Stockout Management

Stockout Management

Mauro Sampaio

Abstract

Despite efforts to improve product availability, retailers have been struggling with considerable stockout for decades with little evidence of improvement. The magnitude of these losses depends on specific consumer responses, which have been found to vary with product, consumer, store and situational factors. Stockout has been, is and will continue to be an important managerial problem. The purpose of this paper is to investigate ways in which stockout losses can be mitigated for the retailers. Using the laboratory experiment date, we found that some recovery efforts had a large impact onthe consumer response to stockout situations. Finally, we discuss the theoretical and managerial implications of the findings.

Keywords: Stockout, Consumer Behavior, Logistics, Supply Chain Management

Introduction

Stockout is anordinary phenomenon for shoppers. Stockout levels of 5-10 percent in retail settings have proven to be the norm, rather than the exception. At the same time, consumer surveys have shown that consumers consider product availability to be a very important factor. The percentages of stockout occurrences regularly vary among five percent (The Netherlands), seven percent (France), and eight percent (United States and Brazil) of the total stock-keeping unit level of supermarkets (Andersen Consulting 1996; Kooistra 1999; ZinnLiu 2001; ECR Europe 2003 and Azevedo & Araujo 2004). These rates of stockout are significant and require the attention of manufacturers and retailers, because their causes are distributed along the whole length of the supply chain. The analysis by Gruen et al. (2002), which is a compilation of several global studies, shows that 70-75 percent of out-of-shelf situations are a direct result of retail store practices, with 47% of the cases attributed to wrong store ordering and forecasting and 25% to cases where the product was in the store but not on the shelf. The resulting gross margin losses for retailers due to stockout are estimated to lie between $7 and $12 billion per year in the United States (Coca-cola Research Council/Andresen Consulting, 1996).

The issue is not confined to traditional retail settings, as demonstrated by a 1987 consumer report study that found that mail-order customers reported stockout items as their most frequent problems. More recently, stockouts have become a major problem for on-line merchants, due to both traditional forecasting problems and poor links between their inventory system and their web site (Fitzsimons 2000).

In response, stores implemented management systems such as Efficient Consumer response (ECR), Quick Response (QR) and Collaborative Planning, Forecasting and Replenishment (CPFR), projects that could reduce the stockouts levels by 55 percent (Vergin & Barr, 1999). However, despite the potential (or theoretical) attractiveness of these projects, a substantial decrease of stockout levels has not yet been observed in practice. Vergin & Barr (1999) affirm that this happens because not every manufacturer is willing to assume the management of retail inventories unless retailers transfer to them an important part of the increase in profits.

Considering the tendency of increasingin the variety of new products and assuming that the shelf space is fixed in the short and mid-term, stockout occurrences are likely remain a regular phenomena for shoppers. Therefore, the retailers need to develop additional insights to mitigate the effect of a stockout in their stores. Until today, few academic efforts have been made to apply the theory of customers' recovery in a useful way to managerial decision making. In this paper we aim to fill this research gap, studying the relationship between different policies of customers' recovery in different purchase situations and evaluating their effect on the consumer response. Two central questions are posed in this study, as suggested by Fitzsimons, 2000. First: how best to manage the consumer when a stockout happens? Second, relativedissue is: are there customers' segments which are more or less sensitive to stockouts and are they identifiable? Stockout Management is a strategic subject, with the potential to promote competitive advantages tothose companies that are capable of averting the stockout and / or recovering consumers in an efficient way.

We continue this article with a review of the prior literature about Stockout. It provides major building blocks for the conceptual framework presented in the third section. The objective is to shed some light into areas where further research can be fruitfully undertaken. Next, we describe the research methodology and empirical results, and we end with a discussion of the managerial implications, research limitations, and directions for further research.

Literature Review

Stockout management is a theme of great relevance among academics and executives. The consumer response to stockout situations has implications for retail assortment, shelf space allotment, pricing, and logistics. A representative volume of studies has discussed this subject in publications, such as: Journal of Business Logistics, International Journal of Logistics Management, Journal of Marketing, and Journal of Operations Management among others.

Shelf management is a difficult task in which rules of thumb rather than good theory and hard evidence tend to guide practice. A company's efforts to set levels of product availability often get bogged down in debate over the cost of stocking out. The sometimes controversial nature of this cost and its hard-to-quantify components (such as loss of customer goodwill) make it a difficult number for people from different functions to agree upon (Chopra & Meindel, 2003). The cost of stockout must include a loss in the current and future sales margins. The stockout situation can generate a number of problems: a negative image of the brand and the store, errors in demand estimates, deficient stock policies, among others.

The great majority of stockout studies that have been published are fragmented and empirical in nature (Table 1). The purpose of this literature review is to identify the state of the art of these studies and the gaps of knowledge. Some of these publications will be briefly reviewed in the next section of this text.

Tab1: Overview of studies about consumer responses to Stockout

Authors / Focus / Nature of study / Stockout
Level / Consumer Response
Peckham
1963 / This study discusses the potential loss ofbusiness to manufactures and retail storesbrought about by out-of-stock at the store level / Survey / 7 a 10% /
Progressive Grocer
1968 / Thisstudy was a sequence of two papers documenting the frequency of stockouts observed for items sold in supermarkets / Survey / 12.2 % /
Walter and Grabner
1975 / In examining the reactions of liquor store customers in Ohio, this paper proposed a formal model that charted all possible responses to stockouts. / Survey / Not applicable /
Schary & Christopher
1979 / SDL response was compared to store image, brand loyalty, and demographic variables. Differences in consumer behavior were observed / Survey / 29.4% /
Emmelhainz et al.
1991 / Stocks were removed from the shelf of a discount grocery store. Consumers were interviewed at the checkout lane about intended SDL behaviors / Field
Experiment / Not applicable /
Andersen consulting
1996 / It is a survey on the problem of retail stockouts that combined store audits, scanner data, and personal interviews with industry and consumers / Survey / 8,2 % /
Verbeke et al.
1998 / This study Identifies consumer SDL reactions for high-selling brands and explainsSDL reactions by store-related and situational characteristics / Field
Experiment / Not applicable /
Campo et al.
2000 / This paper presents a conceptual framework that integrates the major determinant of consumer reactions to stockouts. / Survey / Not applicable /
Bell & Fitzsimons
2000 / This study presents a experiment to understand how consumers respond to stockout in terms of their satisfaction with the decision process / Laboratory
Experiment / Not applicable / Not applicable
Zinn & Liu
2001 / This research reported results of a study of consumer short-term response to stockouts. The majority of variables that are significant correlates of SDL behavior are situational / Survey / Not applicable /
GMA Study
2002 / It is a study on direct-store delivery in the USA / Survey / 7,4 % /
Gruen et al
2002 / It is a worldwide study of the extent, causes and consumer responses to stockout in the fast-moving consumer goods industry / Survey / 8,3 % /
Roland Berger
2003 / It is a ECR study on Optimal Shelf Availability that shows the importance of improving product availability. / Survey / 7 a 10% /
Sloot et al
2005 / This study investigates the effect of brand equity and the hedonic level of a product on Stockout responses. / Survey / Not applicable /

Category

The objective of early studies about consumer responses to Stockout was mainly to define and measure Stockout reactions and their financial impact. Peckham 1963 was the first author to discuss the potential loss of business to manufactures and retail stores brought about by out-of-stock at the store level. The main contribution of 1968 study developed by the National Association of Food Chains, A. C. Nielsen Company, & Progressive Grocer, was to identify the behavior of the consumer in response to a stockout, which may be:

• S - Switch to another product
• D- Delay the purchase until the next trip to the same store
• L- Leave the store, buying the missing item in a competing store.

Ever since, the acronym "SDL"- Substitute, Delay or Leave - is used to collectively describe consumer reaction to stockout. Another important contribution was the Walter & Grabner (1975) study that established the scheme for systematically classifying all possible consumer responses to stockouts, which influenced most SDL studies that followed.

Academic research has identified and categorized up to 15 possible consumer responses to stockout, though typically, managerial researchers measure five primary responses (Gruen, Corsten and Bharadwaj, 2003). These are:

1Buy item at another store (Leave)

2Buy later at the same store (Delay)

3Substitute – same brand for a different size or type (size switch)

4Substitute – different brand (brand switch)

5Do not purchase the item (lost sale)

Empirical studies reveal that "brand switching" is the predominant reaction, followed by size switching (Charlton & Ehrenberg 1976; Emmelhainz et al.1991). The magnitude of product substitutability has been estimated from 22 % to 83 %, depending on the specific retail environment (see tab.1). Store switching (31 %) and canceling/deferring (15%) the purchase are less often observed, like you can see in table 1, yet remain important as they may entail serious negative consequences for manufacturer and/or retailer.

A unique approach was developed by Charlton & Ehrenberg (1976). Instead of using surveys, they conducted an experiment. For 25 weeks, 158 consumers were visited at home and given the opportunity to purchase from a selection of three brands of detergent and of tea. The brands were especially created for the study. Stockouts were introduced during the study, and the reactions of consumers were measured. Consumers usually substituted the stockout brand but returned to it with the restoration of supply. This study did not consider the possibility of switching stores in response to the stockout. Finally, Charlton & Ehrenberg (1976) reported no long-term effects on sales.

Straughn (1991) was the first to use scanner data in a stockout study. She attempted to estimate the effects of stockouts on brand share for candy bars. The short-term effect was negligible. The long-term effect, defined as more than five weeks following the stockout condition, was substantial. Decline in brand share averaged 10 percent. The revenue loss not only stems from lost product sales during the stockout period, but also to later periods or other product categories

These empirical studies reveal that, the consumer reaction to stockout varies strongly from case to case. Table 1 reports the percentage of times consumers reacted to stockouts by each of the SDL behaviors. It is easy to notice that the results published of SDL behavior did not converge.

More recently, researchers began to build a theoretically based conceptual framework to explain consumer reactions to stockout. An improved understanding of consumer response to stockouts thus provides important managerial insight, and may help to determine the items for which stockouts should be avoided or ways in which stockout losses can be mitigated. Table 2 is a summary showing all significant relationships observed in several separate studies between the independent variables and SDL behavior. The signs in the table refer to the direction of the relationship. A positive sign indicates a direct relationship, while a negative sign indicates a reverse relationship. For example, the positive sign for Brand Loyalty and Leave indicates that the more Brand Loyalty someonehas,the more likely he or she willbe to buy the item at another store.

Tab.2. Summary of significant relationships (significance p < .05)

Description of
Characteristic / List of Variables / Switch / delay / Leave
Product
Variables related to the specific product category or brand in which the stock-out appears / Brand Loyalty (ZinnLiu 2001)
Brand Loyalty (Campo et al. 2000)
Private label (Campo et al. 2000)
Private label Merchandise (Schary &Christopher 1979)
Different Brand (Emmelhainz et al.1991)
Different variety (Emmelhainz et al.1991)
Different package size buyer (Campo et al. 2000)
Availability of acceptable alternatives (Campo et al. 2000)
High-equity brands (Sloot et al 2005)
Hedonic level of a product (Sloot et al 2005)
Stockpile products (Sloot et al 2005) / +
+
+
+
+
+
+ / -
+
+ / +
+
-
+
+
Store
Variables related to the store or retail chain in which the stock-out occurs / Perception of store prices (ZinnLiu 2001)
Perception of store prices (Verbeke et al.1998)
Store loyalty (Campo et al. 2000)
Part of the week (Sloot et al 2005) / +
+ / +
+
+ / -
+
Situation
Variables related to the specific shopping trip in which the stock-out appears
: / Urgency in purchase of item (ZinnLiu 2001)
Impulse purchases (Sloot et al 2005)
Specific time constraint (Campo et al. 2000)
Consumer is upset with stockout (ZinnLiu 2001)
Surprise with stockout (ZinnLiu 2001)
Pre-visit agenda (ZinnLiu 2001)
Pre-visit agenda (Campo et al. 2000)
Required purchase quantity (Campo et al. 2000)
Consumers purchasing small amounts (Verbeke et al.1998)
Major shopping trip (Campo et al.2000) / +
+
-
+
+
+ / -
+
-
+ / -
+
Consumer
Variables related to the consumer who is confronted with the stock-out / Occupational class (Schary &Christopher 1979)
General time constraint (Campo et al. 2000)
Shopping attitude (Campo et al. 2000) / + / +
+

Schary & Christopher's (1979) study was the first to attempt to explain Stockout reactions. In the early 1990s, Emmelhainz et al. (1991) continued to focus on explaining Stockout reactions. Their study reported that customers who were loyal to a store were more likely to delay purchase than non-loyal customers. The perceived risk of the product - "the risk of purchasing a brand other than the preferred brand" - has been shown to reduce brand switching, while the urgency to buy the brand had the opposite effect, that is, it increased the likelihood of consumers switching brands.

Verbeke et al.(1998) found thatretailers should keep in mind that abruptly taking a brand out of store is simply disastrous for their own retail chain because a large percentage of consumers are walking out of the store. This shows the power of the manufacturer in the supply chain for at least in the short term.

Bell & Fitzsimons (1999) reported that not only do consumers react negatively to the omission of preferred brands, but also that they experience dissatisfaction when previously available (but unchosen) brands are removed from the choice set.This result suggests that retailers should seek stability in in-store assortments. Fitzsimons (2000) has also demonstrated that consumer response to stockout is strongly influenced by his commitment to the alternative stockout and any change in decision difficulty caused by the stockout.

Campo et al. (2000) were the first to build a theoretically based conceptual framework to explain consumer reactions to stockout and revealed that consumers react differently depending on their shopping habits and on the specificities of the shopping occasion. In the same way, Zinn & Liu (2001) suggest that demographic variables don't have significant correlations with SDL behavior and that the majority of variables that have significant correlations are situational. A consumer buying something in a situation of urgency, such as something for Mothers Day, has a greater probability of substituting the item than a consumer who is making a planned purchase. Of the six variables included in the Zinn & Liu study (2001), four are situational: Urgency, Upset, Surprise, and Pre-Visit Agenda. The majority of consumer decision making occurs in the store (Hoch & Deighton 1989). Long standing surveys of supermarketing shopping behavior have found that only about 1/3 of purchases are specifically planned in advance of visiting the store (Dagnoli 1989). The apparent importance of situational variables has the potential to explain the SDL behavior divergence in previous studies.

Stockout has been, is and will continue to be a problem. Retailers live with gaps in their inventory because they believe that to fix it is more expensive than tolive with the problem. This is not the most profitable idea. Stock occupancy cost range from about $20/square foot for dry grocery shelf space to over $50/sp ft for dairy and $70/sq ft for frozen foods (Drèze,Hock, Purk 1994).Reducing stockout requires initiatives that cutacross functional boundaries which can require a fundamental rethinking of the retailer process.The most comprehensive study on the extent of retail stockouts is the one by Gruen, Corsten Bharadwaj (2003). They looked at several worldwide studies of more than 71,000 consumers that were conducted across a variety of categories. These authors show that stockout rates vary across retailers but the majority tends to fall in the range of 5% to 10%.

For retailers, brand and size switching do not represent a central problem. First, they need to focus on stockout situations in which consumers do not buy a substitute. Clearly, the retailers cannot sell something that is not in the store. Many consumers were not willing to switch brands when their preferred brand was out of stock: They either switched stores (31%), postponed the purchase (15%) or do not purchasethe item (9%) (Gruen, Corsten & Bharadwaj, 2003). In general, retailers can lose nearly half of intended purchases when customers encounter stockouts. When shoppers permanently switch stores, and either the new preferred store has overall lower stockout levels, or it has lower stockout levels on items of greatest value to the consumer, the result will be even worse.The store with a lower overall stockout level and effective strategies to manage dissatisfaction due to stockout will lose fewer customer and gain more customers from other stores. Those abandoned purchases resultin sales losses of about 4% for a typical retailer. For a billion-dollar retailer, that could mean $40 million a year in lost sales(Corsten & Gruen, 2004).