I Said Pareto Chart Not Potato Chart!
By Charles Hopkins
Published 04/20/2006 | Finance
Does this sound familiar? You were hired for the new management position. You were tasked to turn the numbers around. You take some time reviewing the current situation. Now its time to take a look at the current processes and get your staff together to analyze the data. You tell them that you want to brainstorm; work on a few mind maps, whip out a couple Ishikawas to get started and then have them bring Pareto charts relative to their respective functions.
One of your department heads looks at you and asks Ishiwhat? You know, you reply, a fishbone diagram. Still blank stares. Cause and effect? you say as you scribble out a trout carcass on your white board. Still nothing. Youre starting to think the elevator doesnt go all the way to the top. Youve got your work cut out for you. So you decide to punt. Ok, lets just start with the Pareto charts, you concede. Sir, what is a potato chart? asks another supervisor. Lets take a five minute stretch break and then meet back in here so that I can welcome you to the world of Pareto charts.
A Pareto chart looks similar to a bar chart. It has columns and it also has a line graph. Generally number of occurrences (frequency) is listed on the left side and percentage on the right. This type of chart is used to graphically summarize and display the relative importance of the differences between groups of data. For example, perhaps you have determined, or at least speculate that your widgets are being rejected due to improper fittings, defective sorting machine, too large or too small, or other. If you look at the reports or studies and gather data on each of these reasons for failure, you can then plug the numbers into a chart. You may have assumed the reason for rejection was because the widgets were too large to fit through the tunnel. However your numbers may actually show (the data will validate) that indeed there was nothing wrong with the size of the widget, but rather the sorter was bent, thereby causing the good pieces to bounce into the reject bin.
Typically you isolate five categories to measure. A Pareto chart can be constructed by separating the data into categories. Lets look at another example. If your business was investigating the delay associated with processing mortgage applications, you could group the data into the following categories: No signature, address not valid, illegible handwriting, existing customer and other.
The left-side vertical axis of the Pareto chart is labeled Frequency (the number of counts for each category), the right-side vertical axis of the Pareto chart is the cumulative percentage, and the horizontal axis of the Pareto chart is labeled with the group names (categories) of your response variables. Are you getting the idea? Your bottom row will be labeled: No signature, address not valid, illegible handwriting, existing customer and other. Each title will have a corresponding column associated with it.
Next determine the number of data points that reside within each group and construct the Pareto chart in a spreadsheet program; Excel works very well for these types of charts. The difference between a Pareto and a typical bar chart is that the Pareto chart is ordered in descending occurrence importance.
Once you have your Pareto constructed and you can visually see what the data is telling you, and you will be able to answer a few questions. You will be able to determine the largest issues facing your team, department or business; you will be able to see what 20% of sources are causing 80% of the problems; and lastly you will know where you should focus your efforts to achieve the greatest improvements.
No more guess work. You wont be needlessly wasting more time and money trying to fix problems that werent broken. Call a staff meeting and get to work on your potato, er a Pareto Charts!