A
friend recently sent me a link to a LA Times op-ed piece that used Sandy
Koufax’s 1965 perfect game as a data collection object lesson. This was of interest to me because I was a
big Koufax fan (the only time I ever cut class was to watch him pitch his gem
against the Minnesota Twins in the ’65 World Series) and I have an interest in
data as it relates to the world of FM.
The article is both enjoyable and informative, and makes a couple of
points relevant to those of us who collect, analyze, or use data.
You can never have
too much data
Often
people choose not to collect certain bits of data because they do not have or
know of a current need for it. As the
author points
out, however, we sometimes later wish we had data that we could have had if
only we had collected it. With new analysis systems and protocols we can
learn things from old data that help us today.
The example he uses in the article is how collecting additional pitch count data
changed the way pitchers are used, thus extending their careers and value. This data always existed but was not collected prior to the 1980's. The potential benefit of the data existed but was unrealized because it had not been collected.
Investigating
old data can lead to new learning, which in turn results in improved operations
and services. Our FM group is a case in
point. For years we have been collecting
and using data on the full spectrum of services and operations. We collect much more data than we analyze or
report, but when we need it, it is there.
For example, we have millions of lines of BMS data which allows us to
conduct rigorous analysis and troubleshooting when needed.
The point is that we did not have a specific need for the data when we
started collecting it. We just did
because we could. Now, having that data
available opens the door to important insights.
It allows us to investigate root causes, to map relationships among
various indices, and to gain valuable knowledge and wisdom that contribute to
improved designs and process modifications.
Home run or hit-and-run?
Some
baseball fans love the power game, as evidenced by the popularity and
predominance of power hitters. Others
enjoy the nuance of the tactical game.
In the business of process improvement, however, the advantage goes to the
tactical, the nuanced changes. In business it is rare that you can achieve a large and dynamic change, a “homerun” that
dramatically improves operations. It is
much more likely that improvements will be incremental with accumulated effect
over time, much like a single, hit-and-run, stolen base and sacrifice fly will result in a run scored. None of these represent a run by themselves
but each is a key part of the sequence that produces in the end. Would you rather take a big swing with lower
odds of success, or spend the time to gain clarity and swing at exactly the
right spot, at exactly the right time, with exactly the right force? The answer is obvious, isn’t it?
Foresight and
imagination enable discovery
Understanding
at the earliest stages of a data collection project that you do not and cannot
know how that data may eventually be used is an important point. It allows you to take a wider approach to
collection and a more diligent approach to the care of your data, and suggests
a healthy inquisitiveness and respect for the unknown that lies in the
future. Data
that has been diligently collected over time can be synthesized to investigate
specific issues and discern patterns, leading to understanding and
improvement. This may come in the form
of a revised process or identification of a problem or opportunity. Regardless, the result is improvement.
Finally, understand that data survives individual systems. Systems come and go. They become obsolete and get replaced, or a new vendor may bring a new system into your operation. Regardless, the data is tangible, it is yours, and it is a hard asset. Treat it that way both in how you care for it and how you protect your legal ownership rights to it.
Take care of your data, and your data will take care of you.
Great tips on what to do with too much data (or more so when you THINK you have too much data) - I have some data collection tips of my own to pass along!
ReplyDeleteTake a look at this webinar on Metrics That Matter from an owner and investor, Timothy Donahoe. You learn about identifying key areas for metrics and visibility to mitigate risk and enhance investor returns.
http://blog.buildingengines.com/2011/10/executive-summary-metrics-that-matter-take-aways/