It’s easy enough to find out how many customers come in and out of your location each day, or how many are present at any given time. You just count them. But what if you want to know much time the average customer spends in your store (or in a particular part of your store)? Unlike your run-of-the-mill parts or stock, you can’t very well tag a customer as she enters the store, then scan the tag on the way out. For some reason human beings seem to chafe at that kind of thing. (We think it’s probably because the labels leave an annoying, sticky residue on their foreheads.)

Thanks to modern operations management calculations, there’s a pretty tidy solution called “Little’s Law.” According to this principle, demonstrated by John Little in 1961, the average number of flow units in a system (inventory) over an extended time period is always equal to the average flow rate times the average flow time. What this means is that the average time a customer spends in a store can be calculated given an average head count and an overall arrival rate, both of which could be captured discreetly via anonymous entry and exit tracking. Since L = λW, it follows that W = L/λ. In plain English, the average time it takes a customer to move through a process is the long-term average head count divided by the arrival rate. This calculation works no matter how irregular the data are. It doesn’t make a difference whether customers leave in the same order they came in or arrive at regular intervals.

Try this clever technique if you need to know how much time customers are spending at your location, or even just one area (such as a queue).