“If you want to retain those who are present, be loyal to those who are absent.” -- Dr. Stephen Covey, The Seven Habits of Highly Effective People

“Learning cannot be disassociated from action.” -- Peter Senge, The Fifth Discipline

“The most important measures are both unknown and unknowable.” -- W. Edwards Deming, Out of the Crisis

Working in the Data Mine by Kevin McManus

First published in Industrial Engineer magazine September 2002

It is nice to know that we now have a fancy, 21 st century term for something that many of us have enjoyed doing for years – data mining. I don't know about you, but I collect data like some people collect baseball cards or Beanie Babies. Crunching the numbers at the end of the month is like getting to read the sports section on Sunday morning. If I miss one month of a particular piece of data, I begin to act in a manner analogous to an eight year old that just needs one more Pokemon card to complete his collection.

People have asked me why I collect so much data, and why I am obsessed in general with data collection and analysis. The answer is simple – I collect and analyze data in order to better understand the systems that affect me and that I am responsible for each day. Numbers tell stories about how systems behave over time. Numbers can also be used to predict what might happen in the future, and to create scenarios that better help us understand a variety of possible futures.

It has been said before that systems give us what they are designed to give us. For example, it is unrealistic to expect the same level of run production from two teams playing baseball at Safeco field in Seattle that you get from two teams playing at Coors Field in Denver . The two baseball systems are significantly different. There might be occasions where the levels of scoring output are similar, but on average, one system is capable of producing significantly more runs per game than the other.

Over the years, I have become more successful with getting people to understand the above concept. Once that hurdle has been cleared, I then proceed to helping them use data to better understand the capabilities and limitations of their systems. This is where things usually begin to get challenging. In general, a lot of people don't like to work with numbers. When you tell them equations and computers are also going to be involved, they really begin to get concerned.

At this juncture, you really have two possible directions you could go. You could make things easy for the users by setting up their spreadsheets and reports for them, or you could tell them what the desired results are, provide them with the resources they will need to get those results, and then offer your help if they need it. Actually, I try to split the difference between these extremes, depending on the users involved. Since I strongly believe that supervisors and managers must know these skills, I tend to ‘encourage' them to follow the ‘learn on your own' path.

In reality, we should be much less afraid of data mining and analysis now then we might have been a little as fifteen years ago. I can remember attempting to work through pages of production forms in an attempt to find performance improvement patterns armed with only paper, a pencil, and a calculator. Learning to use spreadsheet software really helped me increase the level of analysis I could do, but I did not really get good at data mining until I really grasped the concept of databases.

Databases are the veins where potential gold lies. Sadly, a lot of us in management roles go through each work day with little awareness of the wealth that is around us in the form of databases. We contribute to their contents each day, both at work and at home, but we often do not realize the types of information that they contain and that we could obtain from them for the purpose of performance improvement.

Every time we fill out a form or complete an entry screen, we are filling in fields in someone's database. As each second goes by, we are both adding data to those databases that already exist and missing out capturing information that might be key to helping us solve a problem that is really driving us batty. Knowing the basics of database design and use, spreadsheet analysis, and statistics is key to being a successful manager today.

In most cases, a supervisor or manager only needs to enter 15-20 values into a database each day in order to gain a much greater understanding of the systems they are responsible for. I have found that by capturing four to five key values for each of the four performance areas I seek to improve daily, I can create a database that forms a foundation for a wealth of systems performance information. I can then use Excel to calculate and chart the performance ratios that give me daily process feedback.

I have become a more skilled data miner by learning two new skills. First, I have learned how to get more of the information I need from the databases that others maintain in my company. Second, I have learned the value of creating my own databases if the technology was not currently in place (or the time available) to do this within the company system. I can now get the key numbers I need to better understand the systems I am responsible for.

Databases are key to understanding correlations, obtaining system averages and standard deviations, and segmenting data into a variety of combinations for further analysis. When combined with a spreadsheet package, statistical analysis can be performed and graphs can be created. What is the probability that you will exceed your performance goals tomorrow? If you have a process average and standard deviation, and a Z table from a statistics book, that question can be answered pretty quickly.

Data mining is critical to forming theories about how systems behave. Learning to mine the type of data you need in a cost effective manner is a key skill to have if you want be seen as a value added contributor in the workplace of the future. Following the numerical performance of your key systems is crucial to learning more about what really drives the current and desired levels of system output that you are after. Are you getting the data you need each day?

Would You Like to Learn More?

Click on one of the following links to learn even more about Great Systems! and the types of systems improvements I can help you make:

	Great Systems! home page
	More articles on performance improvement
	Systems Change: The Key to Getting Better Results
	Do You Need Great Systems!
	Types of Systems I Can Help You Improve

“The only thing I know is that I do not know it all.” -- Socrates