The Three Legged Stool of Data

By Joe Torrago on March 8, 2017

The Three Legged Stool of Data

Ever try to sit on a one or two-legged stool? What happens? It will fall over and not support you or do its job of supporting you or your objectives.

The same thing happens with your data. Let me explain. There is a three-legged stool of data where each leg represents a critical element of the cycle of data. The three legs are as follows -

  1. Collect Data
  2. Analyze Data
  3. Make decisions based on Data

All three legs have to occur equally for the stool not only to support your operations, but also to be balanced, similar to an actual three-legged stool. Let us quickly review the three steps.

Step 1. Data collection should occur as close to the point of occurrence as possible. For example, if you want to track a process condition such as temperature, it ought to be recorded as close to, if not in contact with, the process you are measuring. If you want to collect the temperature of cooling water in a thermoforming mold, it should be captured as close to the point of use, not at the exit of the chiller. It does not matter whether you capture it on a stone tablet and chisel, pencil and paper, computer, or a portable tablet. What matters is it is captured and recorded for purposes of Step 2

Step 2. Data analysis can be as simple or as complex as necessary. Simple can be a scatter gram showing locations of occurrences of a particular characteristic, a production control tracking sheet showing predicted quantity against actual quantity, or a Statistical Process Control chart with an upper and lower limit. The analysis can also be complex such as Analysis of Variance, Two Tailed Tests or other statistical methods. Whatever it is, do it and draw a conclusion. That conclusion should be used in Step 3

Step 3. Making a decision on data analysis should be straightforward based on the conclusion reached in Step 2. Is the process in control? If so, great, continue as before. If not, begin a root cause investigation into why not. Not enough information? It is okay to get more data. If you do, the same three legged stool rules should apply to your new data. Most companies fall down on this rule, so be careful. I cannot tell you how many times I work with companies that have file cabinets full of data collection sheets that have never been either looked at or analyzed.

Every company has some data that can be used either to support or to improve its operation. Collect it, analyze it and make decisions based on it. If you are not willing to do all three steps, do not do any as the three-legged stool will not do its job.

Still confused? I would be happy to discuss your needs and what your three-legged stool should look like.