Blog

Are PPM Metrics Useful?

I belong to several groups on the Linked In web site. I find some of the discussions both fascinating and informative. Not long ago, on one of the group sites a member posted the following topic for discussion: “Some experts say defect PPM metrics are no longer as useful as they once were, but why?”

Another member named Jerry posted the following reply:

“I’m not in the automotive industry, but a company I was at started tracking PPM as part of a new corporate directive for our quality metrics. In the past we had measured quality performance mainly by customer credit dollars and by waste. The new Quality Director added PPMs to our metrics due to increased customer requests for this measure from their suppliers. I did NOT believe it had value as a measure of quality performance in that industry.

“Example: I had a call from this Quality Director questioning why we had a very high PPM for a particular month, and yet our quality costs were low. I explained that our biggest customer claim involved several thousand pieces of a product that was very small, and had a cost of only about two cents each; thus the low quality costs but high PPM. We have had other instances where we had a claim for only one or two items, yet these items were very costly, so our quality costs were high that month along with a very low PPM.

“The quality costs did NOT correlate with PPM, so in my opinion PPM was not a viable measure there as it might be for an industry where all of the components were of similar value and cost. Still customers were asking for it.”

I posted the following comment in response to the original post and Jerry’s point of view.

Jerry’s point about PPM vs. quality costs may not be the best approach. For example, the high volume of very low cost – but nonetheless defective – components may have had a small impact on Jerry’s quality costs, but were the customer’s process and delivery schedules disrupted? If one must monitor PPM, a better correlation than quality costs would be the severity of the failure.

Even a relatively high PPM failure rate might be acceptable if the severity of the effect is minimal; i.e., it could happen, but with little or no disruption to the customer’s process. On the other hand, a low PPM would be a big deal if the severity rating is high; i.e., non-compliance with regulation; or somebody’s going to be injured; or somebody’s going to die!

When all is said and done, however, I prefer quality metrics in the form of variables data. I always encourage my clients to seek evidence of minimal variation around their targets – and reduced variation over time – instead of waiting around for PPM to count.