Retail on-shelf performance of advertised items: An assessment of supply chain effectiveness at the point of purchase

Customer expectations have increased dramatically over the past decade. Indeed, the constant rise in customer expectations is a primary driver of today's supply chain integration initiatives. At the retail level, heightened competition and demanding customers have led many traditional retailers to work to improve two key measures of supply chain effectiveness: on-shelf stock percentage and "total landed cost to the customer's trunk." The fact that channel power has migrated toward the end-consumer which often makes the retailer the channel master, has only increased the importance of understanding the fundamental dimensions of retail supply chain performance.

Retailers are placing more emphasis on on-shelf stock percentage as a bottom-line measure of service performance because of its potential effect on customer satisfaction.' Steams et al. found that stock-outs were the most frequently mentioned cause of frustration for dissatisfied customers.' Traditional fill rates remain important, but they do not capture the all important final customer experience, which occurs when the product is moved from the shelf, to the cart or rack, to the bag. A failure at this point makes all previous supply chain activities meaningless, no matter how flawlessly they have been performed. Quite simply, when customers visit a retail store with the intent to buy a specific product, they expect the product to be readily available. When they find an empty shelf, expectations are unmet, and a certain amount of dissatisfaction results) Moreover, they may decide to seek the product from a competitor. Not only is the current sale lost, but also, if the customer enjoys the purchase experience at the competing retailer, it is conceivable that a lifetime stream of revenue is diminished if not lost entirely. A simple service failure can prove quite costly for both the retailer and the entire supply chain. For example, one restaurant estimates the cost of a dissatisfied diner who chooses not to return loses approximately $25,000 in profit.4 Because the loss of a $10 sale can have enormous consequences, the "last hundred yards" of supply chain performance are absolutely critical.

The potential for dissatisfaction increases when the out-of-stock item is featured in a promotional flier. After all, fliers are designed to entice customers to visit the store. The importance of being well stocked on advertised items is not lost on retailers. According to a Kohl's executive, "in retailing, the biggest single customer-service complaint is not having the item. If Kohl's is promoting Dockers at 25 % off this week, you'd better believe the pants will be in stock. Otherwise, it's like inviting someone into your house and not offering him a seat."'

From the customer's perspective, the advertisement represents an implicit promise that the product will be available throughout the duration of the sale. In short, expectations are raised when items are highlighted in a flier. If these products are not on the shelf, expectations are disconfirmed, and the result is a dissatisfied and perhaps annoyed or angry customer (after all, the store's promise has been broken). When the missing item is the primary reason for the shopping trip, an even higher level of angst is likely to result. Moreover, it is not sufficient to have the product in the store, it must be on the shelf, where the customer expects to find it, since time and convenience are two very important priorities for consumers today. Few shoppers enjoy a "treasure hunt," traipsing through a store looking at end caps and other displays, trying to find a particular item. Even fewer are interested in tracking down a sales representative to ask if the product might be "in the back." Even if the product is on the premises, transaction costs are raised, and the shopping experience is diminished, when the product is not on the shelf.

Of course, a high on-shelf stock percentage is not without cost. In particular, inventory and transportation costs are incurred, and these present a serious tradeoff, since consumers are notorious for demanding low-cost products. As a rule, customers demonstrate little loyalty and are willing to shop around for a consistently lower price. Retailers must have a high level of logistical and supply chain excellence to achieve high on-shelf performance at an acceptable cost. Wal-Mart has become the world's largest retailer largely because of its reputation for consistently having product on the shelf at the lowest prices.6 To mitigate the conflict between the goals of high on-shelf performance and low cost, many retailers adopt postponement-based logistics systems that rely on close relationships with suppliers, advanced information systems, cross-docking, and dedicated transportation services.' These systems include continuous replenishment programs (CRP), efficient consumer response (ECR), and collaborative planning, forecasting, and replenishment (CPFAR).

The research reported here focuses on how well the recent logistical reengineering has helped three types of retailers-mass merchandisers, category killers, and grocery retailers-achieve high levels of on-shelf stock performance, specifically, for items highlighted in national advertising fliers. Our study evaluates the ability of retailers to live up to their implied promise of product availability and thereby meet a prerequisite to outstanding customer satisfaction.

ON-SHELF PERFORMANCE IN THE RETAIL INDUSTRY

Logisticians have long understood the critical role of logistics in assuring customer service,' and several authors have examined customer service performance from the vantage point of manufacturers' industrial or retailer customers. Another stream of research examines the effect of out-of-stocks.10 These studies highlight specific customer responses to stock-out situations. For instance, a 1991 study of grocery store stock-outs, found that 26.8% of buyers did not substitute any product, 20.5% switched brands but kept the same size and variety, 17.6% switched variety but kept the same brand, and 13.7% planned to go to another store." The remaining customers took a wide variety of other actions.

There is almost no recent research on the last hundred yards of supply chain performance. Schary and Becker first noted a lack of academic work on stock-outs in their 1980 article." One of the earliest studies was conducted in 1968 for Progressive Grocer by theA.C. Nielson Co."It determined the stock-out record for 35 grocery product categories at 166 stores throughout the country. For twelve leading chains, a total of 3,372 items were inventoried, or about 20 per store, with each item checked for stock status every day of the week for one week. If the item was not on the shelf, back room status was checked. The data were gathered in a way that allowed Nielson to report on stock-outs for individual items, as well as for all sizes and flavors of an entire brand. During 12.2% of the stock checks, the individual item was not on the shelf, and it was not available anywhere in the store 9.8% of the time. Furthermore, the entire brand (across flavors and sizes) was not on the shelf during 4.4% of the checks and was not available anywhere in the store 3.8% of the time. The research also found wide variations in availability by store chain and by day of week.

In 1971 the Federal Trade Commission (FTC), based on a study of supermarket mispricing and stock-outs of advertised items, issued a rule that made it an unfair or deceptive act or practice for retail food stores to (1) offer products for sale at a stated price, by means of any advertisement disseminated in the area served by any of its stores which are covered by the advertisement, when they do not have such products in stock and readily available to consumers during the effective period of the advertisement; and (2) fail to make the advertised items conspicuously and readily available for sale at or below the advertised price."

More than a dozen supermarket chains entered into consent decrees with the FrC. Additional supermarket surveys by the FTC during 1973 found an unavailability rate of 5.5% across 640 stores in 160 cities. The rule has since been repealed, but it formed the basis for a subsequent rule that requires retailers of various types to offer rainchecks for out-of-stock advertised items."

Given the FTC interest, Mason and Wilkinson undertook an unavailability and mispricing study of discount department store chains in 1981 to determine whether such practices exist in this retail sector." They collected data on nine stores from five discount chains in the Midwest and Southeast. A total of 1,360 advertised items were checked every weekday for two to three weeks to see if they were in stock anywhere in the store. Items were unavailable during 4.53% of the checks made.

A study by Kelly, Smith and Hunt in the 1990s contained some information on stock-outs but was focused on planned versus unplanned shopping behavior." Shoppers were asked why planned and unplanned purchases were not made, and a stock-out was listed as the reason in 13.5% of the cases. Finally, a 1996 study on stock-out performance by Arthur Anderson for the Coca Cola Retailing Research Council suggests that on average 8.2% of items are out of stock." The out-of-stock rate ranged from 3.9% for baby diapers to 11.1 % for yogurt.

Research Hypotheses

Given the importance of in-stock performance and the limited recent research, it was decided that a new study could help answer a number of questions. The first issue is whether there is a difference between stock-out percentages for advertised versus nonadvertised items. The stock-out ratio is simply the number of out-of-stock occasions divided by the total number of occasions the stock status is checked. Although both the academic and trade press report many improvements in logistics service as a result of postponement-based systems, retail industry reports and anecdotal observation indicate that stock-outs continue to be a problem.' In the 1970s, research by the FTC and others suggests that, for whatever reason, retailers have a more difficult time assuring on-shelf availability of highly advertised items as compared to nonadvertised items. In fact, the grocery industry's ECR initiatives highlight the need to improve performance on advertised products."

Hypothesis 1: The "on-shelf," "on-shelf or display," and "anywhere in store" out-of-stock ratios for advertised items will be significantly higher than the ratios for nonadvertised items.

The second issue concerns differences in stock-out levels across a variety of store types. For the three types of retailers examined here (which represent a large portion of all U.S.-based retailers) varied supply and logistical practices are required for success. Specifically, given the wide variety of SKUs that mass merchants must carry, they must manage not only a large and diverse supply base but also a complex logistical support network. Therefore, a typical mass merchant should have a more difficult time maintaining on-shelf availability on advertised items than grocers and category killers. Yet, a well-managed and focused mass merchant could leverage its resources to build a strong logistics infrastructure and use superior technology to allow it to post better on-shelf availability results than other retailers. Likewise, a category killer should have the most focused supply and logistical system and, therefore relatively high in-stock performance. With respect to infrastructural complexity, grocers fall somewhere in the middle, that is, a fairly large number of SKUs but fewer total suppliers than a mass merchant. We suggest that category killers should have the highest in-stock ratio, followed by grocers and mass merchants.

Hypothesis 2: Category killers have the lowest on-shelf stock-out percentage, followed by grocers and mass merchants.

A third issue deals with on-shelf availability by day of the week. Given the Sunday to Saturday time frame of weekly advertising fliers, one would expect availability to deteriorate as the week progresses; it should be worst on Saturday, at the end of the promotional period. A modicum of planning should allow for very high on-shelf availability at the beginning of the promotion, but actual sales are hard to forecast, so on-shelf availability should decrease later in the promotion. Managers can alleviate this problem by maintaining very high inventory levels, but that is generally too expensive. Another solution is to re-supply stores midweek, but without highly advanced information systems, tight supply relationships, and a world-class logistical infrastructure, matching replenishment to flowthrough rates is quite difficult (and expensive).

Hypothesis 3: On-shelf stock-out rates will worsen through the week and will be the highest at week end.

The final issue relates to on-shelf stock-out rates by region. We obtained data from the Midwest and the Mountain West, which represent two very different infrastructural environments. Due to population density, the Midwest is an excellent location for a major distribution center, so it is likely that most retailers are relatively close to one. Also, store density is high, which facilitates frequent deliveries. Yet, stores tend to be older (with less efficient docks and so on) and located in congested urban areas, which complicates delivery. The opposite conditions prevail in the Rocky Mountain West. Several other factors can effect differential stock-out rates. For example, many retail managers believe that unionization as well as the number and quality of store personnel who monitor stock-outs and effect solutions influence in-stock performance. The mix of factors makes prediction difficult, so for exploratory purposes we offer a simple hypothesis.

Hypothesis 4: On-shelf stock-out rates will not vary by region.

Methods

To investigate the on-shelf availability of advertised products, an empirical method involving multiple visits to diverse retailers was chosen. The sample consisted of three large, national mass merchants; four category killers involved in the office supply, building, and electronics subcategories; and three retail grocers. Selection was based on reputation and market dominance. To enhance the overall sample size and increase the opportunity to draw comparisons, two matched sets of stores also were identified: ten in the Midwest and another ten in the Mountain West. The matching was based on chain affiliation whenever possible, such as Wal-Mart in both regions, but not all retailers are part of a national group, especially in the grocery segment. In these instances, comparable stores were selected as the matched pair, that is, two dominant regional grocers and two smaller, independent grocers. This matching was based on overall size and presence of grocer in each region.

For each chain, the national fliers were reviewed for type of product and 20 typical control items were selected for each store (fast moving items similar to those found in the fliers). These remained the same throughout the evaluation period. Promotional items, selected on the day the advertising flier came out, were drawn from the following categories: general grocery, pantry (high-demand staples commonly found at mass merchants), cleaning, personal care, electronics, kitchen, building products excluding bulk products, and computer products. Soft goods lines were excluded because they typically come in a variety of sizes, which greatly complicates the on-shelf determination (i.e., four out of five sizes might be available). The 20 promotional items were selected randomly by the research team in the Midwest. The exact details of the SKU were identified, including a specific size, color, and flavor. Faxes were then sent to the Mountain West research team so that the same or comparable items could be identified (because in some instances different chains were evaluated). Next, the week's data collection instrument was prepared for each store. This had one column for the item description, three columns to identify the in-stock position (on-shelf, on display, in back room), and a final column for comments. The instrument also asked the following questions regarding the level of customer service.

* Was a customer service representative easy to find?

* Was the customer service representative pleasant?

* Was the customer service representative knowledgeable?

* Did the customer service representative volunteer to check the back room?

* Was the customer service representative willing to look when asked?

* Was a raincheck available at the shelf?

* Did the customer service representative offer a raincheck?

* Was a raincheck available upon request?

* How long before stock would be in?

Finally, for each store, a typical, high-volume aisle was randomly selected, and the total number of stock items was counted (the products on this aisle were similar to the types of items found in the weekly advertising flier). During each visit, the number of out-of-stock items was counted. An average on-shelf stock percentage was calculated for this random aisle on each visit.

Each store was visited three times during the week: two days after the flier was distributed, two days later, and after another two days which typically was the last day of the advertised sale. Each store thus had a total of 60 (3 visits x 20 items) purchase opportunities in a given week (if control items are included, 120 purchase opportunities). Because this pattern was followed for four weeks, each store had a total of 480 purchase opportunities. The method was pretested for two weeks before the actual collection period so that researchers could become familiar with the stores and learn where to look when an item was not found on the shelf. At the end of the collection period, brief interviews were conducted with store managers to capture any of their insights about the store's on-shelf stock percentage. Questions about their firm's replenishment model were asked at this time.

To summarize, on-shelf performance was evaluated over four consecutive weeks at ten stores each in the Midwest and the Mountain West. Mass merchants, category killers, and grocers comprised the sample population. Each store was visited three times during the week, and data were collected for promotional items, control items, and a random aisle within the store. The total number of purchase opportunities for the promotional and control items was 9,600. Overall, a fairly comprehensive view of on-shelf stock performance was captured.

Findings and Discussion

The first hypothesis deals with the overall stock-out ratios for advertised versus control items. Figure 1 summarizes the results for overall in-store, on-display, and on-shelf stock-out percentages. The on-shelf figure is the most important indicator of a supply chain's ultimate performance from the vantage point of the consumer. We found that, across the board, the advertised items had a stockout percentage almost twice as high as the control items, regardless of the availability measure. The differences in stock-out percentage between the advertised and control groups were all significant at the .01 level, which confirms hypothesis 1. For advertised items, the in-store stock-out ratio was 12%, and the tougher on-shelf measure was 16.5%. It is important to note that this means an item was not in stock on this percentage of the possible purchase occasions, not that this percentage of the items were out of stock. For instance, one item might have been out of stock all three days of the week. Furthermore, this is the stock-out rate for heavily advertised products, and for similar control items not on promotion during the four-week examination period, so it is not representative of a store's overall stock-out levels.

This out-of-stock ratio is quite high in view of the technology used by retailers today. Every single store and chain in the study captures point of sale (POS) data and uses it with some form of automated replenishment system (ARP). All the retailers are major players in their industry or geographic location and take a proactive approach to supply chain management. The fact that all the items were included in the chain's promotional flyer (free-standing insert) strongly suggests that each chain had major incentives to be sure these products were in stock for the duration of the promotion. Although the FrC once believed that retailers purposely advertised some items as loss leaders when they had no intention of having them available, we do not believe this is the case in today's extremely competitive environment. Information-empowered customers exhibit little loyalty to companies that do not live up to their service expectations. We simply do not believe that a retailer would spend money to raise expectations while planning to disappoint consumers by not stocking promotional items. It is much more likely that these stock-out percentages reflect a failure somewhere between the supplier and the final store personnel responsible for placing goods on the shelf. Because of the critical importance of the on-shelf stock-out measure, in the interest of brevity the remaining discussion focuses on this single performance indicator.

The second hypothesis concerns differences in product availability across retailer type-grocer, mass merchant, and category killer. Table 1 summarizes the on-shelf stock-out percentage for each category. Interestingly, mass merchants, with perhaps the most sophisticated replenishment logistics operations in retailing, had the highest stock-out levels for advertised items; these were not available on-shelf on 22% of the possible purchase occasions. This compares to 11.5% for grocery items and 16.1 % for the category killers (p=.05 for differences between retail categories). Within each retail category, the out-of-stock incidence rate was far higher for promotional than for control items (p=.05). At the individual chain level, all differences between promotional and control items were also significant at the .05 level, with the exception of grocery chain A and category killer chains L and M. Finally, it must be noted that comparative performance within each retail category varied widely, which suggests that management policy and practice greatly influence on-shelf stock performance.

Why do mass merchants have the worst performance for advertised items, followed by category killers, and grocers have the best performance? Although mass merchants are perceived to have the best logistical infrastructures, they also carry a wider assortment than grocery retailers; also, the demand for their advertised goods probably is far less stable than is the case for grocers. Both factors create serious forecast integrity and inventory management challenges. Furthermore, the gap in on-shelf performance between advertised and control items provides strong support for the strategy of offering everyday low prices. With respect to category killers, they face the difficult task of needing to stock very deep lines of specialty goods with very low unit demand.

Hypothesis 3 targets differences in stock-out percentage by day of week for advertised items. Although percentages did deteriorate as the week progressed, the decrease in on-shelf performance was relatively minor. The average stock-out rate was 13.3% on Tuesday, 15.3% on Thursday, and 15.6% on Saturday. The Tuesday - Thursday and Tuesday - Saturday differences were statistically significant (p= .05), but, the Thursday - Saturday difference was not. Figure 2 summarizes day-of-week data. Overall, there appear to be opportunities both to improve forecast accuracy and to re-supply once a sale has begun. The key is to identify "hot" items accurately and quickly and then have the information system and logistical abilities to get these items on the store shelf in minimal time. Interestingly, for grocery and mass merchants, the stock-out percentage was somewhat lower on Saturday, which suggests that some re-supply is taking place.

The final hypothesis looks at regional differences in on-shelf stock performance. Surprisingly, the Mountain West substantially outperforms the Midwest across all three retailer categories (p

As noted earlier, a high-volume aisle was selected for each store. Each day the total items out of stock were counted and a percentage for the aisle was calculated. Examples of selected aisles were cleaning products, grocery pantry items, and computer products. The stock-out incidence for these aisles ranged from 1.1 % for grocery stores, to 5.5% for mass merchants, to 7.8% for category killers. Part of these differences may reflect the far greater volume of items in the grocery store aisles, which ranged from 300 to 2000 depending on the store. The mass merchant stores contained far fewer items, and the category killers had as few as 46. Especially in the electronics category, the aisle might have very fast moving high-value items that are hard to keep in stock.

After data collection, store managers were interviewed to gain some insight into possible causes for out-of-stock situations. Table 2 highlights the basic reasons identified. Managers made such comments as:

* "we don't have enough staff to check on out of stocks and update records;"

* "we don't have staff to get merchandise out fast enough after delivery;"

* "we get too many small deliveries;" and

* "the warehouse was out of the item."

The reasons can be grouped into five categories: (1) supplier, (2) chain buyer and inventory control personnel, (3) distribution centers, (4) transportation provider, and (5) store personnel. From the perspective of store managers, the greatest single challenge lies within the retail outlet, in terms of ordering and getting the product on the shelf. The critical role of store operations points out the importance of managing the whole supply chain right to the final shelf location.

Implications

This study found that advertised items are not available in their normal shelf location on about one out of six purchase occasions. In an era of tough competition and demanding customers, this frequency is surprisingly high, especially in view of the recent emphasis on state-of-the-art logistics and retail ARP systems. The figure for advertised items (16.5%) is almost double that for similar control items not on promotion. Furthermore, managers at several of the chains note that they target 94% on-shelf availability, so performance is falling far short.

Another finding is substantial differences in on-shelf performance by type of store. Mass merchants perform the worst with stock-outs on 22% of all possible purchase occasions compared to category killers (16.1%) and grocers (11.5%). The stock-out rate for one of the mass merchandisers skewed the data somewhat, but, even when the worst performer in each category was dropped, mass merchants still had the highest rate. A likely reason is the wide assortment of goods they carry and the large but erratic volumes of shopping activity generated by their promotional fliers. Also mass merchants frequently place special buys for promotion, and when the product sells out, it cannot be replenished. This practice drives up stock-out percentages and potentially leads to higher levels of customer dissatisfaction.

Other Findings are Noteworthy

* Stock-outs worsen as the week progresses, although the differences between Thursday and Saturday were not statistically significant. Some re-supply takes place, but current information systems, supply relationships, and logistics infrastructures are not capable of matching replenishment to sell rates.

* On-shelf stock-out rates in the Midwest are almost double those in the West. The cause is probably a combination of factors: urban location, older age of stores, low staffing levels, and restrictive union work rules in the Midwest.

* Stock-out rates for all items in a single aisle in each store ranged from 1.1% for grocery retailers, to 5.5% for mass merchants, to 7.8% for category killers. This finding corroborates the notion that stock-outs for advertised items greatly exceed those for nonpromotional items. This raises the question of priorities. If the retailer is going to reduce inventory costs at the risk of a stock-out, is it better to stock out on advertised or nonadvertised items?

* Individual chain performance varies dramatically within retail category. This suggests that policies, logistical infrastructures, and investments in information systems can have a profound effect on on-shelf performance.

* When advertised items are out of stock, it is generally not very easy to investigate their status. Customer service representatives are often hard to find and uninformed "the item might be available when the next truck arrives." Furthermore, employees often seem impatient when asked to check for the item in the back room. Finally, rain checks are seldom available at the shelf or voluntarily offered by the customer service representative (although available when requested). Rather than emphasize service recovery, retailer policies and/or employee training tend to aggravate consumer dissatisfaction.

Overall, it appears that retailers are doing a poor job of living up to the implicit promise they make to have advertised items on the shelf, where customers expect to find them. We doubt that they purposely plan to be out of stock. Too much is at stake in the highly competitive world of retailing, where customer service and relationship marketing are key goals. It is not likely that retailers spend millions of dollars to entice shoppers and then disappoint them.

What is the problem? Several companies have the rather modest goal of being in stock at the shelf 94% of the time; but even this standard is not being achieved. The poor performance raises the question of whether inventory levels may be kept too low in this era of just-in-time replenishment, but the worst performer in this study had one of the lowest inventory turn levels of any chain in the sample. The fundamental problem appears to lie with inadequate policies and procedures as well as poor communication within the store and throughout the supply chain. It seems that some retailers pursue efficiency at the distribution center or even dock level but give much less attention to the last hundred yards.

If supply chains are going to compete successfully, they must go the final distance to the "customer's trunk" and beyond, to after-sales service. This is the only way to assure satisfaction, develop loyal customers, and establish a "lifetime stream of profits." Although supplier, advertising department, and warehouse operations can be the cause of poor performance, our interviews with store managers suggest that store operations' shelf monitoring, special orders, and re-supply of shelves are important as well. Supply chain management must span boundaries to assure the communication and support necessary to achieve in-stock performance. Success across 95% of the supply chain is simply not good enough if the system fails to deliver where it really counts, in the heart and mind of the customer.

Limitations and Future Research

In this study, stock-out rates are for advertised items, which often were at high seasonal demand points. It is also important to note that the items were all hard goods and, especially with mass merchants, items on the random aisles were concentrated in very popular categories, such as home electronics, pantry, and personal care. In addition, stock-out rates are expressed as a percentage of all possible purchase occasions. A single product might have been out of stock on multiple days, so the percentage does not directly equate to the percentage of actual items out of stock. Furthermore, stock-out rates for control items should not be considered indicative of the store's overall stock-out rate. The control items were chosen to approximate the kinds of popular items being advertised and most likely would have a higher stock-out rate than total store inventory. Control items might even be purchased in lieu of advertised items, which would increase their stock-out role.

This study is limited by the relatively small number of chains, store locations, and items studied. It is also possible that the stores in the Midwest, in older urban areas, with low staff levels and a clientele that may be prone to shop advertised items, are more susceptible to stock-outs than the average U.S. store. This may explain the regional difference found. Other limitations of the research relate to the lack of control over the exact time of day items were monitored, as well as the frequency and actual day of the week of re-supply.

Future research should attempt to assure that the stores selected are representative in terms of age, staff levels, patron income, and location. An attempt also should be made to control for the time of day when stock status is checked. It would be useful to increase the number of stores and items studied. Extending the study to a wider variety of retailers is another opportunity. Both mail order and E-commerce retailers are presumed to have an advantage in assuring in-stock performance through centralized single-warehouse inventories, but that needs to be confirmed by empirical evidence. A study of the in-stock performance of such retailers would provide interesting comparisons with more traditional retailers.

NOTES

'Stanley E. Fawcett and M. Bixby Cooper, "Logistics Performance Measurement and Customer Service," Industrial Marketing Management 27 (July 1998):341-57.

2James M. Stems, Lynette S. Unger and Jack A. Lesser, "Intervening Variables between Satisfaction/Dissatisfaction and Retail Re-patronage Intention," in Bruce J. Walker et. al., eds., An Assessment of Marketing Thought and Practice (Chicago, IL: American Marketing Association, 1981), pp. 179-82.

'A. Parasuraman, V. A. Zeithaml, and L. L. Berry, "A Conceptual Model of Service Quality and Its Implications for Future Research," Journal of Marketing 49 no.4 (1985): 41-50; and V. A. Zeithaml, L. L. Berry, and A. Parasuraman, "The Nature and Determinants of Customer Expectations of Service," Journal of the Academy ofMarketing Science 21 (Winter 1993): 1-12.

IR. D. Blackwell, From Mind to Market (New York: HarperCollins Publishers, 1997).

'Anne Faircloth, "The Best Retailer You've Never Heard Of," Fortune, (March 16, 1998): 112. 6G. Stalk, P Evans, and L.E. Schulman "Competing on Capabilities: The New Rules of Corporate Strategy," Harvard Business Review 70, no. 2 (1992): 57-69.

'Susan S. Fiorito, Eleanor G. May, and Katherine Straughn, "Quick Response in Retailing: Components and Implementation," International Journal of Retail and Distribution Management 23, no. 5, (1995): 12-20; Tony Seideman, "What Sam Walton Learned from the Berlin Airlift," Audacity (Spring1996): 53-61; and Michael Hartnet, "Managing Change in Retail Logistics," Stores 76, no. 9 (September1994): 60-2.

$Arun Sharma, Dhruv Grewal, and Michael Levy, "The Customer Satisfaction/Logistics Interface," Journal ofBusiness Logistics 16, no. 2 (1995): 1-21; Jay U. Sterling and Douglas M. Lambert, "Customer Service Research: Past, Present and Future," International Journal of Physical Distribution and Materials Management 19, no. 2 (1989): 1-23; and Lloyd Rinehart, M. Bixby Cooper, and

George Waggenheim, "Furthering the Integration of Marketing and Logistics through Customer Service in the Channel," Journal of the Academy of Marketing Science 17 (Winter 1989): 63-72. Also International Journal of Physical Distribution and Logistics Management, special issue on customer satisfaction 20, no. 2 (1990).

9Theodore P. Stank, Patricia J. Daugherty, and Alexander E. Ellinger, "Voice of the Customer: The Impact on Customer Satisfaction," International Journal of Purchasing and Materials Management (Fall 1997): 2-9; Daniel E. Innis and Bernard J. La Londe, "Customer Service: The Key to Customer Satisfaction, Customer Loyalty, and Market Share," Journal of Business Logistics 15, no. 1 (1994):1-27; and Lisa M. Ellram, Bernard J. La Londe, and Mary M. Weber, "Retail Logistics," International Journal of Physical Distribution and Materials Management 19, no. 12 (1989): 29-39.

"Margaret A. Emmelhainz, James R. Stock, and Larry W. Emmelhainz, "Consumer Responses to Retail Stock-outs," Journal of Retailing 67, no. 2 (1991): 138-46; Larry W. Emmelhainz, Margaret A. Emmelhainz, and James R. Stock, "Logistics Implications of Retail Stockouts," Journal of Business Logistics 12, no. 2 (1991): 129-41; Paul H. Zinser and Jack A. Lesser, "An Empirical Evaluation of the Role of Stock-Out on Shopper Patronage Processes," Advances in Consumer Research 8 (1981): 221-24; Phillip B. Schary and Boris W. Becker, "The Impact of Stock-Out on Market Share: Temporal Effects," Journal of Business Logistics 1, no. 1 (1980): 31-43; Philip B. Schary and Martin Christopher, "The Anatomy of a Stock-Out," Journal of Retailing 55, no. 2 (1979): 59-70; and C. K. Walter and John R. Grabner, "Stockout Cost Models: Empirical Tests in a Retail Situation," Journal ofMarketing 39 (July 1975): 56-68.

"Larry W. Emmelhainz, Margaret A. Emmelhainz and James R. Stock, "Logistics Implications of Retail Stockouts," Journal of Business Logistics 12, no. 2 (1991): 129-41.

"Phillip B. Schary and Boris W. Becker, "The Impact of Stock-Out on Market Share: Temporal Effects," Journal of Business Logistics, 1, no. 1 (1980): 31-43.

""Growing Problem of Stockouts Verified by Nielson Research," Progressive Grocer, (November 1968): S17 - S32.

"Federal Trade Commission, Trade Regulation Rule Including a Statement of Its Basis and Purpose: Retail Food Store Advertising and Marketing Practices, July 12, 1971.

"J. Barry Mason and J.B. Wilkinson, "Unavailability and Mispricing: Are Discount Stores Also Offenders?" Journal of Consumer Affairs 15, no. 2 (1981): 325-39.

16 J. Patrick Kelly, Hugh M. Cannon and H. Keith Hunt, "Customer Responses to Rainchecks," Journal of Retailing 67, no. 2 (1991): 122-37.

"Same reference as note 15.

11J. Patrick Kelly, Scott M. Smith and H. Keith Hunt, "Fulfillment of Planned and Unplanned Purchases of Sale and Regular Price Items: A Benchmark Study," a working paper, Wayne State University, 1998, pp. 1-24.

19"Out-of-Stocks.. the Details," Supermarket Business 51, no. 5 (May 1996): 33-41.

"Same references as notes 13, 14, 15, 16, and 19. Also see Thomas A. Foster, "Editorial: Just How Good is Logistics in the Retail Industry?" Distribution (October 1994): 4.

"Kurt Salmon Associates, Inc., "Enhancing Consumer Value in the Grocery Industry," a report prepared for the ECR Working Group January 1993.

by

John C. Taylor

Wayne State University

and

Stanley E. Fawcett

Brigham Young University

ABOUT THE AUTHORS

John C. Taylor is Assistant Professor of Logistics and Transportation, Wayne State University, Detroit, Michigan. His research interests are international logistics, transportation policy, and logistics performance and effectiveness. His research has appeared in Transportation Journal, International Journal of Logistics Management, International Journal of Purchasing and Materials Management, Regulation, and International Journal of Physical Distribution and Logistics Management.

Stanley E. Fawcett is the Donald L. Staheli Professor of Global Supply Chain Management, Marriott School, Brigham Young University, Provo, UT. An active researcher, he has published more than 75 articles in the areas of global supply chain management, comparative manufacturing systems, strategic purchasing, performance measurement, and global network design and management. His work has appeared in Journal of Business Logistics, Transportation Journal, International Journal of Logistics Management, and International Journal of Physical Distribution and Logistics Management.

Copyright Council of Logistics Management 2001
Provided by ProQuest Information and Learning Company. All rights Reserved