Football Season

As this issue goes to press, we’re coming up on the two-year anniversary of the Sept. 11 terrorist attacks. Although national security has increased markedly since then, it hasn’t protected us from threats that are vastly different from those we faced back in 2001.

Paranoia about foreigners stealing American jobs has always been a subject of national debate, but until a couple of years ago it stayed in the background. Now, as the economy flounders and U.S. firms send jobs abroad in droves, both to save costs and benefit from what they perceive as higher-quality labor and output, the use of offshore resources is sparking a public outcry and push toward legislation to protect domestic workers (see “Short Takes,” p. 8).

The frightening thing is that this time, the paranoia may be justified. Back in July, research firm Gartner released a study stating that 10% of U.S. technology jobs will move offshore by 2004, and that jobs in other sectors are expected to follow suit. Although India is the largest recipient of business process outsourcing, with 66% of the world’s total and nearly 90% of the IT work sent from the U.S., countries such as China, the Philippines, and Russia aren’t far behind.

Two factors make offshore outsourcing ominous: It’s profitable for U.S. companies, and it’s permanent. The first means that no matter how unpopular their move overseas, businesses will defend their choice despite the growing backlash from labor unions and politicians. The second means that American employees can’t — as they could in previous recessions — count on getting their jobs back when the economy turns around. According to an article in CIO magazine, “even economists who support globalization agree that the transition for white-collar workers in the coming years will be difficult until new industries arise to take the place of ones with jobs being sent overseas” (Sept. 1, 2003).

Caught unprepared, legislators face a tough battle in the upcoming election year. “This issue is a political football,” says Frank J. Casale, president of The Outsourcing Institute, an industry group. “Jobs are always a hot topic. Whoever supports jobs gets votes.”

Football Season

Whether you’re a star quarterback or a less glamorous order fulfillment manager, getting back to the basics is key to performance and performance measurement. Although broad, boundary-spanning assessments receive more attention from top management, it is the basic measurements that need to stay in the focus of the logistics processional.

When I was growing up, my father would advise me and my brother to think before acting. As typical children, we had a tendency to act, and then think. (Some would argue that we still have that tendency.) Yet, the mantra of thinking before acting still rings in my ears. In a world that is spinning ever faster, with more demands on personal time, it is difficult to get any time to first think, then act. Acting on too many seemingly important issues ruins your focus.

So, let’s get ready for football season — a lesson in the basics. This article focuses on just the few things you need to know. You should take three basic steps to be successful in measuring performance in your company:

  1. In the huddle

    As part of our research for the book Keeping Score: Measuring The Value of Logistics in the Supply Chain (Council of Logistics Management, 1999), which surveyed the logistics measurement practices of 355 companies, we conducted over a score of case studies.

    One interview clearly stands out in my mind today. As part of our standard in-depth interview, we asked a manager to describe his division’s business strategy. After a few minutes of stammering, the manager could not articulate the strategy of his unit. If he — as the manager — could not define the strategy, how could the people that reported to him know and understand their role in supporting the vision of the organization?

    Strategy matters. Communicating the strategy also matters. If people don’t know what the strategy is, they really won’t know how to respond when faced with a problem that needs to be solved.

    According to Michael Treacy and Fred Wiersema in their book The Discipline of Market Leaders (Perseus Books, 1995), companies can follow three basic strategies. First, they can work toward operational excellence, or cost efficiency. The firm focuses on reducing costs as much as possible. Second, the company can become customer-intimate, or customer-focused. Third, merchants can strive to make and bring new products to the market that will differentiate them from the competition. Although many firms pursue multiple combinations of these strategies, when faced with a decision, one of these approaches will typically emerge as the dominant one.

    The reason that strategy matters in logistics measurement is that the strategy defines and determines the emphasis placed on the tools used to measure performance. If the firm is focused on cultivating intimacy with customers, then the measures must reflect this objective. If the firm focuses on operational excellence, the measures must be set to achieve this goal.

    Such an alignment has not been realized for the most part. A recent study conducted by Cap Gemini Ernst & Young, the University of Tennessee, and Georgia Southern University found that the link between a company’s business and logistics strategy is tenuous at best. According to Mary Holcomb, Ph.D., associate professor of logistics at the University of Tennessee, “This is most prevalent in firms that still do not understand the ‘value’ dimension of logistics. These same firms will primarily focus on cost as their approach to managing this functional area.”

    Strategy should not only help managers determine what measures to use, but also assist in determining how much weight or emphasis to place on each measure. For instance, a firm that is working to woo customers shouldn’t ignore transportation costs. Likewise, a firm that is operationally focused, or cost-oriented, shouldn’t reduce costs at the expense of customer expectations.

    The table on this page provides a guide to some of the more commonly used measurements. Based on your strategy and your logistics emphasis, you should be able to select a series of standards — sometimes referred to as “cockpit” measures — that will help you achieve your goals.

  2. Line of scrimmage

    One of the challenges of being a parent is to keep up with the slang and strange phrases that erupt from the mouths of our children. Many times we have to slow down and ask them to repeat the phrase, and define it for us; typically, this is best done while alone, and not in front of their friends.

    Slowing down and defining the metrics should also be done at work, but preferably in front of everyone. In the Keeping Score study, respondents were asked what measures they used and how the standards were defined. The measure could have been defined by the customer, or defined jointly with the customer. Two other choices were also given: The measures were being defined, or the respondent did not know how the measure was defined. The results of the question are shown in the table above.

    What is striking about these findings is the absence of any agreement as to what the measures mean. A full 40% of the respondents have not yet defined, or are in the process of defining, on-time delivery.

    The implications for this are immense, especially given the consequences of not having clear definitions. More than one top logistics professional told us of instances where their measures were in stark contrast to the measures of their customers. Lack of agreement causes confusion, which can be compounded if each of your major customers defines each measure differently. Logistics managers must be aware of these differences and strive to meet the needs of customers based on measures agreed upon by all parties.

    Confusion is more common than we would like. At a recent conference, one of the attendees said that he had visited a major customer, a retailer, to discuss overall performance. Although the supplier’s numbers indicated a high level of service, they were at odds with the customer’s. The problem was that the measures were being defined differently by each party. “That will never happen again,” was the only other comment the attendee would make.

    “If you could do one thing today, it would be to determine the most critical measure for your top customer, and how he operationally defines the measure,” says Dave Durtsche, a senior executive with TranzAct Technologies. “Then, take that information and communicate it to the rest of your team. Maintaining that information and sharing it with your key personnel will assist you in meeting and managing the expectations of your most important customers.”

    Defining measures should be based on more than just feelings. A good measure has several key attributes, described in the table on page 44.

    Coming up with clear standards is much more difficult for direct-to-customer retailers. Instead of dealing with just a handful of major customers, the retailer is faced with thousands of people whose expectations may change based on their last experience with the firm, a competitor, or even another unrelated provider. The challenge here is to understand what is critical to key traditional market segments, such as age, income, and so on. Another way is to segment the customer group based on usage of products or services. For instance, some customers may order a significant portion of their items to be delivered the next day. By contrast, another segment may rarely demand next-day service. In this situation, on-time delivery for the first segment is much more critical than for the second; a service failure in the former case may result in a lost customer.

  3. First down

    The alignment of compensation with overall performance sounds fundamental, but it is probably one of the hardest issues to tackle. When firms focused on functional performance, measuring how well an individual performed was rather elementary. Outputs for the function could be monitored and evaluated. The problem becomes increasingly complex as one focuses on the role an individual plays in a process, not just a function.

    An example might help. Suppose you were charged with evaluating a buyer for a large grocery chain, and were going to look only at how well he performed at a functional level. In this case, one of the measures might be the discounts he got, or the money he saved. A good buyer would have the lowest per-unit cost and would keep that number low season to season.

    Yet, as many of us know, this is only half the story. To get these lower costs, the buyer purchased large quantities of merchandise, which ended up being stored for several months. (One interviewee admitted to having several years’ worth of a raw material purchased on the basis of this type of incentive program). But the need for additional warehouses was viewed as inefficiency on the part of the warehouse manager, who was not given a bonus even though he had nothing to do with the buyer’s behavior.

    How closely related are performance measures to employee compensation? The answer to this question, asked in the CGE&Y survey mentioned earlier, is not encouraging. Only 17% of the respondents believe that measures and compensation are highly integrated, while twice as many (35%) believe that the two are not connected at all.

    “Integrating across functional silos is not easy and ultimately requires the development and acceptance of a new set of measures,” says Rich Thompson, vice president at Cap Gemini Ernst & Young, in the CGE&Y survey. “People act very rationally. When implementing change, it should not be a surprise that people will respond according to how they are rewarded. Executives need to better understand how they would like to reengineer their business processes, and then build new incentive programs to reinforce the expected behavior. Aligning an individual’s variable compensation with specific performance measures that incent [sic] enterprise-wide effectiveness is critical to sustained results.”


The problem with taking the three basic steps is that they are challenging and difficult to implement. But they are necessary to build a successful program in measurement. And the bonus is that working on the basics will highlight other opportunities for improvement in your company.

Concentrating on fundamental logistics principles should not keep one from looking to the future. The hurdles facing companies in the next decade are indeed large, and at times intimidating. Each of them, however, can be used as an opportunity to improve and reshape the firm.

Technology is an area that can be viewed as a threat or as a powerful tool. New service offerings from a wide range of software providers are assisting in the capture of data for performance measurement. For instance, it should be possible in the near future to capture dock-to-dock transit time using GPS data. Detention and demurrage can be tracked based on the movement (or the lack of movement) of the assets. Electronic proofs of delivery will speed the billing process and make it more efficient. Technology will enable the use of more integrated measures that cross several functional areas within the organization.

Unfortunately, the level of technological use is not as ubiquitous as we would hope. According to the recent study conducted by Cap Gemini Ernst & Young, Georgia Southern University, and the University of Tennessee, the majority of companies still rely on spreadsheets, manual methods, or software packages developed in-house to manage transportation and distribution. “Compared to the results of 2000, more respondents are migrating to commercially purchased software; use of manual [methods] or spreadsheets in transportation was around 27%,” notes Dr. Holcomb of the University of Tennessee.

Finally, a word of encouragement. Regardless of where you are in the logistics measurement process, it is never too late to start, and it is never time to stop. Many organizations that we have interviewed and worked with have turned their businesses around based on solid information about their performance.

Karl B. Manrodt, Ph.D., is an assistant professor in the Department of Information Systems and Logistics, College of Business Administration, Georgia Southern University, Statesboro, GA. He can be reached by e-mail at

The Metric System

Effectiveness Measures % Use
Involving Trading Partner
Customer complaints 76.6
On-time delivery 78.6
Over/short/damaged 72.3
Returns and allowances 69.1
Order cycle time 62.3
Overall customer satisfaction 60.8
Days sales outstanding 58.7
Forecast accuracy 54.4
Perfect order fulfillment 39.5
Inquiry response time 29.6
Average 59.5
Internal Focus
Inventory count accuracy 85.8
Order fill 80.8
Out of stocks 70.5
Line item fill 68.5
Backorders 64.4
Inventory obsolescence 62.7
Incoming material quality 61.6
Processing accuracy 45.0
Case fill 39.1
Cash/cash cycle time 32.2
Average 61.1
Efficiency Measures % Use
Outbound freight cost 87.3
Inbound freight cost 68.9
Inventory carrying cost 60.4
Third-party storage cost 58.6
Logistics cost/unit vs. budget 52.4
Cost to serve 37.4
Average 60.8
Finished goods inventory turns 80.2
Orders processed/labor unit 43.3
Product units processed per warehouse labor unit 47.6
Units processed per time unit 37.2
Orders processed per time unit 36.1
Product units processed per transportation unit 21.8
Average 44.4
Space utilization vs. capacity 46.5
Equipment downtime 46.0
Equip. utilization vs. capacity 40.4
Labor utilization vs. capacity 35.8
Average 42.2
Source: Keeping Score: Measuring the Business Value of Logistics in the Supply Chain (CLM, 1999)
Note: Based on 355 responses

Defined to a T

Measure Jointly
In process of
being defined
On-time delivery 31% 29% 32% 8%
Order fill 25% 33% 31% 11%
Line item fill 29% 29% 33% 9%
Invoice accuracy 28% 30% 26% 16%
Order cycle time 25% 25% 38% 12%
Source: Keeping Score: Measuring the Business Value of Logistics in the Supply Chain (CLM, 1999)


Tool Transportation Distribution
Commercially purchased software pkg. 36.5% 40.7%
Software pkg. developed in-house 32.1% 35.0%
Manual methods/spreadsheets 18.6% 14.3%
Third-party provider 11.5% 8.6%
Other 1.3% 1.4%
Source: CGE&Y, GSU, UT Annual Survey (2001)
Note: Based on 434 responses

For Good Measure

Attribute Description
Quantitative Provides an objective value of performance
Easily understood Conveys at a glance what is measured and how
Seeks appropriate behavior Is balanced to reward productive behavior
Visible Measure and causal effects readily available
Defined Has been described and mutually understood
Output and input Combines factors from all aspects of the process
Relevant and important Focuses on key aspects of process performance
Multi-dimensional Makes the proper trade-offs among disciplines
Collects economically Designed to easily capture relevant information
Facilitates trust Validates participation among various parties
Source: Keeping Score: Measuring the Business Value of Logistics in the Supply Chain (CLM, 1999)