Dollars of purchases divided by gross sales – This metric really does not tell management much, in my opinion. Most importantly because it is very difficult to correlate the sales dollars to the purchase dollars in time. Raw materials or goods for resale purchased today may have their revenues dispersed over two or three months or, in the worst of cases, even two or three years. There is almost no direct connection between purchase-dollars spent today and revenues today—or in any given period in the future.
Buying costs divided by purchase dollars – Here is another mostly useless metric. In theory this should be some measure of “purchasing efficiency,” but if you have purchasing agents who are responsible for purchasing a range of goods from high-cost components to pennies-per-thousand nuts-and-bolts, it is not likely that the efficiency on one end of the scale bears any resemblance to efficiency on the other other end of the scale. And, what does the “average” tell you? Precisely nothing.
Also, consider the fact that most companies cannot tell you what their “buying costs” are. Where does the process of “buying” begin and end? If the company assumes that it is the costs associated directly with the “purchasing department,” who gets charged with the time if the buyers’ actions increase costs in the Quality Assurance department or the Receiving Department? Also, who gets charged with the expense of I.B.W.A.—inventory by walking around—when the numbers in the system can’t be trusted and somebody has to verify quantities on-hand in various locations before a purchase order is issued?
Buying costs divided by the number of purchases – This measure is probably better than the “Buying costs divided by purchase dollars” metric (above) because the number of purchase order lines is probably a better root metric than purchase dollars (due to differences in per-unit costs of purchased goods). Nevertheless, this metric is plagued with the same issues with regard to the firm’s actually knowing what their “buying costs” really are.
Dollars of purchases divided by number of purchases – This metric amounts to the value of the average purchase order or purchase order line (depending upon how you define “number of purchases”). But, what does it really tell you? What can you manage by knowing this number? Are you really going to change your purchasing processes based on this number, or will changes actually be based on some other metric?
Percent of purchases rejected – This is likely a solid metric for quality improvement, but before applying it you need to ascertain and evaluate the causes of the purchase rejections.
Value of rejects and monetary adjustments divided by dollar-value of purchases – This, too, is a reasonable measure of quality purchasing. However, the more important metric would be the quality failures’ impact on Throughput. That is, did the quality failures actually keep the firm from shipping product or collecting on shipped product?
Watch for more KPI evaluations in the future. In the meantime, let us know your thoughts about what’s been said here. We look forward to hearing from you.