Printer Friendly

Time and product variety competition in the book distribution industry.

EXECUTIVE SUMMARY

Product variety and time have emerged as prominent dimensions in global competition. Marketers have emphasized the strategic advantages accruing from offering increased variety; on the other hand, operations managers have stressed reduced operations cycle times and faster response times to customer needs. Although substantial documentation supports the impact of such activities on costs, the only empirical evidence concerning actual rewards in the marketplace from such strategic endeavors is anecdotal.

This study of the book distribution industry demonstrates a two-phase methodology for valuing the rewards from faster response and improved variety. First an industry analysis was conducted to determine at a macro level how time and variety have been treated strategically and how it has affected the industry. The conclusion was that the very existence of the industry depends on the ability of these firms to offer a greater variety of titles shipped in smaller quantities in a relatively rapid fashion. Within the industry, those who can do this best have increased their market share substantially.

To quantify the potential rewards accruing to time-based and variety-based strategies, conjoint (or trade-off) analyses were conducted via telephone to estimate the specific trade-offs that retailers were willing to make among price, delivery time, and product variety. By understanding the relative value of different services or attributes to the consumer (in this case the book retailer), the impact of changing these services by either the firm or one of its competitors can be evaluated. For example, in our study, to overcome retailer's loyalty to a competitor, distributors would need to deliver one day faster than the average and fill about 3% more of the retailer's title needs. Similarly, one can determine what order fill rate increase or delivery time reduction would be necessary to offset a specific price increase and not lose customers.

In conclusion, the empirical research confirms the importance of time and variety as critical strategic dimensions. The emergence of a distributor that distinguishes itself from the competition along one or both these dimensions can alter the structure of the industry. Consequently, managers need to be cognizant of the relative values of time and variety in configuring their distribution strategy.

INTRODUCTION

A number of academics and practitioners have detected a recent, but significant shift in strategy by leading global competitors: firms that once profitably focused on providing products with the most value for lowest cost were now placing increasing emphasis on providing the quickest response to customer demands (Blackburn (1991); Stalk (1988); Stalk and Hout (1990)). These time-based competitors pursue a market-driven strategy that places a premium on response time and product variety in building a sustainable advantage.

In this study we examine the strategic power of time and prOduct variety in distribution, a critical function for the time-based competitor. Our approach is to measure, quantitatively, how the customers of wholesale book distributors--retail bookstores--make trade-offs among price, response time and product variety. The results provide an economic foundation for a time-based and variety-based strategy.

STRATEGIC CONFLICT BETWEEN MARKETING AND OPERATIONS

A firm's emphasis on time and variety can create strife between operations and marketing. Marketing managers, driven by a sales objective, historically have advocated that the firm should offer the customer virtually infinite variety and instant response. On the other hand, operations managers, measured in terms of output and cost, have lobbied for variety-minimizing strategies anchored in long production runs and low cost. Until recently, this conflict usually went unresolved, smoldering beneath the surface while corporate performance suffered. The development of JIT systems and "lean production" has changed all that. These systems provide operations with the flexibility to be responsive and offer product variety without incurring the magnitude of cost that traditionally accompanies such capabilities.

However, in the subject of this study, distribution, faster response and increased variety(1) usually mean increased operating cost. Facing increased competition along these dimensions, firms must make explicit trade-offs in formulating their marketing strategy and in executing the operations plan to support it. This study provides insight into how distributors should deal with those issues.

RESEARCH ON TIME AND VARIETY AS COMPETITIVE DIMENSIONS

Almost all studies of the impact of time-based competition have been conducted by researchers in operations management and have focused on the effects on costs. A number of studies have documented the kinds of savings one can realize whether they be logistics costs (Moskal (1984)), changeover costs (Frazier et al. (1988)), or shipping costs (Ansari and Heckel (1987); Blumenfeld et al., (1987)). In addition, various studies have pointed out the advantages accruing particularly to those buying from Just-In-Time (JIT) suppliers. These include: reduction in inventory (Ansari and Modarress (1986)) and inventory amplification (Rosenberg and Campbell (1985)), minimizing speculative finished goods (Greenberg (1988)), and increased quality (Hahn et al. (1986); Schonberger and Ansari (1984)). Speed of delivery also allows some firms to handle an increase in variety of product offerings, e.g., in perishable goods or through incremental innovation (Gomory and Schmitt (1988)). Finally, research has been devoted to demonstrating advantages of being first to market with new products (e.g., Lieberman and Montgomery (1988); Wernerfelt and Karnani (1987)).

The impact of product variety in determining competitive advantage has been addressed primarily by researchers in marketing. A number of models examine the question: What is the optimal product variety that a producer should offer (examples include Bental and Spiegel (1984); Green and Krieger (1987); McBride and Zufryden (1988)) have been developed. In addition, there is considerable literature on variety-seeking behavior by consumers. These studies have demonstrated how such behavior varies by demographic group (Hager (1988)), product class (Handelsman (1983)), variety drive of individuals (Hoyer and Ridgway (1984)), and how variety-seeking behavior changes over time (Handelsman and Munson (1985)). At the same time, some have argued that the effects of variety will vary by market conditions (Shugan (1989); yon Ghyczy (1986)).

In fact, there have been few studies that have examined the demand side impact of time and variety. The evidence supporting a strategic focus on time and variety is mostly anecdotal or drawn from a theoretical model of consumer behavior. With respect to time, it has been argued that customers are willing to pay a premium to the firm that can minimize the time between customer need and customer satisfaction (see Davis (1987)). Therefore, those firms that can significantly reduce the cycle times to carry out various functions of the firm, such as new product development, manufacturing, or customer service, will gain a competitive advantage and reap appropriate rewards. Others have noted the value of product diversity (e.g., Johansson and Nonaka (1987); McKenna (1988)) as a strategic weapon. In response, others have argued that consumers are more interested in quality and low price (Schonberger (1987)). This controversy over the value of time and variety was the genesis for this study.

METHODOLOGY

This study addresses the issue of how the market values response time and product variety in the distribution industry.(2) Specifically, the study focuses on book distribution because initial field studies indicated that time and variety were becoming dominant competitive dimensions in this industry segment. Distribution also brings into focus critical issues of strategy conflict between marketing and operations. The key issues to be investigated were determined through extensive interviews with key industry executives, supplemented by an analysis of existing data from previous research. In addition, conjoint, or trade-off, analysis was performed on industry survey data collected to quantify the measurable results of efforts to increase response time or add variety to the product line.

EMPIRICAL ANALYSIS: AN INDUSTRY STUDY OF THE WHOLESALE BOOK DISTRIBUTION INDUSTRY

Our study of the wholesale book distribution industry led us to conclude that both time and variety have been crucial strategic dimensions in the birth and growth of the industry and will continue to figure prominently in its future. It is precisely because of the value placed on these dimensions (and others) by the book retailers that the industry has performed so well.

Three key players are involved in the delivery of the books from authors to consumers: publishers, wholesalers, and retailers. These entities constitute a value-delivery chain connecting the author and the reader. Authors produce the intellectual property which publishers convert to book form. Books are then distributed to retailers either through wholesale distributors or directly by publishers. Finally these products are purchased from the retailers by individual consumers or institutions such as libraries or schools. Each of these has contributed to the current strategic situation facing the industry.

Publishers

Publishers have contributed to the ascendance of both time and variety as competitive dimensions in the industry. On the supply side, publishers have fueled variety-based competition by creating a proliferation of product. The number of titles in print has grown from 163,000 in 1963 to over 751,000 in 1987.(3) There are now over 50,000 new titles produced each year. This increase is due to several factors.

* Growth of publishers. One of the contributing factors is the explosive growth in the number of publishers in recent years, from just over 2,000 in 1982 to over 29,000 in 1988.(4) Growth can be attributed to a number of factors, but our research indicates that the most influential factor has been technology: it is simply much easier now for budding publishers to publish books at a reasonable cost due to the advent of desktop publishing. This has also contributed to the growth of small publishers. There are approximately 2,000 firms who publish at least four titles each year who are responsible for more than 90% of the industry sales volume.(5)

* Competitive response to growth and new entrants. As a result of such growth, most publishers have been very profitable in recent years, with average profit margins close to 20%, return on sales of averaging just under 10%, and price earnings ratios ranging from 20 to 30. Perhaps because of this healthy growth in sales and earnings, a number of foreign firms have been attracted to the industry and have either acquired or invested in United States book publishers. With these competitive pressures, there is an increasing focus on finding new competitive dimensions by which firms can position themselves not only to insulate themselves from competition but also to be in a position to take advantage of anticipated growth.

* Variety of Product Segments. The major product categories include trade, mass market paperback, religious, professional, book clubs, university presses, and elementary/high school/college texts. These segments create opportunities for publishers to specialize and focus production and marketing efforts on different niches.

At the same time that the industry has seen an explosion in product variety, publishers have failed to rise to the challenge of improving their distribution of product. Typically, publishers are very skilled at production and initial sales of their products, but they are not adept at on-going customer service. As one retailer stated: "Publishers do not make good distributors." Publishers typically deliver large drop shipments of newly-minted books to retailers very quickly, and, with 50,000 new titles every year, they spend most of their resources on distributing the initial printing. And because they ship to the retailer at the same time as they ship to the wholesaler, publishers can provide new titles faster than the wholesale distributors. But it is the continual inability of publishers to offer speedy delivery on follow-up orders that has contributed to the growth of the wholesale distribution industry.

Retailers

At the other end of the product delivery chain, growth in the retailer segment has also contributed to the increasing importance of time and variety. Along with the growth in the number of publishers, growth in the number of retail establishments selling books has been equally strong--from 5,689 in 1972 to 12,675 in 1984 to over 20,000 in 1989.

Among these establishments, there are four main types of retailers: national chains, regional chains, independents, and college bookstores. Each possesses significant strategic differences as they have tended to pursue different generic strategies. Figure 1 portrays the fundamental differences, mapped in terms of the dimensions of time and variety. The figure indicates how different firms have carved out turf in time-variety space.

With over 800,000 books in print, for a bookstore's stock to be representative and attractive to customers, it must have more titles than in the past. Although space and sales have increased over time, these have not increased nearly as fast as the number of titles. Essentially, each store is carrying a lower inventory per title, so that more frequent and quicker restocking is necessary. Also, modern communications and advertising have made the consumer more quickly aware of new titles. This has made books much more of a fashion item, i.e., where demand ramps up rather quickly.

When ordering books retailers must first decide whether to purchase from a publisher or a wholesaler. In selecting a vendor, retailers generally are driven by several factors including (in order of importance) higher discounts, free freight, unlimited returns, and early notice of new titles.(6) Ease of ordering, multiple selection and terms of trade are also key determinants. Retailers prefer "one-stop shopping," that is, they prefer to place their entire order with a single supplier. Wholesale distributors have a clear advantage over publishers because they have the opportunity, with sufficient stock, to be a single-source supplier.

Clearly, technology has played an important role in the retail ordering process, and it has been a significant factor in thrusting time and variety into the competitive forefront. Electronic ordering which reduces the shipping time delays is becoming more and more prevalent, and this has substantially changed ordering patterns. Our research indicates that approximately 40% of all orders are still placed orally over the phone, with an average order size of 55 titles. In contrast, electronic orders have an average order size of 125 titles. This obviously has created powerful incentives for distributors and publishers to be more technologically progressive in order to garner larger orders.

To strengthen the link with retailers and to create greater incentives for single-source ordering, distributors often supply retailers with microfiches of their inventory, usually updated monthly. In addition, major distributors will sell systems to customers which include CD-ROM with hardware and software. Ingram Book Company, for example, has over 300 such systems installed with retailers. The American Publishers Association (APA) developed an electronic ordering system known as PUBNET that is fairly prevalent in the college bookstore market. The APA is now trying to make inroads into the trade bookstores, but has experienced limited success, due primarily to different ordering patterns related to discounts. Also, in 1989, at least one major publisher began equipping its sales force with laptop computers for taking orders.

It is anticipated that technology will continue to cause radical changes in the industry. In a recent survey of booksellers, for example, the impact of new technology headed the list of probable trends that will affect the industry. Specifically, a significant number of booksellers expressed the belief that in the future, 80% of books will be ordered by electronic means and there also will be a great deal of ordering by telephone.(7)

The president of the National Association of College Bookstores states that in the near future he expected that customers would soon purchase by phone and the book would be delivered (by the wholesaler) directly to the customer.(8) Thus, the distributors would be only electronic middlemen. In a sense, this would move the industry to the ultimate in customization and variety, dramatically crunching time in the delivery chain.

Wholesale Distributors

Originally, the distribution industry began by servicing libraries which could not be adequately serviced by publishers because they wanted only one or two copies of any given title. Thus, the industry arose, not out of a specific need for either speed of delivery or variety, but to fill the needs of these institutions for small orders. However, while distributors have maintained a large proportion of library sales, they have made steady inroads into other markets as well. Today, there are approximately 40 distributors of significant size, with the two largest being Ingrain Distributors and Baker & Taylor.

Wholesale distributors have made substantial penetration in the total market (approximately 35%), yet this varies substantially by product segment. In fact, part of the success of wholesalers has been due to their focus on particular products delivered. They have only 15% of the trade books business, for example, but almost 70% of mass market paperback sales, and these percentages have remained fairly stable over the past decade.

Wholesalers have generally focused on lower ticket items, such as paperbacks, and thus when we compare share in dollars versus units, we see considerable differences. In particular, while wholesalers hold a 35% share in units sold, they have only a 20% share in dollars sold, i.e., wholesalers are selling 35% of the units but are receiving only 20% of the dollar revenue.

In addition, there is considerable variation in sales through distributors to the different types of retailers. In general, those retail categories with more of a focused, lower variety, lower cost strategy (particularly the chains) are more concerned about higher discounts. As discounts by wholesalers are no larger than 43%, these retailers typically choose publishers over wholesalers in order to take advantage of the higher discounts they offer.

However, we conclude that the wholesalers' gaining share was due as much as to the lack of the publishers' service as to strategic activity of the wholesalers. Generally speaking, most retailers are considerably less satisfied with the level of customer service provided by publishers as compared to that offered by wholesalers. Figure 3 indicates retailers' marked dissatisfaction with publishers' service compared with that of wholesalers. Once again, we can note variation across retailer type. This lack of satisfaction has created an opportunity for wholesale distributors to maintain and, in some cases, increase their share of the retailers' business.

In addition, the respective success of the different distributors in each of these market segments has differed considerably. Figures 4a-4c show, for the retailers sampled, the percentage of stores using a given distributor, and then, for those using that distributor, the percentage of business the retailer gives that wholesaler.

As shown in the preceding figures, Ingram was used by almost all the retail stores sampled and received approximately 56% of this business. Second in popularity was Baker & Taylor, which was used by 55% of retailers but had only a 28% share of the market.

It is also interesting how this changes for the non-independents, especially for the distributors other than Ingram. Among college bookstores sampled, for example, Ingram increased its share, while others shrunk to nothing. Also, some of the regional distributors have a very low share among the chain stores. We conclude that this is due to the title selection available and to higher discounts available from these wholesalers with national scope.

This disparity is always due to a number of structural and strategic factors. But we would argue that it is significantly affected by the strategies they have chosen to pursue and their ability to implement them. In fact, there are a number of strategic operating choices around which the industry participants are arrayed: number of titles carried, number and location of warehouses, ordering procedures, shipping policies, discounts, return policies, merchandise carried, and freight, among others. As well, there is variation around the quality of their operations and level of service as perceived by the retailers. Table 1 presents a recent ranking of distributors by independent retailers.

As can be seen, some distributors rank high for fast delivery and relatively low for variety (inventory selection and fill rate). For example, Koen is rated highest for speed and very low for variety (5-6). Ingram is rated highest for variety and somewhat lower for speed (3). It is not clear whether speed and variety are strategic trade-offs consciously made by distributors or whether they are an issue of quality of implementation. Furthermore, these data do not tell us the relative value of these purchase aspects to the retailers. Given the fact that nothing is free, and given a choice, how would retailers make trade-offs with respect to these different aspects?

TABULAR DATA OMITTED

REWARDS TO TIME AND VARIETY: EVIDENCE FROM CONJOINT ANALYSIS

To address this question about how retailers make trade-offs among delivery time, price and variety, we undertook a survey of retailers. The survey used conjoint analysis to determine more quantitatively the specific rewards one might expect from competitive actions undertaken by wholesalers vis-a-vis retailers. (For a complete description of this methodology, see Elrod (1990)).

Attributes Investigated

This study investigated five attributes of the services offered by book distributors:

* Discount. Distributors supply books for resale. The price of the book is expressed in terms of the percentage unit margin to the retailer reselling the book at list price. This is the "discount," which is typically between 40 and 43%.

* Speed of delivery. This is the time in days required for the shipment of books to arrive from the day the order was placed, It is the average time for the retailer's usual distributor, Same-day delivery is not uncommon, but some must wait as long as a week. As discussed earlier, the delivery time is largely a function of the location of the retailer relative to the distributor.

* Delivery reliability. This was measured as the percentage of the time that deliveries arrived on time. In some industries reliability in delivery is at least as important as speed.

* Number of titles carried in inventory. Retailers typically approach distributors for a few copies of each of many titles in an order. Distributors that stock a large number of titles can fill a greater portion of the retailer's order, which is convenient for the retailer (the "fill rate" discussed earlier). Respondents reported the percentage of titles in their typical order that their primary distributor carried in stock.

* Loyalty. This final attribute is qualitative. As is common in industrial marketing, long-standing relationships between retailers and their primary distributors are typical. Unless given a reason to switch, retailers prefer to remain with the same distributor. Again, our prior interviews showed this to be an important factor.

Survey Results

At the start of the interview, book buyers were asked to evaluate the level of service received from their primary distributor. The results are shown in Table 2.
TABLE 2
DESCRIPTIVE STATISTICS FOR CONJOINT ANALYSIS
 Discount Delivery Time Reliability Variety
Mean 40.7 3.4 95.4 84.9
St. Dev. 3.0 2.2 8.4 15.8


Thus, among those surveyed, the average discount is currently 40.7%. The average delivery time is 3.4 days from the day the order is placed, which would appear to leave room for improvement. Reliability in delivery, however, exceeds 95%. At present, the retailer's primary distributor can supply 85% of the titles requested in a typical order from existing stock.

The Importance of Time and Variety Improvements

The remainder of the interview consisted of a sequence of five questions to assess quantitatively the importance to retailers of improvements in service relative to the level of service they already receive from their primary distributor. This procedure is known as choice-based conjoint analysis (Louviere and Woodworth (1983)); it has been shown to yield results comparable to the more common ratings-based conjoint analysis (Elrod, Louviere and Davey (1991)). The choice-based methodology was adapted by Elrod (1990) to allow administration by telephone.

The questions asked respondents to choose between improvements in existing service. The importance of the four quantitative attributes was assessed by asking each interviewee three questions, such as:

Suppose your primary distributor could improve service in one of two ways, either by increasing its discount by one percent or by delivering one day faster. Which would you prefer?

Loyalty to the existing distributor was assessed by asking two questions such as:

Suppose another distributor promised service comparable to your current distributor in the ways we have talked about and in addition promised to deliver one day faster. Would you switch to the new distributor or would you remain with your current distributor?

The importance of the five attributes was assessed by estimating binary logit, which is a regression-like model for the analysis of a binary dependent variable. In this case, the dependent variable is whether the respondent chose the first of the two alternatives. The model allows for tests of statistical significance, as in ordinary regression analysis. The results of the estimation for the full model and for the final model (which contains only the statistically significant variables) are shown in Table 3.

TABULAR DATA OMITTED

The full model contains two coefficients with t-values with absolute value less than two and therefore statistically insignificant. The first of these is the intercept, which tests for a tendency on the part of respondents to pick the first or second of the two alternatives. The other statistically insignificant coefficient is reliability in delivery. This does not mean that reliability in delivery is unimportant. It does indicate, however, that given the near-perfect performance in reliability at present, further improvements are not as important to retailers as improvements on other attributes,

Special measures of goodness-of-fit are required for logit models, such as McFadden's pseudo-|R.sup.2~ (cf. Maddala (1983, p. 40)), which are shown in the table. These values may appear low by regression standards, but the role of four of the variables is highly significant, as indicated by the t-values.

Discussion of Results

The coefficients indicate how retailers value changes in four attributes relative to existing service. Examining the final model shows that all coefficients are of the expected sign: increases in discount and variety are positively valued, while an increase in delivery time is negatively valued. The positive coefficient for loyalty suggests that retailers favor maintaining their existing relationship with their primary distributor. The coefficients can be used to calculate an index of customer satisfaction (CS) subject to changes in existing service, given by:

CS = 0.65 (Discount) - 0.25(Delivery) + 0.10(Variety) + 0.65(Loyalty).

This CS index predicts the proportion of respondents that would choose a new distribution service relative to the existing service. That proportion (P) is given by the equation:

P = exp(CS) / |1 + exp(CS)~.

The coefficients are such that the predicted proportion P is as close as possible to the observed proportions for the interview data.

Of particular interest are those changes in service that leave retailers indifferent to their present condition. This index shows the relative effect of a change (relative to the current level of service) in these strategic factors on consumer satisfaction, both in terms of direction and magnitude. The positive coefficients for discount and variety and the negative one for delivery indicate an increase in the discount rate (i.e., in effect, a price decrease), a decrease in delivery time and an increase in variety of titles available would raise retailer satisfaction. The coefficients also show the magnitude of the effect of such a change from the current level of service.

The value of a specific change from the existing service can then be computed. For example, suppose a distributor is contemplating a one-day improvement in average delivery time. (Given that delivery is at present 3.4 days on average, this improvement appears to be feasible.) Such an improvement would increase the CS index by 0.25 x 1 = 0.25, given no other changes in service. If, however, the one-day improvement in delivery time were to be combined with a decrease in the average discount, then the CS might not increase.

At issue is the payoff of either reducing the delivery time or increasing the number of available titles. By using a discount rate variable or a loyalty variable, the model can estimate the payoffs in either increased prices to existing retailers or new business from other retailers.

Speed of Delivery

The payoff from existing retailers to delivery time changes is determined by the discount rate offsetting the decrease in delivery time so as to leave the CS index at zero (and therefore leave the average retailer as indifferent to the current situation). For example, a one-day reduction in delivery time (which raises the CS index by 0.25) could be offset by a decrease in the discount rate of 0.38% (0.65 x -0.38 = 0.25). This is the effective price increase retailers would be willing to pay for this reduction in delivery time to maintain the same level of satisfaction. Because this relationship is linear, we can graph the relationship for an average distributor as shown in Figure 5.

The graphed function in Figure 5 indicates the discount rate appropriate for a given level of service. For example, if a wholesaler has an average delivery time of five days, it would need to offer a discount rate of 41.28%. If priced above this, sales would diminish.

The loyalty coefficient indicates how a competitor must perform relative to the existing distributor to have an even chance of winning away the retailer. The competitor must overcome the loyalty disadvantage, which corresponds to a CS index of -0.65. Thus, to have the opportunity to gain new retailers, a competitor must reduce delivery time by at least 2.6 days (0.25 x 2.6 = 0.65), ceteris paribus.

Variety

The payoff for an increase in variety (here an increase in titles available) can be determined similarly An increase in inventory sufficient to fill one percent more titles from stock (fill rate) would increase the CS index by 0.10.(9) This would be advisable if the distributor could do so while passing along a decreased discount rate of less than 0.10/0.65 = 0.16 percent. Figure 6 depicts the linear relationship between the discount and fill rates.

Figure 6 shows the appropriate pricing scheme for a given level of service for a distributor with an average level of service. A competitor would need to increase the fill rate by 6.5% (since 6.5 x 0.10 = 0.65) in order to offset the loyalty disadvantage of 0.65. Therefore, one should not expect to gain new retailers from any increase in the fill rate less than this amount.

Combinations

The above discussion has isolated each of the issues of delivery time and title selection. However, combinations of improvements may be equally effective in overcoming the loyalty disadvantage or a price increase. For instance, delivering one day faster and filling 3.1% more of a retailer's title needs also increases the CS index by 0.65, a value that exactly offsets the loyalty advantage of the retailer's current wholesaler. In general, any differences in service can be evaluated by the CS equation.

Applications

How might this model be used practically by managers? Suppose a firm wants to pursue major market share increases. It can construct a combination strategy based on its own capabilities that would be sufficient to dislodge retailers loyal to a competitor. For example, the distributor could increase the discount rate by 0.50%, reduce delivery by half a day, and add inventory to increase the fill rate by slightly over one percent. Another use of the model is in evaluating competitive moves and potential responses. For example, suppose a competitor can reduce delivery time to a certain area by a given amount. The model projects that if a competitor can deliver to our customers 2.6 days faster, the competitor would be an equally attractive supplier to our customers. Such a move could be countered in several ways--by increasing the fill rate by 6.5% or by increasing our discount rate by 1.0%.

Similarly, a manager could evaluate the effect of a competitor offering to pay the freight for shipments to retailers. For any given retailer, the value of such a move compared to a change in discount rate can be evaluated and then compared to the current level of service (time and/or variety) offered by the wholesaler to determine its ability to overcome the loyalty disadvantage.

IMPLICATIONS OF CONJOINT ANALYSIS FOR OPERATIONS STRATEGY

The results of the conjoint analysis suggest that book distributors are under intense competitive pressure along several dimensions. Industry standards for delivery times and reliability of delivery are already quite high, and, to serve customers better, retailers are pushing distributors to offer more variety--that is, carry more titles. Carrying additional titles is a key to market dominance because retailers, when placing replenishment book orders, prefer "one-stop shopping"--ordering all of their books from a single wholesaler. Other things being equal, the wholesaler with the most titles in stock ("the most variety") is most likely to get the order--to become the retailer's first-choice supplier.

Distributors are under increased competitive pressure along two dimensions:

1) to move closer to the customer in terms of delivery time;

2) to offer greater variety by carrying more titles in inventory.

The distributors who thrive in this environment are the time-based competitors who have the lean, flexible distribution systems to outperform the competition. The strategy of one centralized distribution center for the entire U.S. is no longer viable because air freight is cost prohibitive and surface delivery is time prohibitive (retailers are unwilling to wait five to seven days for delivery).

To improve delivery speed, the leading distributors are winning market share by shrinking time to the customer in two ways. One, wholesale distributors are investing in smaller, more flexible regional distribution centers (DCs) and moving geographically closer to their customer base. One major distributor we interviewed stated that their strategy is to have DCs located within an overnight delivery zone of 95% of the book buyers in the U.S. This means six to seven regional distribution centers to cover the country. With such a decentralized warehousing/distribution system, a firm must exercise even tighter control over their operations--inventories, information, shipments, etc.

In addition to moving physically closer to the customer, leading distributors are developing logistics capabilities to shrink the delivery time within the zone of a DC. Firms are developing "lean order-entry processes" to move retailers' orders to the picking line as quickly as possible. In order entry, our research indicates that the "best-in-class" distributors are found in computer software distribution. Orders phoned in during the evening to these distributors can be processed and picked within two hours and delivered by air the next day. Faster shipping to customers is also achieved by supplementing standard United Parcel Service (UPS) deliveries with private carriers and UPS zone-skippers.

Unless operating changes are made, increased product variety translates into increased operating cost. Inventory costs and order-picking costs increase as titles are added to the product line. In book distribution, leading distributors have worked assiduously to shrink the time to pick, load and deliver orders. As found in other studies we have conducted on time-based competition, reducing time in a warehousing operation does not automatically mean increased cost. Speed can be achieved through elimination of non-value adding activities and developing simpler processes that can mean even lower costs. Faster inventory turnover through better picking systems and electronic linkages with retailers means faster asset turnover. Astute distributors are investing some of the financial benefits of faster turnover in increased variety--that is, broadening their product line.

CONCLUSIONS

Our investigation of the book distribution industry confirmed the strategic value of response time and variety. Supply and demand side forces have stimulated the increase in product variety and the speed at which they are delivered. Clearly, technology, such as desktop publishing, has played a major role in increasing variety; and electronic ordering and data interchange have reduced the time required to process and deliver orders.

Retailers have responded to the market demand for more titles and this pressure is transferred to the wholesale distributors. To compete with large national chains, independent booksellers were forced to emphasize service and variety to differentiate themselves and maintain market share. Given their limited space and financial limitations, independents needed to have a multitude of titles available on short notice. At the same time, publishers were negligent in meeting their restocking needs. Consequently, retailers turned increasingly to wholesale distributors as intermediaries: inventory holders and distribution channels for product.

The book wholesalers, who had originally serviced mostly libraries who needed small order quantities and did not need them particularly fast, were then faced with a tremendous opportunity to increase sales. Those who were able to respond to the changing marketplace prospered. Specifically, this meant offering increased variety of product and reduced delivery times. Twenty-four hour turnaround of orders became almost a necessity. To solve the resulting logistical problems, new delivery and ordering methods were explored. Warehousing became a critical issue, and the costs associated with it created new barriers for expansion to existing players. In effect, the dimension of time created new boundaries for markets, shrinking or expanding them according to the technological and logistical capabilities of competitors.

One important aspect of this our research uncovered is exactly in what segments of the business the wholesale distributors have made their inroads. Specifically, they have found a receptive market among independents who focused on variety as a means of differentiation. Also, in terms of product, they have a much stronger component of lower priced items. Finally, with the increased ease and speed of ordering, average order size has decreased. However, with increased variety, and thus availability, they are able to get a greater number of orders from competitors.

Distribution is an industry in which a significant portion of the product delivery chain has been overtaken by new players competing on speed and variety. These players achieved operational efficiencies enabling them to make a sizeable mark in the book industry. However, they have been able to make more substantial inroads among certain types of competitors and with certain product segments. Among wholesalers, those who have been able to do so most effectively have gained a larger share of their specific segments. However, their ability to make further gains in other segments, especially in light of renewed competitive efforts by publishers, is still in question. Nevertheless, the time and variety-based competition has unalterably changed the structure of the industry.

Finally, our conjoint analysis confirms that indeed there are specific trade-offs retailers are willing to make for increased variety and reduced speed of delivery. We have applied a relatively simple yet robust methodology for mapping customer trade-offs so strategic decisions can be made with respect to time and variety in any given industry. Managers can use the results of such analysis to better understand appropriate prices for changing services, or likely effects of competitive actions, and better perform cost-benefit analysis of key decisions.

There are many such industries where new technologies will provide opportunities for new players to enter and capture business due to their ability to provide greater product variety in less time, more efficiently. Certainly there are ample opportunities in distribution in other industries, and we predict the evolution of such systems. We also have seen how the rising importance of these dimensions has redefined markets altogether. Successful companies recognize this and carve out markets and niches to suit their own particular strengths.

ENDNOTES

1. Variety has many connotations. We define variety here as number of product offerings. We recognize that the degree of actual variation may be debatable, that is, one could argue that, although the number of sku's could be high, for example, there is not a good deal of real difference in terms of style, etc.

2. Industry studies were also conducted in machine tools and plastic injection molding. Reports on these industry studies are available from the authors.

3. Center for Book Research, Book Industry Trends 1988, p. 11.

4. Center for Book Research, Book Industry Trends 1988, p. 11. 5. United States Department of Commerce, U.S. Industrial Outlook, 1989, Washington, D.C.: U,S. Department of Commerce, p. 37-7.

6. Anthony et al. Book Retailer Survey 1989.

7. Cetron, Marvin, and Rebecca Lucken "How Booksellers See the Future," American Bookseller, July 1989, pp. 52-54.

8. Personal interview.

9. This study is concerned with estimating the demand implications of improvements in time and variety, so we do not estimate the costs associated with faster delivery time or increased inventory. Distributors are in a better position to estimate the incremental cost in their own operations. What they have heretofore not been able to do is predict customer response to such improvements.

ACKNOWLEDGMENTS

The authors gratefully acknowledge partial funding for this research by the Boston Consulting Group and the Operations Roundtable of the Owen Graduate School of Management, Vanderbilt University.

REFERENCES

Ansari, A., and J. Heckel. "JIT Purchasing: Impact of Freight and Inventory Costs." Journal of Purchasing and Materials Management, vol. 23, no. 2, Summer 1987, 24-28.

Ansari, A., and B. Modarress. "Just-in-Time Purchasing: Problems and Solutions." Journal of Purchasing and Materials Management, vol. 22, no. 2, Summer 1986, 11-15.

Anthony, M., J. Brody, D. Ingram, W. Robinson, T. Simpson, and B. Sitton. "Book Retailer Survey, 1989." Unpublished market research study, Owen Graduate School of Management, Vanderbilt University, 1989.

Bental, B., and M. Spiegel. "Horizontal Product Differentiation, Prices, and Quantity Selection of a Multi-Product Monopolist." International Journal of Industrial Organization, vol. 2, no. 2, June 1984, 99-104.

Blackburn, J.D. (ed.) Time-Based Competition: The Next Battleground in American Manufacturing. Homewood, IL: BusinessOne Irwin, 1991.

Blumenfeld, D.E., L.D. Burns, C.F. Daganzon, M.C. Frick, and R.W. Hall. "Reducing Logistics Costs at General Motors." Interfaces, vol. 17, no. 1, January-February 1987, 26-47.

Center for Book Research. Book Industry Trends, 1988. Scranton, Pa.: Center for Book Research, 1988.

Cetron, Marvin, and Rebecca Lucken. "How Booksellers See The Future." American Bookseller, July 1989, pp. 52-54.

Davis, S. Future Perfect. Reading, MA: Addison-Wesley, 1987.

Elrod, T. "Conjoint Analysis by Telephone." Unpublished manuscript, University of Alberta, Edmonton, Alberta, Canada, 1990.

Elrod, T., J.J. Louviere, and K.S. Davey. "An Empirical Comparison of Ratings-Based and Choice-Based Conjoint Models." Journal Of Marketing Research, vol. 29, August 1992.

Frazier, G.L., R.E. Spekman, and C.R. O'Neal. "Just-in-Time Exchange Relationships in Industrial Markets." Journal of Marketing, vol. 52, October 1988, 52-67.

Gomory, R.E., and R.W. Schmitt. "Step-by-Step Innovation." Across the Board, vol. 25, no. 11, November 1988, 52-56.

Green, P.E., and A.M. Krieger. "Consumer-Based Approach to Designing Product Line Extensions." Journal of Product Innovation Management, vol. 4, 1987, 21-32.

Greenberg, B.A. "Implications of 'Quick Response' on Commercial Vendor Patronage." Developments in Marketing Science, vol. XI, 1988, 486.

Hager, C.J. "Shopping for Variety in Red Meat, Poultry, and Fish." Advances in Consumer Research, vol. 15, 1988, 19-21.

Hahn, C.K., K.H. Kim, and J.S. Kim. "Costs of Competition: Implications for Purchasing Strategy." Journal of Purchasing and Materials Management, vol. 22, no. 3, Fall 1986, 2-7.

Handelsman, M., and J.M. Munson. "On Integrating Consumer Needs for Variety with Retailer Assortment Decisions." Advances in Consumer Research, vol. 7, 1985, 108-112.

Hoyer, W.D., and N.M. Ridgway. "Variety Seeking As an Explanation for Exploratory Purchase Behavior: A Theoretical Model." Advances in Consumer Research, vol. 11, 1984, 114-119.

Johansson, J.K., and I. Nonaka. "Market Research the Japanese Way." Harvard Business Review, vol. 65, no. 3, May/June 1987, 16-22.

Lieberman, M.B., and D.E. Montgomery. "First-Mover Advantages." Strategic Management Journal (UK) (Special Issue), vol. 9, Summer 1988, 41-58.

Louviere, J.J, and G. Woodworth. "Design and Analysis of Simulated Consumer Choice or Allocation Experiments: An Approach Based on Aggregate Data." Journal of Marketing Research, vol. 20, November 1983, 350-367.

Maddala, G.S. Limited-Dependent and Qualitative Variables in Econometrics. New York: Cambridge, 1983.

McBride, R.D., and F.S. Zufryden. "An Integer Programming Approach to the Optimal Product Line Selection Problem." Marketing Science, vol. 7, no. 2, Spring 1988, 126-140.

McKenna, R. "Marketing in an Age of Diversity." Harvard Business Review, vol. 66, no. 5, September-October 1988, 88-95.

Moskal, B.S. "Manufacturing: Just-in-Time; Putting the Squeeze on Suppliers." Industry Week, vol. 222, July 9, 1984, 59-60.

Rosenberg, L.J., and D.P. Campbell. "Just-in-Time Inventory Control: A Subset of Channel Management." Journal of the Academy of Market Science, vol. 13, no. 3, Summer 1985, 124-133.

Schonberger, R.J. "Frugal Manufacturing." Harvard Business Review, vol. 65, no. 5, September/October 1987, 95-100.

Schonberger, R. J., and A. Ansari. "'Just-in-Time' Purchasing Can Improve Quality." Journal of Purchasing and Materials Management, vol. 20, no. 1, Spring 1984, 2-7.

Shugan, S. "Product Assortment in a Triopoly." Management Science, vol. 35, no. 3, March 1989, 304-320.

Stalk, G., Jr. "Time--The Next Source of Competitive Advantage." Harvard Business Review, July-August 1988, 41-51.

Stalk, G., Jr., and T. Hout. Competing Against Time: How Time-Based Competition Is Reshaping Global Markets. New York: The Free Press, 1990.

United States Department of Commerce. U.S. Industrial Outlook, 1989. Washington, D.C.: U.S. Department of Commerce, 1989.

von Ghyczy, T. "Product Diversity and Proliferation As a New Mode of Competing in the Motor Car." International Journal of Technology Management, vol. 1, no. 3/4, 1986, 515-518.

Wernerfelt, B., and A. Karnani. "Competitive Strategy under Uncertainty." Strategic Management Journal (UK), vol. 8, no. 2, March/April 1987, 187-194.
COPYRIGHT 1991 American Production and Inventory Control Society Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Specail Issue on Linking Strategy Formulation in Marketing and Operations: Empirical Research
Author:Lindsley, William B.; Blackburn, Joseph D.; Elrod, Terry
Publication:Journal of Operations Management
Date:Jul 1, 1991
Words:7515
Previous Article:Manufacturing and marketing interdependence in the new venture firm: an empirical study.
Next Article:Factory focus: segmenting markets from an operations Perspective.
Topics:

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters