The Infosys global supply chain management blog enables leaner supply chains through process and IT related interventions. Discuss the latest trends and solutions across the supply chain management landscape.

« Beware! Do not mix Procurement and Marketing | Main | Is “Supply Chain function” a critical function to business? »

The Perfect Order – The Danger of Aggregate Metrics

The term “perfect order” has become a well-worn metric for anyone with a supply chain.  Typically, most people think in terms of the basic measures of “on time, in full, in spec,” which means have the customers gotten what they wanted when they wanted it, in the full amount ordered and with the expected level of quality?  Over time, the metric has expanded to specifically call out subsidiary metrics like Right Product, Right Packaging, Right Documentation, delivered to the Right Location, etc.  And all of these metrics make perfect sense.

What gets interesting – and I think a bit misleading – is when you aggregate these metrics  to come up with a single perfect order index, the “POI”, by multiplying all the individual KPIs together.  Some have argued that this makes sense, because looking at just individual KPIs could “lull you into a false sense of security.”  Perhaps. 

But looking at an aggregate index like POI can cover up specific issues in your supply chain performance.  For one thing, the underlying factors could be shifting around significantly from month to month but with no overall change to the composite POI.  You would have no sense if some metrics were trending better while some were simultaneously trending worse. What’s worse, you could be implicitly assuming that all of the component metrics carry equal weight, which is almost never true. 

Bottom line: There’s just no getting away from looking at the individual metrics and their interactions to have a clue as to what is happening in your supply chain.


I believe all metrics suffer from the same issue, but if we look at all metrics in detail, we will soon be drowning in data.

I also think that aggregate values are very useful for comparison against other companies. This means that companies within the same industry may compete in different ways which in turn affects their comparative metrics.

Let us take Cash-to-Cash, which is made up of DSO, DPO, and DOI. At the executive level, especially for the CFO, C2C gives a quick overview of how quickly the company is turning purchases into goods which are sold and cash is collected. Clearly history, and therefore trending is very important. Equally important is the ability to drill down to the details, not only of DSO, DPO, and DOI, but, for example, to be able to look at the constituents of DOI, RM, WIP, and FG.

The same is true for POI. People need a "quick" overview and to trend the customer service performance as measured by POI. Once a trend is observed, the need to drill to the details becomes necessary in order to determine the root cause.

From a supply chain perspective, I think the most important missing link when talking to the execs, particularly the CFO, is being able to articulate the benefit of getting PROJECTED values for metrics. Too often the history is used at this level, whereas the supply chain can give an estimate of what these metrics will be for, say, the next 2 quarters.

I'm in line with Richard's concern over aggregation of KPI not only on order execution but also on other metrics.

Aggregation of KPI is a necessary evil as the management is interested to view top level metrics due to time constraints. These top level metrics project rosy pictures which the management would like to review. The middle management who is responsible to analyze the thorny issue of the KPI components take appropriate action to weed them out.

However, I would like to reserve my comments on Trevor's view related to projected values for metrics as I am not sure how these add value to this issue.

I agree that we have to be careful not to go overboard in trying to follow routinely too many detailed metrics. The Cash to Cash Cycle Time is a perfect example of a good aggregate metric. For one thing, it is computed in a straightforward, mathematical way from DSO, DPO and DIOH. A day less in Cash-To-Cash Cycle Time, whether it is the result of lower DSO or Inventory, or higher DPO, is just as good from a cash flow perspective. (Though, achieving this may have uneven results on one’s relationships with customers or suppliers.)

Changes in the Perfect Order Index metric are not so easily interpreted. The underlying measures of quality, timeliness, completeness, etc. are quite different in nature and are of varying levels of importance to the customer.

Perhaps we could get the best of both worlds by creating control charts, in true Six Sigma fashion, for each of the metrics that comprise the POI, and automatically alert users to unstable, “out of control” processes, thus allowing users to focus on aggregate measures unless alerted otherwise.

The Perfect Order index has some wonderful research quantifying it's payback to the shipper. AMR has done this, and I'd be glad to share slides with anyone that would want them.

AMR Research shows:

10% improvement in the POI leads to 50-cent increase in earning per share.

5% POI improvement leads to 2.5% better ROA

3% POI improvement leads to 1% increase in profit margins.

Further, we can show the value to the CFO immediately simply on the value of invoices paid on time. It takes determining number of pallets per shipment, number of shipments, average value of each shipment, projected accuracy rate, projected damage rate and an internal rate of return. My calculations show that any company with a value of more than $30K per truck and a combined inaccuracy and damage rate approaching 1.75 percent of total can pay handsomely on a shipment for verification of creation of a 100 percent accurate invoice. Again, I'd show anyone that wants to see it.

Changes in the Perfect Order Index can and should be monitored on a shipment-by-shipment basis. Anyone that's not working to determine this as soon as shipment is completed will soon find themselves behind the 8-ball.

There are also benefits to the carriers. They can be found at:

I'd question any carrier that doesn't have you and your shipment's best interests at heart. There ARE methods for solving issues identified by the Perfect Order Index. However, if you truly want to improve, you have to be willing to share the benefits obtained.

Dave, very useful information. Could you please share the AMR research slides?

Are these slides posted somewhere?

Side note: great blog - good topics.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.

Subscribe to this blog's feed

Follow us on

Blogger Profiles

Infosys on Twitter