"Unknown or Unknowable"...yet SHOCKING! [Data Sanity (Part 2 of 5)]

Back to my last post's review meeting scenario [Click here to read]

IChart of "red, yellow, green" tabulated data shown at meeting:

Goal: 10 percent or less. If people are open to a new (and correct) way of defining performance versus a goal, your colleague has met her "tough" goal all year – because its average was 10 percent. And, rather than using the special cause strategies of (1) treating every individual cancellation or no show as a special cause and (2) treating any deviation from the goal as a special cause – especially any individual month that isn't green – you can interpret the chart for appropriate action:  it's all common cause! 

An Example of "Unknown" and "Unknowable"?

"The most important figures that one needs for management are unknown or unknowable, but successful management must nevertheless take account of them." – the estimable late Lloyd S. Nelson, a true statistical giant.

Routine use of traffic light displays unwittingly destroys organizations because it pretty much treats all common cause as special. How much organizational time is consumed by this?

Remember the 0.6 percent alleged increase in the difference between the current year's average of 10.0 versus the previous year's average of 9.4? Given that the plot of the 32 months of data indicated no special causes, there is no difference!

Then there is the additional time-wasting nonsense of leaders making suggestions on what they believe to be the "low-hanging fruit" and holding her accountable by another, even "tougher," goal for the next year. 

Mark Graham Brown has estimated that 50 percent of time executives spend in meetings involving data (such as this one) is waste, which leads to even more waste: the consequence after meetings of people spending time looking for (and no doubt finding!) and fixing the reasons why they didn’t meet some arbitrary numerical goal. 

What is objective of these meetings, then? Is the goal to meet goals or improve processes? 

As Lao Tzu might say: If all of your focus is on meeting the goal, you will not meet the goal.

Frightened people are very clever. They will waste a lot of time and energy doing their best to meet a goal by any means possible – the game of working on the number instead of the process. What do these games cost you? "Plot the dots!" to see whether your interventions are working.

As you will see, any goal has absolutely nothing to do with approaching how one would improve a process.

This is only one example demonstrating how "the everyday organizational use of data" process is a staggering hidden opportunity.

If you can't right now think of at least a dozen similar examples in your work, take my advice from the last post:

1. Think of Lao Tzu's famous line: The journey of 1000 miles begins with a single step.
2. Rewind and restart that journey, but this time take that first step: just plot over time a number that makes your organization "sweat." Begin to apply some of what I have talked about to your chart and watch how the conversations and the ways you think about data change.

"But Davis, what are these common cause strategies you are talking about?"

Be patient. If you have truly rewound, you are going to be amazed at what you are about to learn by just plotting a few critical numbers over time.

This raises some powerful questions you may have never considered:

• What data should be plotted?
• Where are the data? 
• What do they mean?
• To whom? 
• Who should see them? 
• Why? 
• Do they have a clear objective?
• How are they being used?
• Is there anything "wrong" with this data that should be fixed to use it properly? For example:

-- Does everyone calculate it exactly the same? (Probably not – and you will be surprised at the extent of it)

-- If it is a count of incidents, does everyone agree on the threshold that makes a situation go from a "non-incident" (= 0) to an "incident" (= 1)? (Probably not)

-- Simple example: is Pluto a planet (= 1) or not (= 0)? It doesn't matter, just agree or your data are contaminated! Which way is best for your objective that will let you take the appropriate action?

Don't these questions help to integrate and clarify aims and systems all at once?

Unless you ask these questions, you may become a victim of (or even perform!) an alleged statistical analysis I call PARC. Not sure what PARC means? You should – it's the most commonly used statistical analysis! See my post: "Can you prove anything with statistics? Yes...if you use PARC analysis." A hearty laugh awaits you!

Oh, and how many statistical tools have I used so far? 

Next time, two important common cause strategies. 

Meanwhile, just "Plot the dots!" please.