First Thoughts
  By Dan Gilmore - Editor-in-Chief  
     
   
  Feb. 17 , 2012  
     
 

Metrics Strategies and Supply Chain Performance

 
 

With the levels of metrics and KPIs most companies have today, which sometimes almost boggle my mind, how is it that supply chains ever really get out of whack? I have some ideas you may find interesting.

A few years ago, "Supply Chain Performance Management" was one of our top 10 supply chain megatrends, and with good reason. However, performance management it isn't going to make our new Megatrends list (coming soon) because the practice of robust performance management it now seems to me is almost ubiquitous today.

Gilmore Says:

I look at what Sara Lee is doing, and I just don't see how it could get very far off course for very long.


Click Here to See
Reader Feedback

In fact, I often hear various pundits say the problem in many organizations is that there are too many metrics. Meaning, by trying to measure everything, you lose sight of those few that really matter.

I agree there are probably cases where there "too many metrics" syndrome is true, but I am not so sure that having detailed, interlinked metrics up and down the supply chain isn't a good thing. Below are just a few impressive examples of what companies are doing with supply chain metrics that I think illustrated where a growing number of companies are today:

A year or so ago I saw a presentation from a Sara Lee supply chain manager that was nominally on its Sales & Operations Planning process - which seemed excellent enough - but what blew me away a lot more was the goal and measurement process behind it. It was really more of an "integrated business planning" type approach versus traditional S&OP, and all the key financial and operational goals were cascaded down multiple levels, with direct linkages back to the top - "line of site metrics," as it is called. Truly integrated performance management, assuming they walked the talk.

Spoke with a supply chain executive at network equipment giant CISCO around the same time, and it is just amazing what all it measures across its supply chain. It would take several columns to do close to justice to how it has instrumented its network, but to cite a quick example, CISCO has developed a number of strategies and tactics to enhance it supply chain flexibility (e.g., a "range forecasting and planning" process by SBU, product family, and individual SKU), and it then creates detailed reporting as to how well those strategies are working - and being used.

At the sort of other end of the spectrum, but illustrating the same basic theme, Procter & Gamble closely tracks what it calls something like "Cases not shipped because of warehouse." That refers to situations when cases are shorted on an order/shipment that were actually in the DC. When that happens, lots of alarm bells start going off, and the warehouse manager gets real nervous. If it happens more than a few times, he or she can expect a visit from the experts in Cincinnati.

I hear these sort of metric stories over and over again now - far different than even five years ago. Virtually every supply chain related case study now includes a detailed "what were the metrics" component. Part of it all this also comes from increasingly capable and integrated IT systems.

So my point is: how, with this level of reporting, can companies possibly go off the supply chain rails? How can there continue to be such wide gaps between the performance leaders, the average, and the laggards, as most recent studies have found there to be?

I mean, I look at what Sara Lee is doing, and I just don't see how it could get very far off course for very long. It seems to be measuring everything, directly tied to its overall goals and objectives. I suppose we could propose that it and other companies don't take those metrics and goals seriously, letting failure to hit the goals continue to just slide, but I just can't believe that is true in the vast majority of cases today, as it might have been in the past.

So, I am left with two choices:

1. Too many companies are being rigorous around the wrong metrics. They are hitting their metrics, but they are the wrong measures.

2. The right metrics are largely in place. But the targets are set at the wrong levels.

My friend Jim Tompkins of Tompkins International has some interesting things to say about number 1, but I am going to hold that for a future column. But I will say his and my thoughts on what should be the right metrics to use is in the end closely relate to point number 2. Too many companies must simply be setting their targets too low.

Who, if anyone, talks to a company in this era that says it is consistently missing its supply chain targets? You last today how many quarters as VP of Supply Chain or a functional manager if that is the case? About two, from what I see. Three quarters if you are really lucky.

So, that would seem to say that companies are not missing their metrics by much or for long. Leaving the logical conclusion that the difference in supply chain performance between similar companies is that they have set different targets, some of them way to low.

I will acknowledge that some difference in targets and thus results could - and often should - be related to different value props, such as whether the company's main differentiation is based on product innovation, cost or service, using the famous and still used framework promoted in the book "The Discipline of Market Leaders" more than a decade ago, by Michael Treacy and Fred Wiersema.

If service is your thing, than obviously metrics like fill rates, cycle times, etc. would have more focus than cost-related metrics (yes, we could debate whether this holds up in practice).

More recently, the concept of supply chain segmentation has re-emerged, and one of the fundamental principles of this strategy is that the supply chain metrics that matter will be different for the different segments being served with differentiated supply chain strategies and service policies.

I must admit in examples such as Sara Lee, CISCO and others that I didn't hear any discussion of segmented metrics, or how their main overall value prop impacted metric targets.

So in general, I am sticking with my conclusion that likely the main driver of supply chain performance between companies is that the leaders set much higher expectations than the middle and the laggards. The driver of performance difference isn't the wrong metrics, or the inability to hit the targets, but that the targets are simply set too low.

After all, how do you really know what the targets should be? More on that, and Dr. Tompkins interesting framework, in a couple of weeks.

What are your thoughts on the relationship between metrics, reporting, and performance? Do you agree with Gilmore's conclusion that the key difference in performance between similar companies often may be that the targets set are very different? Let us know your thoughts at the Feedback button below.

 
 
     

Recent Feedback

Editor's Note:

Thanks – I am more in agreement with the second factor. I think the not be actionable at most larger companies today at least is something mostly of the past. I just don’t think you last too long these days, as I noted, if you miss your numbers for more than a couple of quarters.
 

I will say that there may be some of this because there are so many variables (oil, unit volumes, etc.) that are different than what the targets were set at, so managers and execs can sometimes explain away the misses based on these changes



Dan Gilmore
Editor
Supply Chain Digest
Feb, 18 2012

I think some of the differences in performance are due primarily to two factors. First, companies don’t take action on their scorecards.  In many cases, scorecard metrics are not actionable.  The best scorecard in the world can’t produce results without a company taking action on it. Many firms focus on scorecard creation and less emphasis on taking action on their scorecards. Second, companies measure those things that can be easily measured rather than what is important to the business.  This can be due to not having readily available sources of data to create the desired KPIs.  Or, a firm has not aligned its scorecards with company strategies and goal, by following the approach that you describe at Sara Lee, resulting in metrics that are a hodgepodge of what’s available in their system and what they think they should measure, but not based upon a robust metrics development process.

My experience has focused on measuring supplier performance, but the challenges are similar to supply chain performance.


Sherry
President
Value Chain Group LLC
Feb, 18 2012

I loved your piece on metrics – so true. Actually Debra Hofmann has this discussion with clients all the time and basically people don’t know how to use the metrics, it’s not that they don’t measure enough! Also Matt Davis is doing some fantastic work on SC segmentation with many of our clients, target setting then becomes critical by SC. But first – how many SC’s do you have/need? Is SCDigest familiar with Debra’s hierarchy of SC metrics (below)?

This is a good place to start with the right set of end to end SC metrics to help manage tradeoffs, because that’s ultimately what it’s all about – using the right metrics to make the right conscious decisions, understanding the consequences, in order to achieve the business goal. Because as we all know, if you squeeze something (like costs) somewhere, they typically don’t go away, they pop up somewhere else. 



Jane Barrett
GVP
Supply Chain Research, Gartner
Feb, 18 2012

Hi Dan

Great article focused on a key aspect of supply chain management.  I have 2 different, though related, perspectives of why supply chains go off the rails.

1.    To me the craziest supply chain metric is plan conformance.  In truth I mean measuring plan conformance at more than a daily level, perhaps allowing that in fairly stable industries this could go out as far as a week. This is crazy because of the assumption that he plan was correct in the first place. Several studies, including one by Terra Technology, have shown that a weekly level forecast is typically about 52% accurate, whereas during NPI this drops down to 35%.  Admittedly Terra’s customers improve this to 70% accuracy for mature products, but this still leaves 30% forecast error. (According to Terra, the typical forecast accuracy has improved only a few % over the past 8 years.) Since the demand plan is typically used to drive the supply plan, how accurate is the supply plan?  So why are we forcing the organization to conform to a plan that is inherently wrong, well inaccurate anyway.

2.    All plans are based upon assumptions of both demand and supply.  These are assumption and, in many cases, assumptions are wrong to some degree. And stuff happens too.  Trucks break down, machines fail, yields drop, ships get delayed, containers fall off of forklifts, etc. Today’s extended, global, and outsourced supply chains amplify the effects this disruptions have on the supply chain because there is less time to react because it takes longer to understand the impact of an event because of lower visibility. While every effort is made to buffer against uncertainty and disruptions, most companies cannot afford the inventory required to mitigate against all the risks.

What I am missing in the discussion about supply chain management is what do we do when the plan is wrong.  The discussion for the past 25 years has been about creating the perfect plan, the optimal plan. This is all fair and well if you can get anywhere close to a perfect forecast, but every indication is that we’re not going to get there any time soon.  If we can’t get to a perfect forecast, and therefore we can’t get a perfect plan, what skills and capabilities should we develop to deal with the inevitable mismatch between actual demand and planned supply?  The first is the ability to detect when actual demand is different from planned demand quickly. More important is the ability to determine the impact this mismatch has on the supply plan. How much of the actual demand can be satisfied by supply plan?  Where are the major mismatches? The second is that once these have been identified, how does the supply chain respond quickly and effectively? 

Our view is that planning is the first step. The question is where will the next breakthrough come from: Learning to plan better, or learning to detect and respond to real demand quickly and effectively.


I look forward to your response.


Best regards

Trevor Miles



Trevor Miles
VP, Thought Leadership
Kinaxis
Feb, 23 2012

Note from Editor:


Thanks.


I have noted to several others that the “things change” dynamic, as they do, and how that is a common and in reality often justifiable reason why Targets aren’t met.


Your points as always are good, and the point about what to do when the plan is wrong is of course a great one.


But having said all that, I do believe there in the end of lots of companies that have measures all over the place and think there supply chains are well performing because they usually hit those targets (esp. when adding the “things changed” for all those where the metric was not met). Because, let’s face it, if you are perceived rightly or wrongly as not hitting your metrics and delivering to the company you are gone very quickly.


So:

 

Targets are often set in a way that allows them to be consistently achieved


In general,  and also due to the above dynamic, often leads companies to think they are performing well in supply chain when they are average or worse.


Most companies look at this in aggregate and do nothing relative to supply chain segmentation or service policy optimization


Dan Gilmore
Editor
Supply Chain Digest
Feb, 23 2012

Dan,

Another great article that poses a great set of questions.

Actually, I think both hypotheses are probably correct.

Wrong metrics: I have seen many organizations track performance to a set of metrics, doing things they knew were not optimal for the business, but that drove a payoff, such as a bonus entitlement.  There are others that have not worked through what really counts, and how to measure it. This might be a small company that hasn’t taken the time to work out how to measure customer service as they grew. When the warehouse got full, they couldn’t get orders out, customers started complaining, they realized they might be missing something. Or management only expect purchasing to place orders, so they don’t measure savings or risk mitigated, because no one cares.

Wrong Targets: I am reading “Good by Choice” by Jim Collins, and the three c ore concepts are highly relevant. I think they can help companies develop the right plans, and select the right metrics to track progress and identify issues early.

Fanatic Discipline: Pace your organization to work at a reasonable tempo all the time, and they will become really good at doing a solid job. Continuous improvement and lean/6 sigma initiatives can stretch that every day, major upgrades can change the game.

Empirical Creativity: small investments with clear objectives allow you to work out what the right big investment is, whether it’s new equipment, a new product or service, or a new market.

Productive Paranoia: Understanding risks and evaluating how significant they are and what to do about it if they happen. I am not suggesting that anyone could have forecast the Iceland volcano eruptions and the effect they had on flying, the Japanese Earthquake/Tsunami, the Thai and Australian floods, or the unrest in the Middle East. But brainstorming on what major uncertainties could certainly have informed contingency plans. Air Travel was halted after 9/11; the 2004 tsunami disrupted southeast Asia for months – Haiti is still suffering from the destruction; There have been signs of worker unrest in China (apparently many employees have not returned to work after they went back to celebrate the New Year). Great companies, if you believe Collins, have a plan to deal with this unpredictable disruption. Surges and collapses in activity can be assessed this way too.

Great job Dan, keep up the good work.


Nick Seiersen
Founder
Seiersen Enterprise Inc
Feb, 25 2012

Dan,

I believe that all too often Supply Chain metrics aren't measuring the items that truly impact Customer Service and Profitability. We all know that people react to how they are incented so the real trick for the Supply Chain practicioner is getting the metrics aligned with the true levers on the business and having the entire supply chain working to improve those metrics - that "line of site" you mention - that's much, much easier said than done. I'd also argue the the average and laggards don't really have true Metrics - they may be measuring something (and often they are not) but it's not a metric that is linked to business success. I also agree with Trevor - measuring performance to plan is meaningless (unless you're evaluating your ability to forecast) - from an operational success point of view you might be 'nailing the plan' but if your customers aren't ordering exactly to plan you may have unmet demand (or almost as bad, unsold inventory).

Thanks again for a great blog! Bill Seliger, CPIM, CSCP
http://billseliger.blogspot.com http://www.linkedin.com/in/billseliger


Bill Seliger
Director Supply Chain

Feb, 27 2012

Hi,

I would like to argue that Supply Chain Performance Management is not a topic to take off the list – but instead focus the discussion on ‘which metrics matter and why’, and steering customers to knowing that it doesn’t have to be that hard – especially with analytics technology and applications in the market today.

Our SCPM product which won the SCC technology innovation award – makes it easier for supply chain professionals focus on metrics that matter, and provides them an easy path to identify the root cause of a metric trend or an exception generated by the system based on a combination of metrics – and proactively manage the situation.  Even more compelling if focused on metrics that matter is then the comparison to cascading goals – financial and operational.

You’re right – target setting which starts to help establish the framework, would help compare performance against a certain goal/objective.  The app also has a way to capture these targets – example: what should my cash-to-cash cycle time be?  But it requires an organization change in mindset, which I do hope your columns can steer the companies to.



Padmini
Analytics for Industries and Lines of Business
SAP Solutions Marketing
Feb, 28 2012
.