The process behaviour chart is the easiest way to differentiate between routine and exceptional variation. This is everything you need to know to use it well.
At the end of this piece, I wrote:
One of the remaining questions that I have is āwhy arenāt process behaviour charts more widespread?ā Of course, one of the ironies here is that while XmR charts may have started in manufacturing, most modern lean manufacturers no longer use them due to the sheer deluge of data they have from computerised production. They have moved on from the kind of vanilla SPC weāre learning here. But statistical process control is used in other fields: medicine, for instance ā the asthma case study in Making Sense of Data is but one example ā which perhaps gives you a sense of the kinds of ānaturalā processes it might be useful to use process behaviour charts with. If you do a google search for ārun chartsā ā which is a precursor form of the XmR chart ā and youāll find a ton of resources from medical websites. Here is one example.
But why not other fields? Why not tech? Why not business in general? Process behaviour charts are simply a tool to separate exceptional variation from routine variation, to help you tell signal from noise. That seems applicable to a lot of domains. I donāt have good answers here. I will say that I find it irritating to look at a raw time series today, without at least an average line plotted through the middle, because I do not know if Iām looking at something meaningful. I canāt tell if itās signal or noise. I yearn for process limits, and I feel mild irritation when I hear someone say āok thatās good right?ā when they see a number go up.
I think I have part of my answer. I recorded a podcast interview with @eric yesterday where we talked about his work on Googleās revenue forecasting team (which heās kindly shared some anecdotes on over at the comments thread on How To Become Data Driven.) And one thing he pointed out is that when youāre working with a process that is exposed to so many exogenous factors, it actually becomes more trouble than itās worth to use an XmR chart.
So for instance, in the comments thread he talked about how Googleās revenues suddenly blipped in the months before the bank meltdowns in the 2007 global financial crisis, and also how their revenue per query experienced a sudden dip when Michael Jackson died. When you operate a large enough business, your business is basically affected by the entire world, which means any geopolitical or macroeconomic upheaval becomes a source of exceptional variation. In such a situation, thereās no alternative but to do the hard work of figuring out variation from āsplit everything down to the smallest variables, then study year-on-year changes for each of those variablesā.
Iām willing to bet that Amazon has the same thing.
Incidentally, this turns out to be consistent with the principles in Wheelerās books. Wheeler repeatedly says, in both Understanding Variation and Making Sense of Data, that āthe purpose of data is insight, which means the best analysis is the simplest analysis that gives you needed insight.ā
If you donāt need an XmR chart to tell you a change has occured, or if an XmR chart is too difficult for your context, donāt use an XmR chart.