are not stupid questions. You have to make these judgment calls
all the time. The trouble is that without the right tools your response
remains a judgment call – gut feel.
What you are really asking is "are these numbers so different
that I care"? This means that there is some point beyond which
you care about the difference between two numbers. A point beyond which
you want to know because you care. We will call that `the caring point'.
Another way of thinking about this is that as far as you are concerned,
all the numbers close to one number are really the same as that number.
It would be good to find that `caring point'. One way to find the `caring
point' is to ask `what if' questions. Just as we did above.
If the numbers are different and you say they are different, or if
the numbers are the same and you say they are the same, everything is
OK. No problem.
But what if you make a mistake? Do you care? If you fail to identify
that the numbers are different, and you care, you have a problem. Remember
that you only care when the numbers are different enough that it makes
a difference to you. In this case, you may have failed to identify a
problem that could cost you a great deal of money.
What if you made the other mistake and said the numbers are different
when they are in fact the same? Usually when people care about some
thing, they act. In this case, you thought you had to leap into action,
when in fact you did not have to do anything. A wasted effort. And the
wasted effort can cost you a great deal of time and resources. More
chasing smoke.
The decision table below shows the results of our decision – especially
the errors.
| |
The numbers are actually different (and you
care)
|
| |
|
Yes
|
No
|
| You say the numbers are different |
Yes
|
OK
|
The numbers are actually the same,
but you mistakenly say they are different.
Over-reacting
"Much ado about nothing"
|
| |
No
|
The numbers are actually different,
but you mistakenly say they are the same.
Under-reacting.
No action when action was needed!
|
OK
|
An answer that "I care because my boss makes me care" is
not a good one. Bosses who make you care about things that you
do not need to care about are being wasteful – of your time and
resources, and the company's time and resources.
Shareholders should also be very wary about CEOs and senior executives
who waste resources by fighting unnecessary fires and cannot find the
real fires to fight. In other words, shareholders beware if the CEO
and senior executives do not know how to find the `caring point'. They
are probably wasting money chasing smoke or failing to act when they
should. Over-reaction and under-reacting.
We would like to be able to prevent these two types of errors. To do
that we must be much more certain about the `caring point' than just
`gut feel'.
|