Interactive Home Shopping: Consumer, Retailer, And

Interactive Home Shopping: Consumer, Retailer, and Manufacturer Incentives to Participate in Electronic Marketplaces Author(s): Joseph Alba, John Lynch, Barton Weitz, Chris Janiszewski, Richard Lutz, Alan Sawyer and Stacy Wood Source: Journal of Marketing, Vol. 61, No. 3 (Jul., 1997), pp. 38-53 Published by: American Marketing Association Stable URL: http://www.jstor.org/stable/1251788 . Accessed: 27/11/2013 12:51

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Marketing Association is collaborating with JSTOR to digitize, preserve and extend access to Journal of Marketing.

http://www.jstor.org

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions Joseph Alba,John Lynch,Barton Weitz, Chris Janiszewski, RichardLutz, Alan Sawyer,& Stacy Wood InteractiveHome Shopping: Consumer,Retailer, and ManufacturerIncentives to Participatein Electronic Marketplaces

The authors examine the implications of electronic shopping for consumers, retailers, and manufacturers. They assume that near-term technological developments will offer consumers unparalleled opportunities to locate and compare product offerings. They examine these advantages as a function of typical consumer goals and the types of products and services being sought and offer conclusions regarding consumer incentives and disincentives to purchase through interactive home shopping vis-a-vis traditional retail formats. The authors discuss implications for industry structure as they pertain to competition among retailers, competition among manufacturers, and retailer-manufacturer relationships.

Aconfluence of technological, economic, and cultural impact this new retail formatcould have on the retail indus- forces has made possible a new and revolutionary try. In the first half we analyze the demand-side issues, distributionchannel known generically as interactive examining what IHS offers consumers that could motivate home shopping (IHS). Although only in its infancy, IHS has them to alter their present shopping behavior.In the second the potential to change fundamentallythe mannerin which half we examine the impact of this new channel on industry people shop as well as the structureof the consumer goods structure and the competitive positioning of individual and retail industries.Projections about the diffusion of IHS firms. are sometimes breathtaking:Forecasts of IHS sales range from $5 billion to $300 billion by the year 2000 (Reda 1995; Interactive Home Shopping Defined Wilensky 1995). In contrast to such projections, current In defining IHS, we conceptualize interactivityas a contin- sales are barely perceptible.Internet sales in 1996 were esti- uous constructcapturing the of communi- mated at $500 million-less than 1% of all nonstore quality two-way shop- cation between two parties. (For an elaboratedtreatment of ping (Schiesel 1997). Combining Interet, other online ser- interactivityin the context of electronic media, see Hoffman vices, television home shopping, CD-ROM catalogs, and and Novak 1996.) In the case of IHS, the parties are the conventional catalogs, all nonstore combined retailing buyer and seller. The two dimensions of are accounts for only 5% to 10% of all retail sales, with little interactivity response time and response contingency. Because IHS growth in recent years. Therefore, IHS will need to offer involves electronic communication, the can be benefits superior to current nonstore channels in order to response immediate-similar to the time in face-to-face realize the more ambitious sales forecasts response that have been set communications. for it. Response contingency is the degree to which the responseby one partyis a function of the Our goal is to examine the effects of consumer, retailer, response made by the other party.We use the term home to and manufacturerbehavior on the diffusion of IHS and the merely indicate that the customercan engage in this interactionin a location other than a store. Figure 1 illustrates a somewhat futuristic JosephAlba is HollowayProfessor of Entrepreneurship,University of form of IHS. Florida.John Lynch is HanesCorporation Foundation Professor of Busi- The scenario portrayedin Figure 1 is highly interactive. ness Administration,Duke University. Barton Weitz is JCPenneyEminent Judy, the consumer, using an electronic shopper, BOB, can Scholarof ChrisJaniszewski Marketing, is AssociateProfessor of Market- specify the type of merchandisesought and then screen the RichardLutz and Alan are ing, Sawyer Professorsof Marketing,and Stacy located alternativesto a smaller set of that Woodis a doctoralcandidate in of Florida.The develop options Marketing,University she can view in detail. The interaction the to authorsgratefully acknowledge the support of the IBM Retail requires parties Group,espe- each other's databases.In ciallyDan Hopping and Dan Sweeney. The views expressed in the article query contrast,this level of inter- representthose of theauthors. activity and selection is not available from currentInternet retail sites, which functionas an unwieldy collection of elec-

Journal of 38 / Journalof 1997 Marketing Marketing,July Vol. 61 (July 1997), 38-53

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions FIGURE 1 *anefficient means of screeningthe offeringsto findthe most Illustration of IHS appealingoptions for moredetailed consideration, *unimpededsearch across stores and brands, and Judy Jamison sits in front of her home electronic center ?memoryfor past selections,which simplifiesinformation reviewingher engagement calendar displayed on her televi- searchand purchase decisions. sion screen. She sees that she has accepted an invitationto a formalcocktail party on Fridaynight and she decides to buy Our scenario implies that the consumerowns the intelli- a new dress for the occasion. She switches to her personal gent search agent BOB, which might be a software package electronic shopper, BOB, and initiates the following bought by Judy and parameterizedto fit her needs on the exchange: basis of data she However, other search BOB: Do wish to to a or provides. engines you browse, go specific store, also be owned and controlled the retailer buy a specific item? might by (e.g., Judy: Specific item http://www.landsend.com)or an independentthird party,as BOB: Type of item? in Continuum Software's "Fido the Shopping Doggie" Judy: Blackdress (http://www.shopfido.com)or Anderson Consulting's Bar- BOB: Occasion? (menu appears on screen) gain-Finder(http://bf.cstar.ac.com/bf). The consumer might Formalcocktail Judy: party enter a site to be by the retailer'ssearch Price interrogated engine. BOB: range? (menu appears) the search be a $300-$500 Finally, engine might operatedby third-party Judy: in a as in BusinessWeek'sMaven BOB: 497 items have been identified.How manydo you expert product category, want to review? agent for finding personal computers (http://www.maven. Judy: Just 5 businessweek.com). The consumer might pay a service [Five pictures of Judy in each dress appear on the screen charge to use the site, or retailers might pay to have their with the price, brand name, and the IHS retailerselling it informationavailable at the site. listed beneath each one. clicks on one of the dresses Judy We assume that all of these types of search agents will and it is enlarged on the screen. Another click and Judy views the dress from different Another click and exist but will have differentmixes of informationdesired by angles. the other specifications such as fabric and laundering instructions parties. However, consumers must have access to appear. Judy repeats this routine with each dress. She vendors'databases if the scenario portrayedin Figure 1 is to selects the one she finds most appealing. BOB knows her become reality. In the current transitionalperiod, product measurements and picks the size that fits her best.] search often is dictated by the vendor. Moreover, global BOB: How would like to for this? you pay (menu search across vendors can be thwartedby actions taken by appears) individual vendors. In the and market American end, technological Judy: Express forces will determine the extent to which consumers can BOB: Nieman Marcus [the firm selling the dress Judy selected] suggests a Xie scarf and Koslowbelt to gain access to the informationthey desire. In the latter half complementthis dress. of this article we consider vendors' incentives to inhibit [Judy clicks on the items and they appear on the screen. informationexchange and their likelihood of success. First, Judy inspects these items as she inspected the dresses. She however, we consider the critical attributesaffecting con- decides to both accessories. BOB then asks purchase Judy sumers' incentives to adopt IHS. about delivery.Judy selects two-day delivery at a cost of $5.00.] BOB: Just a reminder.You have not purchased hosiery The in 30 days. Do you wish to reorderat this time? Demand Side: Consumers Judy: Yes and IHS BOB: Same shades? Consumer Trade-offs Judy: Yes Similar to any innovation,IHS will need to match or exceed the utility provided by traditional formats to succeed. In Table 1 we compare six retail formats in terms of benefits tronic catalogs (Rigdon 1996). Consumers cannot search and costs to the consumer.The three in-store formats are a and for items of nor can quickly easily specific merchandise, prototypicalconvenience-goods store (supermarket),a spe- screen and merchandiseon the basis of their they compare cialty-goods store (departmentstore), and a shopping-goods desires. Individualretailers road idiosyncratic provide maps store (category specialist) (cf. Copeland 1923); the nonstore to facilitate search within their sites but avoid formats that formatsare the traditionalcatalog, the presentInternet offer- would consumers' needs. satisfy comprehensive However, ing, and the IHS format described in Figure 1. Although the such as those described in with capabilities Figure --along scenario in Figure 1 is intriguing,department and specialty the and of customized design production clothing-soon stores afford buyers the opportunityto touch and feel mer- could become available to consumers (Cortese 1996; Hill chandise and obtain informationfrom sales associates. "Cat- Maes 1995; 1995; Negroponte 1995). egory killers"such as Best Buy and Office Depot offer com- The scenario illustrates the critical attributes following parisonsacross a wide arrayof alternativesin a specific mer- the of IHS: affecting adoption chandise category. Also, all in-store formats allow immedi- *faithfulreproduction of descriptiveand experientialproduct ate delivery. information, It is importantto clarify our orientationand assumptions ?agreatly expanded universe of offeringsrelative to whatcan before discussing the relative merits of these retail formats be accessednow through local or catalogshopping, in detail. First, our analysis assumes that technology has

InteractiveHome Shopping /39

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions TABLE 1 Dimensions Affecting Relative Attractiveness to Consumers of Alternative Retail Formats Current Department Category Internet IHS Dimension Supermarket Store Specialist Catalog Retailer Format Providing Alternatives for Consideration Numberof Categories Medium Medium Low Low Low Low or High Alternativesper Category Medium Low Medium Medium Low High

Screening Alternatives to Form Consideration Set Selecting ConsiderationSet Medium High Medium Low Low High

Providing Informationfor Selecting from Consideration Set Quantity Medium Medium Medium Medium Medium High Quality High High High Medium Low Low or High ComparingAlternatives Medium Medium High Low Low Depends on Supplier

Ordering and Fulfillment:Transaction Costs DeliveryTime Immediate Immediate Immediate Days Days Days SupplierDelivery Cost Low Low Low High High High CustomerTransaction Cost High High High Low High Low SupplierFacility Costs High High High Low Low Low Locationsfor Placing Orders Few Few Few Everywhere Many Many Other Benefits Entertainment Low High Medium Low Low Medium Social Interaction Medium High Medium Low Low Low Personal Security Low Low Low High High High

developed to the point in which a highly evolved IHS sys- Dickson1982). For example, most apparelis sold in depart- tem is readily available to a significant number of house- mentand specialty stores because these outlets offer the ser- holds. Therefore, as characterizedin Table 1, IHS enables vice andaccessories sought by customersbuying clothing. In consumers to access merchandiseunavailable in their local contrast,apparel sales makeup a smallerpercentage of total sales at discounters.Catalog apparel sales skew toward unfit- markets, veridical informationabout merchandiseat gather ted clothing items. Catalogs are especially attractivefor occa- a a low cost, efficiently screen the offerings of broad cross- sions where the purchasercannot achieve a superiorfit by vis- section of suppliers by avoiding unwanted alternativesand iting a store (as when buying a gift for a relative in a distant unimportantfeatures, and easily locate the lowest prices at city). which a specific item is offered. As we discuss in the fol- *Catalogs dominate current Internet retailers. It is therefore lowing sections, IHS retailers currentlyenjoy considerable unsurprisingthat there are so few examples to date of busi- latitude in designing their offerings to exploit or subvert nesses making significantrevenues by selling merchandiseon such activities. the Internet. Second, the values used to describe a formatare illustra- *The IHS formatdiffers from currentInternet retailers primar- tive. It is not our intention to argue the specifics, which can ily by providingmore alternativesfor consideration,the abil- vary across retailerswithin a given format(e.g., across those ity to screen alternativesto form considerationsets, and information to facilitate selection from the considerationset. selling products that can rather than cannot be digitized, those ratherthan breadthof emphasizing depth selection). We expect changes in the benefits relating to consumer Our assessment of of the six retail formats is performance informationacquisition to drive any change from the cur- based on the retail structurein the well-developed industry rent, nearly nonexistent penetrationof Internetretailing to urban and suburbanUnited States. In less retail developed the more optimistic sales projectionsfor IHS. Consequently, environments,nonstore formatscould be much more attrac- we focus our analysis on the dimensions in the first three tive and Klein (Quelch 1996). sections of Table 1, which bear on the cost of information Table 1 illustratesthree main points: search, ratherthan on those in the bottom half of the table. *Fora given productcategory, a comparisonof traditional Retailers and retail formats compete in the types of retail formats (e.g., departmentstores, category specialists, informationthey convey effectively to customers. Just as in or consumer makes catalogs selling clothing electronics) Erlich and Fisher's (1982) analysis of "derived demand for apparentthe basis for competition. The benefits provided by advertising," we "derived demand for retailer infor- different formats influence the types of merchandisethat can analyze mation about Erlich and Fisher note be sold successfully; product,situation, and consumercharac- products." that infor- teristics affect the relativeweights of these benefits when con- mation reduces the wedge between the market price sumers select a format (Day, Shocker, and Srivastava 1979; received by the seller and the "full price" paid by the buyer.

40 Journalof Marketing,July 1997

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions The wedge between marketprice and full price includes the opportunityto consider a thousand alternativesversus ten costs of obtaining informationabout productsand of dissat- alternativescould be enough to switch some of them from isfaction from disappointingpurchases. Consumers demand in-store shopping to IHS. However, other consumers could informationthat reduces this wedge. Such informationalter- find it too tedious and stressful to look throughinformation natively can be derived from their own prior knowledge, on hundredsof productsidentified for consideration,unless advertising,or "otherselling efforts"-notably information there is reason to expect that the added alternativesare sys- from retailers. tematically different from the first ones considered, with a Although we focus on retail competition throughinfor- different distribution of utilities. Consequently, the mere mation, we recognize that retail formats differ on many fac- capability of IHS to increase the universe of potential tors, such as entertainmentand personal safety, that con- options is not a majorreason for its adoption. tributeto the utility consumers obtain from the "total shopping experience" (cf. Tauber 1972) and that transaction Screening Alternatives to Form Consideration Sets costs related to and fulfillment are an ordering important The attractivenessof the opportunityto inspect an expanded basis for For and competitive advantage. example, Verity number of alternatives is dependent in part on the con- Hof that it could be 25% less to (1994) suggest costly sumer's ability to sort efficiently through a potentially in direct with electronic channels. engage marketing daunting amount of information.A particularadvantage of Although consultants and the popular press widely draw IHS over alternativeformats is that consumers can screen similar we this as an conclusions, regard open question. On informationso that they can focus on alternativesthat match the one the is hand, IHS retailer not burdenedwith the cost their preferences. of convenient stores. On the other hand, the IHS locally In most productcategories, consumershave priorbeliefs retailer faces the cost of merchandise in small delivering and preferences about alternatives(Hauser and Wernerfelt to individualconsumers. It is to assess quantities premature 1990; Ratchford1982; Roberts and Lattin 1991; Simonson, the relative efficiencies. and electronic Using catalogs gro- Huber,and Payne 1988). Consumersuse this informationto as cery shopping (e.g., Peapod [Donegan 1996]) guides, make purchase decisions more efficiently by forming a however, it is not clear that consumers will mon- enjoy large small consideration set and then evaluating alternatives cost IHS. etary savings by using within this subset in more detail. The savings in searchcosts However, here, we focus on the informationaleffects of involved in using this two-step process often overwhelms electronic commerce as to retailer-consumer they pertain the potential opportunity cost of overlooking the "best" interaction.Excellent discussions of enhancedconsumer-to- alternative that would have been uncovered by carefully consumer interactionand the implicationsfor marketingare inspecting the entire universe of alternatives. available elsewhere and 1996; Hoffman (Armstrong Hagel Interactivehome shoppingenables the formationof con- and Novak 1996). sideration sets that include only those few alternativesbest suited to a consumer'spersonal tastes. This can be Providing Alternatives for Consideration screening done almost instantaneouslyusing electronic agents that use A benefit of IHS with significant compared other retail for- informationabout an individualconsumer's specific prefer- mats is the vast numberof alternativesthat become available ences and the alternativesavailable (Maes 1994). In Figure to consumers. Through IHS, a person living in Florida can 1, for example, BOB located 497 "suitable"black dresses shop at Harrod'sin London in less time than it takes to visit from a potentially much larger universe and rank-ordered the local Burdines departmentstore. these dresses on the basis of criteria (black/for- Economic search theory implies that if there are N alter- mal/$300-$500) supplied by Judy. An additional screening native brands or sellers available in a market,and the con- phase that is based on criteriaderived from priorinteractions sumer considers only a subset n < N, the utility of the cho- and stored in the agent's memory (such as the style she sen (best) alternative from the subset increases with n prefers and her trade-offs between price and quality) might (Hauser and Wernerfelt 1990; Ratchford 1980; Stigler reduce the set dramatically.The remainingalternatives then 1961). However, in terms of the benefits of search, there are could be searchedin more detail to choose the "best"of this strong diminishing returns. As additional alternatives are reduced set. If the screening criteria are highly correlated examined, the potential increase in benefits offered by the with Judy's full utility function,Judy can be reasonablycon- next alternativeis small. Inasmuch as the cost of searching fident that the alternativechosen after screening has utility for and evaluating new alternativescontinues to increase, a close to that associated with the choice she would have point is reached at which the expected cost of considering made if she had inspected all 497 alternativesexhaustively additional alternativesis greater than the potential increase (Feinberg and Huber 1996). in benefits. At this the point, consumer terminatessearch for Othershave noted that consumersoften rely on memory additional alternatives. Research also indicates that con- for the generationof alternativesfor consideration(Alba and sumers reach this point quickly: Consumers rarely visit Chattopadhyay 1985; Hutchinson, Raman, and Mantrala more than one or two outlets when they are buying expen- 1994; Kardes et al. 1993; Nedungadi 1990). In such cases, sive consumer durables (e.g., Newman and Staelin 1972; memory plays a screening function that is often only imper- Wilkie and Dickson 1985). fectly correlated with the consumer's utilities. An efficient Because IHS searchcosts are low and decline with expe- and dispassionate search agent should produce appropriate rience using the interface, simply providing consumers an brands that otherwise would not have been considered,

InteractiveHome Shopping /41

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions implicitly replacing memory with explicit product criteria Quantity of information. Retail formats differ in the for screening the universe of availableoptions to a manage- sheer amount of informationprovided about the merchan- able considerationset. dise they offer. For example, Lands' End not only provides Note that both BOB and retail store buyers have access faithful visual informationbut often gives great detail about to the same universe of merchandise and screen that uni- the constructionprocess, stitching, and materials.Other cat- verse to offer a subset intendedto appeal to end consumers. alogs provide only a few specifications per item, such as However, the assortments offered by store-based retailers price, weight, and brandor model. More informationcould are developed for marketsegments with significant within- increaseability to predictconsumption utility but add to pro- segment heterogeneity. Store customers are required to cessing costs. expend resources to form smaller considerationsets tailored Store-basedretailers also differ in the informationthey to their needs. Consumerscould find that the set providedby make available to consumers. Specialty and department the retaileris insufficient and opt to visit anotherstore. stores often providetrained and knowledgeablesales associ- Interactive home shopping has the potential to tailor ates, whereasdiscounters do not. Consequently,the effective consideration sets from a much broaderset of alternatives "database"of attributes available to consumers is much for specific individual consumers. The usefulness of these greater at specialty and departmentstores than it is at discounters and Store-based customized approacheswill depend on the consumer effort catalogers. retailershave an additional characteristicthat increasesthe usefulness of necessary to calibrate the screening mechanism and the radically the informationavailable to that accuracy with which the mechanismcorrelates with the con- consumers, is, interactivity. Interactionbetween a customer and sales associate enables sumer's full utility function for meaningfulalternatives. The store-basedretailers to provide informationabout the attrib- lower bound on effort to calibrate screening criteria comes utes that matterto the customer.Such selectivity gives con- from the use of past purchasehistory-as in the Peapod gro- sumers all the advantages of a large database without the cery shopping service, which keeps lists of regularly pur- large informationprocessing costs. Perhapsfor this reason, chased items for automaticrebuy. At the other extreme, the post-purchasereports from buyers of major durables indi- screening criteriain many currentInternet retailing sites are cate that the salesperson was the most useful information cumbersome in requiring the consumer to enter many source consulted, outstrippingConsumer Reports, advertis- responses to calibratethe function (e.g., Money Magazine's ing, and friends (Wilkie and Dickson 1985). Best Places to Live site on Pathfinder.com, at Firefly and Internet for music and Conversely,catalogers, discounters, present http://www.agents-inc.com films). retailersare forced to make decisions about which attributes Some search less data from the con- agents require input to promote on the basis of what is most desired the mar- sumer but at a cost of a few criteriathat col- by including only ket as a whole or by relativelycrude of the market. a segments lectively explain relatively small percentageof variancein However, consumers differ in their needs and therefore in a consumer's overall preferences.A good example is the use the information that will be of interest to them. Conse- of a standardInternet search like Alta Vista to engine shop quently, the information provided by catalogers and dis- for Advanced Photo System cameras. Others strike a better counters will be less valuable because it is not tailored to balance in asking for a compact set of preferences highly idiosyncraticdesires. related to a tastes but allow search of a limited person's only Interactivehome shopping should prove superior even set of alternatives Dell's site (e.g., computer http://dell.com to specialty and departmentstore retailers in terms of the for RackesDirect women's site at computers, clothing sheer quantity of attributeinformation it can provide about Fido the http://www.rackes.com/rackes.html, Shopping each stock-keeping unit. As a result of the interactivity of service for in Doggie shopping a broadcross-section of cat- IHS, retailers need not fear that the provision of informa- egories). Therefore, screening criteriacan be established in tion about an attribute that matters only to a few will different In the BOB ways. example, Judy explicitly stated impose search costs on the majority. In this respect, IHS her criteria when initiating the search. In the Internetsites resembles department and specialty stores. However, mentioned previously, screening criteria are limited to a because attribute information is available consistently small set specified by the retailer. from a central database, IHS effectively becomes a "super sales associate" (i.e., one that never gets sick, is not Providing Information to Evaluate Alternatives in moody, learns quickly, and never forgets). In contrast, the Consideration Set store-based retailers have a difficult time retaining knowl- One of the primarybenefits offered by traditionalretailers is edgeable sales associates, and in many cases it is not cost- informationthat enables consumersto predict how satisfied effective for them to do so. It should cost far less to add they would be if they purchased various offerings. The information to an IHS database than to attempt to dissem- degree to which this information is useful to consumers inate the same informationto sales associates throughcon- depends on the nature of the informationprovided and its ventional training. Consumers should seek out reliability. formats that enable Quality of attribute information. Information econo- them to make selections that maximize consumption utility mists often distinguish among search, experience, and cre- net of and searchcosts and price (Ehrlich Fisher 1982), even dence goods (Darby and Kari 1973), typically in terms of if retail formats offer identical competing merchandise consumers' ability to know quality before and after buying. (Hauser, and Urban, Weinberg 1993). In economic parlance,search goods are those whose quality

42 / Journalof Marketing,July 1997

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions and value to the consumer can be assessed easily prior to Cherry Garcia might be a search good at a Ben & Jerry's purchase. The quality of experience goods is difficult to store, which allows a consumer to taste the ice cream prior assess prior to purchaseand usage; however, because qual- to purchase.It would be an experience good at first if a per- ity can be assessed accurately after one use, the consumer son were buying at a supermarketthat sells ice cream only knows quality when an opportunityarises to repurchasethe in cartons and does not allow tasting priorto purchase.Con- same brand. For credence goods, quality cannot be known sequently, the Ben & Jerry's store initially would have an even after repeatedpurchase and use. informational advantage over the supermarket.However, A tempting conclusion that is based on this trichotomy when the consumer learns that CherryGarcia on the carton is that merchandisenow selected in store environmentspri- label reliably predicts experienced flavor, the supermarket marily on the basis of search and credence attributesis most no longer would be at a disadvantage. Similar dynamics amenable to electronic retailing (because direct experience explain why mail order computer giants Dell and Gateway is not required), whereas merchandise purchased on the have a customer mix dominatedby experiencedusers (Tem- basis of experience attributeswill be purchasedin stores. By plin 1996). similar reasoning, IHS and catalogs should be more suc- Similar principles govern the relative advantageor dis- cessful with merchandisedominated by visual attributesand advantageof store-basedretailers relative to nonstoreretail- should fare less well when touch, taste, and smell are impor- ers that sell throughcatalogs or IHS. For example, critical tant for evaluating quality. The latter senses require direct informationin the purchaseof apparelmight include search experience consuming or sampling the product (Anderson attributessuch as color and style-which ostensibly can be 1995). assessed accurately in a department store or catalog-as However,these conclusions fail to consider the key issue well as experiential attributes such as fit, which can be the of information.The regarding quality quality or useful- searched readily before purchase only in the department ness of information is determined the to which by degree store. However,when buying the item throughnonstore out- consumers their (or agents) can use the informationobtained lets, the ability to assess color depends on consumers'infer- to to their satisfaction from subse- prior purchase predict ences about the faithfulness of photographicreproduction which in turn on quent consumption, depends intimately and piece-to-piece variationin dyeing. Also, fit might seem consumers' inference rules (Alba and Hutchinson 1987; unpredictableunless the nonstoreretailer has consistent siz- Broniarczykand Alba 1994) and consumers' confidence in ing and the consumerhas learnedover time to infer what fit the of these rules and In the reliability (Wright Lynch 1995). is implied by a particularbrand and size. analysis that follows, we and (1995) adopt Wright Lynch's These examples illustrate three importantpoints: First, of the distinc- reinterpretation search/experience/credence consumers make inferences about productattractiveness on tion in terms of consumer inferences. for Specifically, expe- the basis of informationprovided by retailers,and retail for- rience and credence (but not search) there is at first a goods, mats compete on the informationthey provide as cues for low correlation between attributes subjective product these inferences; second, different consumers dif- observable to and benefits at the time of con- possess prior purchase ferent rules, and this affects the extent to which the infor- sumption.For experience goods, brandnames enable highly mation provided by any particularformat leads to competi- reliable inferences about consumption benefits after one tive advantage;and third, the cues that are deemed to pro- purchaseand use. This is not true for credence goods, pre- vide a reliable basis for inference are likely to change with sumably because feedback from the first use takes a long experience with the brand. The issues further time to materializeand is not of util- following predictive consumption the need to consider of satisfaction if the same brand were to be emphasize predictability ity repurchased. ratherthan a classification of of In information economists simple suitability "goods" addition, though initially to IHS that is based on the traditional of and credence search/experience/cre- spoke search, experience, "goods," it is dence distinction: now clear that all goods have some combination of search, experience, and credence attributes.A search good is sim- 1. The(in)adequacy of searchableexperiential information. In ply one for which the consumptionbenefits most important certainpurchase situations, information for some products to consumers are predictedreliably by attributeinformation with importantexperiential attributes cannot be gathered to In such in-store available to them before buying. This reasoning implies prior consumption. cases, shoppingoffers little overIHS. For flowersand wine are that the same can be a search, or cre- advantage example, product experience, consummate consumers who dence on the benefits that are to sensory products. However, good, depending important send flower arrangementsvia FTD must base theirdecisions consumers and the inferences consumers make about how on pictures in the florist's shop, and purchasersof wine fre- well those benefits are predicted by information available quently must rely on labels or advice from a retail sales prior to purchase. associate. Therefore, some products possessing important These observations have importantimplications in the experience attributescould be no less amenableto IHS than to present context because retail formatsdiffer greatly in their traditional shopping. In yet other cases, experiential to informationabout attributeslinked to attribute information could be conveyed more effectively capability provide than in-store. For benefits. electronically example, the electronic consumption Consequently, attributes that are bookseller Amazon has for search attributesin one (http://www.amazon.com) space format might be experience attrib- customers to post their own reviews of books, with positive utes in another-and this dictates patterns of competition wordof mouthclearly influencing sales. retailersover time. For if the among example, key attributes 2. Consistency and predictability.The ability to satis- of ice predict cream relate to experienced flavor, Ben & Jerry's faction from observableattributes is not inherentin the spe-

InteractiveHome Shopping /43

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions cific consumptionbenefits driving satisfaction, nor is it given decision task-so much so that decision makerscould inherentin the retailformat. Actions by retailersand manu- focus more on effort reductionthan on accuracy maximiza- facturerscan increase consumers' to ability predictpost-pur- tion (for a discussion, see Todd and Benbasat 1994). In this chase satisfactionfrom attributesobservable before pur- context, the of IHS are The initial (and chase.Consider the case of runningshoes purchasedby a advantages apparent. consumerwho cares about comfortand protectionfrom effortful) decision phase involving attribute-based,side-by- injury.We might expect that these features could be assessed side comparisons will be compressed if an efficient screen- betterwhen buying from a store,such as AthleticAttic, than ing mechanism is available. This should inspire consumers from a cataloger.Road Runner Sports, however, provides to learn and use more informationin the course of decision informationfor each shoe in its it to catalog,making easy making (cf. Kardes and Kalyanaraman1992; Russo 1977). assesssuitability for and and underpronators overpronators, In the transformationof the decision from a mem- customerscan submittheir old shoesfor a customanalysis addition, andsuggestions for suitablereplacements. ory-based to a stimulus-based choice should enhance the Manufacturers'actions also influencethe customer's precision of the decision process and thereforethe optimal- ability to predictconsumption satisfaction from pre-pur- ity of the ultimate decision (see Alba, Marmorstein,and chase information.If manufacturersbecome moreconsis- Chattapadhyay1992; Biehal and Chakravarti1983; Lynch, tentin thecharacteristics buildinto modelsin they differing Marmorstein,and Weigold 1988). theirproduct lines, consumers' ability to predictsatisfaction will rise accordingly.Comfort and sizing are important The combinationof IHS search, screen, and comparison attributesof runningshoes that requiredirect experience features also should promptconsumers to make their deci- withthe product.However, when a particularbrand is con- sions more rapidly (cf. Greenleaf and Lehmann 1995). sistentin the heightof its archsupport and the roominess of Research shows that the additionof attractivealternatives to its toe the of comfort and size is box, predictability a choice set could consumers to their choice enhanced.In essence, brand name converts prompt delay experience and Shafir because of the attributesto searchattributes that can be effectivelycommu- (Tversky 1992), perhaps perceived nicatedverbally or visually(see Agins 1994). possibility that even more attractiveoptions have yet to be 3. Other determinants of satisfaction. Satisfaction is deter- inspected (Karni and Schwarz 1977). Insofar as search and minedby morethan the consumptionexperience with the comparisonminimize the possibility of regretover choosing product;it also is affectedby the beliefthat one hasexhaus- a suboptimal product,both decision speed and satisfaction searchedthe set tively of acceptablealternatives such that with the decision process should increase. thereis no a missed regretregarding opportunity(Gilovich A caveat is at this Most of an andMedvec 1995). Interactive home shopping provides the appropriate point. aspects potentialfor a moreextensive search than that which con- efficient search engine point to improved decision quality. sumerscould accomplish in a store. However, it has been noted recently that though some deci- These considerationsimply that consumeradoption sion aids could improve decision making, abuse is possible on depends more than the (retail format-independent) (Todd and Benbasat 1994). In particular, Widing and importanceof search,experience, and credence attributes to show that the decision aid most to be the consumer. Talarzyk(1993) likely a part of an electronic search agent (i.e., a cutoff rule that Comparison of alternatives. Retail formats differ in the enables formation of a consideration set containing only extent to which they facilitate the comparison of alterna- those alternativesthat pass consumer-specifiedattribute cut- tives in the consideration set. For example, most in-store offs) can lead to suboptimal decisions in efficient choice retailers stock alternativecolors, styles, and brands in each sets. In addition, a separatestream of researchshows that a product category. An appealing characteristicof category second likely characteristicof IHS-visually rich presenta- specialists such as Circuit City and Office Depot is the tion-can distortthe decision process by divertingattention breadthof selection and customers' ability to make side-by- to peripheralcues and away from informationthat is most side comparisons of brands.Similarly, consumers shopping importantfor the task at hand (Jarvenpaa1989, 1990; cf. for apparel can compare the fit of different alternatives. Edell and Staelin 1983). Current Internet retailers do not offer this opportunity.In addition, currentIHS retailers are selective in the informa- Summary of Key Consumer Factors Affecting Use tion presented, whereas in-store retailers allow the con- of the IHS Format sumer to control the basis for comparison of alternatives. factors will influence a Research shows that consumers acquire and process Many consumer's decision to shop informationin ways made easiest by the constraintsof the electronically versus in-store. We focus on the benefits per- informationformat (Bettman and Kakkar 1977). However, taining to the consumer's informationacquisition and pro- consumers preferformats that promotemaximum flexibility cessing that enable consumersto locate and select merchan- to engage in either attribute-or alternative-basedprocessing dise that satisfies their needs, because the fundamentalben- (Bettman and Zins 1979). This preference for flexibility to efit of IHS is to lower the cost of informationsearch (Bakos engage in attribute-basedprocessing should be stronger for 1991). In summary,then, the growthof IHS is dependenton novices in a product class than for experts (Bettman and the following factors: Park 1980); experts know what levels of an attributeare ?Vastselection: If theformat does not allow for andcom- attractivewithout having to on relative informationto quick rely prehensiveinspection of an expandedset of options,elec- make that assessment (Mitchell and Dacin 1996). troniccommerce will mimic the shoppingexperience now It is that effort looms when argued large decision mak- availablethrough catalogs and achieve a relativelylow level ers consider the effort-accuracy trade-off required in any of penetration.

44 / Journalof Marketing,July 1997

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions *Screening:If consumerscannot screen the largenumber of makes recommendationsand informs consumers of where optionsmade available, the advantagesof vastselection will to find the best deal. be the costs of search. outweighedby In the remaining portion of this article, we discuss that *Reliability:If benefitsare morereli- consumption predicted nature of competition in an IHS environment,approaches from informationsearchable in storesthan ably experiential that firms can take to build in this fromsurrogate information searchable through IHS and con- competitive advantage and sumersare unwillingto bearthe risk,in-store shopping will environment, some importantissues confronting IHS continueto prosper. retailersand manufacturers. *Productcomparisons: To be successful,IHS mustallow the consumersto tailorthe basis for comparison of alternativesin The Role of Price and Quality orderto make the system with the processby compatible To a a whichconsumers prefer to makedecisions. Interactive home complete sale, vendor must be considered by a con- shoppinghas the potentialto providesuperior information sumer and the consumer must fail to consider a superior presentationformats for makingthese comparisons. alternative(Nedungadi 1990). Retailers believe that an IHS presence can increase the probabilityof being considered, Withoutthese benefits, IHS will not develop beyond the but conditional on the achievement of that goal, IHS can relatively unattractivecollection of electronic catalogs rep- have only a negative effect on the Internet In the next section we profits by intensifying price resenting present offering. competition with other IHS alternatives.Inasmuch as estab- review the incentives and disincentives for retailers and lished retailershave less to gain in terms of increasingcon- manufacturersto stimulate the development of the IHS siderationprobability, it is perhapsunsurprising that few of channel and the informationto attract provide appropriate the most aggressive entrants into IHS have a store- consumers. large based presence. But the conclusion that IHS must lower profits throughhigher price competition does not necessar- The Supply Side: Retailers, ily follow. Generally speaking, informationthat is easy to and IHS obtain or that can discriminate unambiguously among Manufacturers, options tends to receive higher weight in the consumer's For retailersthe most threat many significant posed by IHS decision process. Price informationpossesses both proper- is that profits will be eroded drasticallyby intensified price ties, which suggests that the concerns of retail firms are well competition that will ensue as consumers' search costs are founded. However,just as in the debate on economic effects lowered. Consequently, many retailers are making limited, of advertising (Mitra and Lynch 1996; Rosen 1978), IHS experimentalinvestments in electronic commerce that, iron- also can reduce the cost and increase the discriminating ically, have none of the characteristicswe describe previ- power of informationregarding merchandise quality. as ously necessary for IHS to be preferredto existing for- A strong parallelcan be drawnbetween the introduction mats. firms stand-alonesites Many participatethrough (such of IHS into the present retail environmentand the develop- as World Wide Web home that pages) increase the costs of ment of discount stores 40 years ago (Sheffet and Scammon cross-store When conducting comparisons. third-partyelec- 1985). Discount stores offered consumers an opportunityto tronic search such as agents Bargain Finder forgo personalized service in return for lower prices. The are created to (http://bf.cstar.ac.com/bf/) compare prices result was an increase in price competition followed by different charged by vendors for the same compact disc, attemptsto avoid such competition throughfair trade laws. some retailers access. When in deny participating interactive Proponentsof fair trade laws arguedthat, without some pro- some firms malls, require exclusivity agreements that pro- tection for department and specialty stores, discounters tect them from the kinds of cross-store comparisons that would drive them out of business; this, it was argued,would would make IHS truly useful to the consumer. leave a shopping environmentin which price could be dis- It is reasonableto assume that firms that have made sub- cerned easily but nuances of quality could not. Conse- stantial commitmentsto an existing business formator tech- quently, consumers would become more price sensitive, will nology adopt defensive responses to radical change sellers would adjustover time to compete more on price and (Leonard-Barton1995). In the case of IHS and other radical less on quality,and consumers would suffer throughthe lack changes, we argue that these defensive approachesare likely of interest in providing superior merchandise and service to fail in the long run, because the ultimatenature of the IHS quality.Although the advent of discount stores did increase channel and its appeal to consumers is beyond the control of price competition in some merchandise categories, many individual firms. Firms might attemptto build walls around consumersshop at retailers,such as Nordstrom,that provide their offerings that make comparison across retailers and superiorinformation and services even though they charge a manufacturers difficult. However, consumers will prefer higher price. Such inherent consumer heterogeneity sug- retailersthat freely provide such informationand make cross gests that no one retail format can dominate all segments. shopping easy; therefore, isolationist vendors could be The potentialimpact of IHS on the natureof competition bypassed in the search process. Eventually, intelligent in the retail industry should be considered in this context. agents will allow consumers to searchacross vendorsto find Although consumers shopping throughan IHS channel will offerings that possess the set of attributesdesired. Attempts be able to collect price information with little effort, they to limit informationwill be met with new formats that dis- also will be able to review at a low cost quality-relatedinfor- seminate information (Bakos 1991). Therefore, an elec- mation about most search attributesand some experience tronic version of Consumer Reports could emerge that attributes.For example, an electronic merchantof custom

InteractiveHome Shopping / 45

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions oriental rugs can convey clearly real differences in patterns Assortmentsof complementarymerchandise. The oppor- and materials used for construction.An electronic grocery tunity to make multiple-itemsales is importantfor two rea- service such as Peapod can enable customers to sort cereals sons. First, by making multiple-itempurchases from an IHS by nutritional content, thus making it easier to use that supplier,customers reduce the shippingcosts, which thereby attributein decision making. Insofar as (1) quality-related reduces the net price. Second, the IHS retaileris in an ideal informationis importantto consumersand (2) brandswithin position to tailor a secondary offering to a customer on the a category are differentiated,IHS can lead to less price sen- basis of the customer's primary purchase objective. We sitivity at the brand level and more sensitivity to search might suggest that electronic agents will put togethercom- attributesassociated with quality than does traditionalshop- plementary bundles of products from multiple suppliers. ping (cf. Mitra and Lynch 1995). However, to accomplish this task, the agents would need to This is a critical point for manufacturersthat offer dif- possess an extremely broad knowledge base, such as infor- ferentiated merchandise with superior performanceattrib- mation on what ties and shirtsgo togetherand what ingredi- utes. Similarly, retailers that carry unique merchandise ents are needed to make a good Brunswickstew. Even with- and/or provide superior information about merchandise out the presence of electronic agents, IHS offers retailersan could face less ratherthan more price competition. Perhaps opportunityto merchandisetheir wares in ways not previ- this is why vendors cooperating with multiple-category ously possible. Traditional merchandising is limited by search agents such as Fido the Shopping Doggie physical constraints. Floor space and shelf space limit the (http://www.shopfido.com/Vendors.html)are predominantly numberof complements that can be placed in close proxim- manufacturersand retailersselling highly unique merchan- ity to any given product.However, even the Internetallows dise such as arts and crafts, alternativemusic, hot sauces and nearly unlimited cross-referencingthrough hypertext. Inter- spices, and gourmet foods and wines. Conversely,manufac- active home shopping faces no such problems, and the effi- turers of "me-too" brands competing on cost can expect cient IHS merchandisershould realize superiorgains in cus- more intense price competition with the diffusion of IHS, tomer retention and cross-selling-goals that are increas- and retailers carrying nationally brandedmerchandise with ingly important regardless of distribution channel (e.g., limited service also will face increasedprice competition. Reicheld 1993). The opportunityto cross-sell extends well Therefore, the introduction of the IHS channel will beyond shirtsand ties. Diversified vendorsthat own subunits intensify the competitive environment, but this need not that are only modestly related to each other in terms of the shift the emphasis from quality to price. By providing more consumer goal they serve could realize synergies not possi- informationto consumers with minimal search cost, manu- ble with conventional channels (cf. Benjamin and Wigand facturersand retailerswith differentiatedofferings will have 1995 on "virtualvalue chains"). a greateropportunity to educateconsumers about the unique Collection and utilization of customer information. benefits they offer, and consumers will find it easier to Database marketing is an important capability for IHS access and compare the offerings of firms competing on retailing (cf. Blattberg and Deighton 1991; Peppers and price. Rodgers 1993). Interactivehome shopping will increase the and Developing Competitive Advantage in IHS importance acceleratethe developmentof databasemar- keting because more comprehensivecustomer-specific data "Location, location, location" is the classic to the response can be captured.All consumers who can about the three most factors in shop electronically question important retailing. be identified at the individual level. Moreover,unlike other The development of IHS reduces the certainly importance formats,consumer browsingcan be tracked.That is, records of location. The successful IHS retailerwill need to a adopt can be constructednot only of what consumers bought, but strategy that seeks competitive advantagein one or more of also what they inspected and for how long. the following areas: (1) distributionefficiency, (2) assort- Interactivehome shopping retailerscan use these data to ments of complementary merchandise, (3) collection and provide information-basedvalue to the customer utilization of customer information, of by (1) (4) presentation to and considerationsets information electronic and using technology identify display through formats, (5) unique most suited to individualconsumer tastes and merchandise. (2) providing informationabout those options that enables consumers to Distributionefficiency. Consumers perform a majorpor- predicttheir satisfactionafter purchase.Consumers, in turn, tion of the distribution function when purchasing from are likely to become loyal to an IHS retailer offering this stores. They transport merchandise from stores to their service. This loyalty advantagecould be sustainablefor two homes and bring unsatisfactory merchandise back to the reasons: First, consumers who experience high satisfaction store. In an IHS system, these substantial costs of home may not defect to competing IHS retailers;and second, as delivery and returns will be fully borne by the seller and consumers patronize a particular IHS retailer more fre- must be factored into the price. Because these costs are sub- quently, more informationcan be collected. Thus, a cycle is IHS stantial, players that can select and package multiple createdwherein consumer satisfaction provides the opportu- items for delivery to individualhouseholds will have a com- nity to learn how to provide greatersatisfaction. Consumers petitive advantage over IHS competitors that lack such would incur switching costs and an initial decrease in cus- skills. The importanceof this advantagenaturally is greater tomer service if they took their business to a competing IHS when the preparationfor shipping constitutes a large frac- retailer. Insofar as informationabout the consumer is pro- tion of the overall price of the product. prietary,sustainability ensues.

46 / Journalof Marketing,July 1997

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions Presentation of information.Traditionally, some stores Entry Into IHS by Retailers have differentiationon the basis of and sought atmospherics Table 2 leads to some when contrasted service. Both still could a role in and each will interesting insights play IHS, with Table 1. Table 1 that retailersare more a new technical skill as evidenced the recent suggests catalog require set, by vulnerable to IHS than are other retail formats. Interactive acquisitionof softwarecompany Davidson and Associates and home shopping retailersand catalogers share the same limi- interactiveentertainment company Sierra On-Line by CUC tations in terms of delivery timing and providing informa- International,a leadingdirect marketer and interactiveretailer. tion about experience attributes;interactive home shopping merchandise.From the retailer's the Unique perspective, dominates catalogs in terms of the information provided. most method for differentiation straightforward increasing However, Table 2 indicates that catalog retailers are best and is to sell merchandisethat reducing price competition preparedto exploit IHS, inasmuchas they possess orderful- cannot be offered elsewhere. has Uniqueness traditionally fillment systems and databasemanagement skills that match been achieved in several ways: the requirements of IHS. As an example, Lands' End *Privatelabels: IHS retailerscan developtheir own private- (http://www.landsend.com)has a "Specialty Shopper Ser- labelmerchandise that they offer exclusively. vice" that coordinates outfits for a whole wardrobe, helps *Brandedvariants: Alternatively, retailers can work with man- the customer find his or her correct size, and keeps a file on ufacturersto provide "brandedvariants" sold exclusively sizes, tastes, and address and credit card The past purchases, throughthat retailer (Bergen, Dutta, and Shugan1996). numbers. the skills for effective visual intentis to incentivesfor retailersto better Also, necessary pre- provide provide sentation of information in IHS follow the visual servicewhen inter-store(but not inter-brand)competition is closely reduced.(As noted subsequently,however, this methodof merchandisingskills necessary for catalogs. Catalogerscan achievinguniqueness could lose some effectivenessin the reap efficiencies by listing their products electronically contextof IHS.) ratherthan in a more expensive print format when penetra- *Offeringassortments of complementstailored to customer tion of IHS justifies the productionof electronic assets by needs: One for retailersto make their merchandise way savings of significant paperand postage costs. "unique"is by creatingbundles of complementsthat are avail- However,the of successful to ableonly separatelyelsewhere. For example, with each bottle ability currently catalogers of wine offeredby VirtualVineyards (http://www.virtualvin. adapt to IHS can be expected to vary sharply,depending on com), customerscan get complementaryrecipes from noted the strategy the catalog retailer has used to establish com- Bay-areachefs. Although some of thewines are available else- petitive advantage. For example, Spiegel sells primarily Virtual customersto sat- where, Vineyardsallows its anticipate brandedmerchandise, which is susceptible to price compar- isfactionwhen servingthe wine with a particularmeal. In isons. retailersthat brandedmerchandise essence,the bundlerather than the bottleof wine Catalog emphasize wine-recipe will be vulnerable with a retailersuch becomesthe unit of analysis.Interactive home shopping retail- particularly compared erscan use customer information skills noted previously to sug- as Lands' End, which has developed high-quality, private- gest bundlesthat lead to multiplesales and increased customer brandmerchandise. satisfaction-withthe side benefit of reducingshipping costs. Interactivehome shopping is ideal for retailers, such as Nieman-Marcus, Harrod's, Gumps, and Saks, that enjoy strong national reputations for high-quality, unique mer- Implications for Firms in the chandise, but that have only spotty or regional penetration. Retail Industry Such retailers are well positioned to take advantage of the feature of IHS an interna- The success of consumer productmanufacturers and retail- market-expanding by attaining in ers in the IHS environmentwill be determinedby the degree tional presence without making significant investments visual and leases to which their strengthsand weaknesses match the capabili- store locations, merchandising, (Rennie Most of these stores an effective ties requiredto build competitive advantage (Aaker 1989). 1993). currently possess Interactive home In Table 2 we provide such a comparison. In this table, we mail-order catalog operation. shopping a cus- assess each type of firm in terms of the skills previously also is ideal for niche retailersthat appeal to far-flung identified as bases for competitive advantage in the IHS tomer base (cf. Quelch and Klein 1996; Wererfelt 1994). channel. We consider the likely impact of IHS on their busi- For example, Hot Hot Hot (http://www.hothothot.com)is a nesses and how their businesses are likely to adapt. After- specialty store that carries more than 450 brands of hot ward, we examine the impact of IHS on manufacturers. sauce. The Internet gives this firm internationalexposure

TABLE 2 IHS Success Capabilities Possessed by Firms Catalog Traditional Category Merchandise Skills for Developing Advantage Retailers Stores Specialists Manufacturers DistributionEfficiency to Homes High Mediumto High Medium Low Provisionof ComplementaryAssortments High High Low Medium Collectionand Use of Customer Information High Mediumto High Low Low Presentationof MerchandiseInformation High Mediumto High Low Medium Abilityto OfferUnique Merchandise Medium Medium Low High

InteractiveHome Shopping /47

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions without significant advertisingand only 300 square feet of many consumers, shopping is an experience that transcends store space (Carlton 1996). productpurchase. One methodof differentiatinga retail out- Conversely, national chains such as Sears have far less let is to provide benefits that enhance the experience.Tradi- incentive to participate.These chains possess high levels of tionally, this has involved improvements in ambiance. penetration through their ubiquitous stores. Even among Increasingly, the entertainmentvalue of shopping is being national departmentstore chains, there are clear differences emphasized. IncredibleUniverse, Niketown, and the Mall of in incentives to enter IHS. Both Sears and JCPenney have America are possible harbingersof the future.(For a discus- saturatedthe domestic marketwith stores, but JCPenneyis sion of how IHS retailersmight respond to these efforts by also the largest catalog retailerin the United States. This cat- in-store retailersand improve the social experience benefits alog operationprovides the infrastructurefor fulfillmentand for IHS customers, see Armstrongand Hagel 1996.) visual merchandisingthat is well suited to IHS. Sears exited In-storeretailers with an IHS presence can use IHS as a the "Big Book" catalog business largely because its catalog source of advertisingto presell merchandiseand to check its fulfillment operations and technology were antiquatedand availability in local stores. This would enable the customer because the cost of rebuilding these systems was prohibi- to pick it up or have it delivered from the local store. tive. This absence of efficient fulfillment systems for indi- In-store retailers and IHS retailers will need to reduce vidual orders creates a further disincentive for Sears to their reliance on nationally branded merchandise to lure engage in IHS. people into their sites and will need to redoubletheir efforts to develop privatelabel brands.Therefore, the trendseen in Adaptation of In-Store Retailers to IHS store-basedretailers such as JCPenney-which increasingly brands The DEFENDER model (Hauser and Shugan 1983) sug- promotes private-label such as Arizonajeans-could accelerate. gests that in-store retailers should react to emerging IHS retailers by emphasizing attributes of their offering for on which they have a comparativeadvantage. Therefore, store- Impact Category Specialists and Discounters based retailers should (1) focus on merchandise that has In light of the consumer analysis in Table 1, category spe- importantexperiential attributes that are search attributesin cialists appear particularly vulnerable to IHS retailing. a store but experience attributesin IHS, (2) capitalize on Aside from the immediacy of delivery,this shoppingformat their relative advantagein providing informationtailored to offers few informationaland noninformationalbenefits. In the needs of specific customers, (3) emphasize the noninfor- addition, these formats emphasize brandedmerchandise for mational benefits of shopping, (4) complement IHS with which price competition will increase with the advent of their in-store business, and (5) place more emphasis on IHS. However, the nature of these outlets varies greatly in merchandise. unique terms of their operation, merchandise, and relationships Because it is more difficult to provide some experience with suppliers. information through IHS, in-store retailers must focus on Toys 'R' Us enjoys national (and increasingly interna- merchandise that possesses characteristicsconsumers can tional) penetration.If Toys 'R' Us were to sell electronically, assess veridically only through contact with the merchan- it might experience significantcannibalization of its in-store dise. For example, bedding and linens come in standard sales, making IHS less attractiveto it than to an entrepreneur sizes and are amenable to IHS; consequently, department entering the toy business throughIHS or even to an F.A.O. stores might need to decrease space allocated to this mer- Schwartz, which is smaller and more specialized. chandise and increase floor space devoted to tailored cloth- Circuit City appears to be as vulnerableas Toys 'R' Us ing. They also might need to increase resources devoted to is to competition from IHS retailers.However, the structure service personalized associated with those items (e.g., alter- of the consumer electronics industryis considerablydiffer- ations). Similarly, departmentstores should shift their mer- ent from the toy industry.The consumerelectronics industry chandise mix to emphasize items for which immediate,low- is dominated by a few suppliers that make most of their cost access to the merchandiseis important. profits from sophisticated, high-technology products. The To offset the of IHS retailersto ability provide personal- benefits of these productscan be credibly demonstratedonly ized informationat home, in-store retailers should improve in a store environment.To motivate electronics retailers to the personalized information they offer using their sales provide this information to consumers, manufacturers associates or in-store kiosks. For example, Best Buy uses employ several mechanisms designed to protect specialty kiosks extensively to alleviate physical store constraintsand retailers from price competition from mass merchandisers detailed provide product information.Media Play uses in- that sell only the low-end and mid-rangemodels that domi- store stations to enable listening acoustic sampling of com- nate the market. (For example, co-op advertising offers to discs to Used-car pact prior purchase. superstore CarMax mass merchandisers can be made contingent on pricing kiosks that allow flexible provides screening criteria, side- cooperation.)Moreover, distribution of high-endproducts to of screened by-side viewing options, and the printingof car IHS retailers would encourage free riding and reduce in- lot location for candidate cars-all of which maps greatly store retailers' incentive to provide product-differentiating reduce search costs inherent in navigating a huge and het- information. erogeneous on-site inventory. Home Depot is similar to Toys 'R' Us in terms of distri- Because IHS retailerscan provide greaterinformational bution intensity but is less vulnerable because many of its in-store retailers benefits, must emphasize ancillarybenefits goods demand immediacy, highly tailored advice from such as entertainmentand opportunities to socialize. For expert associates, or direct (non-video) inspection of size,

48 / Journalof Marketing,July 1997

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions specifications, or colors. Home Depot also offers a level of believes it would be more profitable to sell directly than hand-holding from expert sales associates that cannot be throughstores, it will hesitate to disintermediatefor fear of duplicated electronically. Moreover, bulky do-it-yourself alienating those stores that currentlycarry its lines. merchandisecan be expensive to ship directly to homes. These considerationsimplicitly identify those manufacturersthat might have an incentive to disintermediate.Man- Implications for Manufacturers and Retailers ufacturerspossessing extremely strong brandnames and the ability to produce complementarymerchandise might con- Disintermediation. The most important structural sider disintermediation.Consider Levi Strauss. Its brand change that could be broughtabout by IHS is disintermedi- names are among the strongestin the apparelindustry. Net- ation, wherein manufacturersbypass the retailer and sell work externalities are weak for the markets it faces, either directly to consumers. Although the IHS channel does offer because it produces complements demanded by consumers manufacturersan opportunity to deal directly with con- (e.g., Dockers slacks and shirts) or because, for core prod- sumers (cf. Benjamin and Wigand 1995; Pine, Peppers, and ucts such as Levi's 501 Blue Jeans, consumers can be Rogers 1995), Table 2 illustrates the limited capabilities of assuredof a match withoutbuying the complementaryitems most manufacturersto succeed as IHS retailers-which from the same seller. In contrast, a maker of dress slacks suggests that the degree of disintermediationwill not be such as Savane would have less incentive to consider disin- significant. termediation because its brand name has less pull and Manufacturerscannot easily and efficiently duplicate a because demandfor Savane slacks benefits from significant variety of services that retailers performfor both manufac- networkexternalities when sold in departmentand specialty turers and consumers (see Sarkar, Butler, and Steinfield stores carryingother manufacturers'lines. 1996). The classic functions undertakenby retailers and The foregoing discussion applies to manufacturersof other firms in a distributionchannel include breaking bulk nationally branded merchandise that distribute through (converting caseload shipments into individual items); pro- store-basedretailers. Small manufacurersand entrepreneurs, viding assortments that permit one-stop shopping; holding conversely, are more prone to disintermediatebecause their inventory to make merchandise available when customers alternativesto IHS are less attractive.Small or new firms- want it; and providing a variety of transactionfeatures and even those with superior new products-find it difficult to services that include credit, alterationand assembly of mer- obtain shelf space or awareness. For these producers, IHS chandise, attractivedisplay, dressing rooms, personal assis- could reduce barriersto entry by making it possible for con- tance in selecting merchandise, repair services, returnser- sumers to locate them. In this sense, IHS functionsjust like vices, and warranties (Levy and Weitz 1995). Although advertising in helping heterogeneous consumer segments these functions can be provided by manufacturersselling find productsthat match their tastes (Rosen 1978). be more effi- directly through IHS, present retailers might Brands and branding.A brandis a search attributethat not cient at performing these functions. Manufacturersare assures consumersof a consistent level of productquality. It to customers. lack the highly skilled at selling directly They might be the only attributeavailable to assess some credence a household level and efficient systems to fulfill orders at goods. Because a brandoffered by different outlets can be have limited capability to offer the complementaryproducts easily comparedby IHS shoppers,manufacturers of branded that increase customer satisfaction and reduce shipping merchandiseare particularlyvulnerable to price competition costs. Similarly,manufacturers may not be able to deal with at the retail level; consequently, IHS retailers will find it high returnrates encounteredin nonstoreretailing formats. unattractiveto sell their merchandise.It is ironic that strong We noted previously that JCPenney'scatalog operation brands increase the attractivenessof IHS to consumers by is the largest in the United States. It is undergirdedby an providingsufficient informationto predict satisfactionwith- extremely efficient and capital-intensivesystem for accept- out experiencingthe merchandise,but that this same mech- ing orders, packaging them together, and shipping them to anism makes these brandsless attractivefor retailersto carry customers to be picked up at local stores and catalog distri- in the face of IHS. bution centers. The difficulty and expense of duplicating In the present retail environment,branded-goods manu- such a system drove Sears from the catalog business; the facturersemploy restricteddistribution in a territory,relying scale economies are high. It seems unlikely that many man- on location to reduce price competition among retailersand ufacturerswould find it worthwhile to build such a fulfill- ensure retailercooperation. This mechanism is not feasible ment operation from scratch or to replace retailers in the in the low search cost environmentof IHS retailing.There- supply chain with outsourcers to handle the functions now fore, manufacturersof brandedmerchandise must focus on performedfor them by retailers. other methods for insulating IHS retailers from price com- These fulfillment-baseddisincentives to disintermediate petition. One method is the production of private-label will be lower among products for which fulfillment costs brandsfor each retailer.Alternatively, the manufacturercan contribute only a small fraction of the sales price to con- produce "brandedvariants" of nationally brandedproducts. sumers. Products such as computer software, brandedjew- These brandedvariants might be retailer-specificmanufac- elry, and high-end perfumes fit this description. turermodel numbers(e.g., Sony Model MA 3150, which is Finally, although manufacturersmight be tempted to sold only by Service Merchandise). generate incrementalsales by adding a direct IHS channel to Neither of these alternativeswill be relished by manu- their store-basedchannels, entry into IHS could alienate the facturers that have developed strong national and interna- stores that now carry their lines. Unless the manufacturer tional brands. It is obvious why such manufacturerswould

InteractiveHome Shopping /49

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions be loathe to find themselves mainly as suppliers of private- sumer search (see Hoch and Deighton 1989). In contrast,the label merchandise.The prospect of employing an expanded potential IHS search environment is highly interactive, branded-variantstrategy also is perilous, albeit in more sub- information intensive, and low in cost. In this alternative tle ways. Increasing the number of branded variantscould environment,research questions in need of attentioninclude have the effect of lowering the average attractivenessof the the following: manufacturer's offerings. The easy search-and-compare of IHS could render the existence of triv- *Whatfundamental changes occur in informationprocessing aspects transparent as a functionof the of electronicsearch ial differences between manufacturersto availablity agents? models, forcing Withfew exceptions(e.g., Widingand Talarzyk 1993), con- create larger differences in their variants to satisfy retailer sumerresearch fails to examinethe heuristicsand resulting demands of noncomparabilityacross retailers.However, if a decisionquality that are enabled by the searchand screening significant amount of purchasing still occurs in store, the operationsthat constitute the mostattractive features of IHS. manufacturerrisks losing sales because the variantcarried A relatedquestion involves the influenceof searchagents on the store is not the variant desired the consumer. It consumerlearning. Traditional shopping affords consumers by by the to learnthe distributionof attributevalues seems that manufacturerswill be driven to variants opportunity produce across alternatives;search agents merely a set of that are exclusive to each retailer with which do busi- produce they alternativesthat satisfy particularcriteria. Thus, on some ness (e.g., "Liz Claibome for Macy's"). dimensionsof productknowledge, search agents can produce The preferredsolution for manufacturersis to create a undesirableoutcomes. level of brandpower that ensures cooperation from retailers in *How does the balanceof memory-basedversus stimulus- termsof resaleprice maintenanceand othertactical mandates. based processingshift as the searchenvironment changes? Manufacturersthat hold such power could threatendefectors Some researcherscriticize research on consumerchoice for on stimulus-based and subtly (Barrett1991). Few brandshold such sway, however, focusing paradigms ignoringimportant and it is that even fewer will be able to maintainsuch memory-basedaspects present in nearlyall consumerdeci- likely sions (Alba,Hutchinson, and 1991).Our with distribution IHS. "brand Lynch assumptions power through Nonetheless, regardingan effectiveIHS system, conversely, argue in favor building"is anotheroption for manufacturersthat fear the lev- of greaterattention to stimulus-basedprocessing inasmuch as eling effects of IHS. On the surfacethis could seem counter- electronicsearch agents will reducememory constraints sig- intuitive:The threatof IHS to vendors is that its information nificantly.An especiallylarge effect should be observedwhen featureswill speed commoditizationand expose paritywhere the optimal choice set includes items from different product it exists; should decreasethe value of the brand. categories (Ratneshwarand Shocker 1991). Although human parity be bounded salient in classes in which memory might by temporarily options, Nonetheless, product technology electronic can retrieveall alternatives with the cannot and for firms that cannot win agents tagged provide advantage consumer's goal or desired benefit (e.g., "gift"). battles, becomes an technological image building option. *Importantquestions also exist regardingshort-term memory For in example, the case of fashion goods, brandscan attain and perceptualissues. Just as the cognitiveimplications of cachet through a carefully crafted marketing strategy. hypertext are virtually unexplored (Rouet et al. 1996), con- Plainly, brandswill have least influence in nonimage, parity sumer researchersmust understandhow memory constraints product classes. However, parity is not a limiting factor affect decision makingas consumersmove from brandlistings when credence attributes are all to brandattributes to third-partyevaluations to complementary important-and nearly and on. From credence attributes For productinformation, so a vendor'sperspective, products possess (Levitt 1981). there is an informationvacuum for- when is difficult to brand name regardingoptimal display example, quality assess, mat. Insofar as search agents efficiently retrieve requested serves as a surrogate(see our previous discussion). And, as alternatives,impulse purchasingwill occur less frequently(cf. marketers long have known, brands can signal quality or Park,Iyer, and Smith 1989). Vendors must understand the cog- other dimensions of differentiation falsely through long- nitive and perceptualrules that can promptconsumers to make term positioning tactics or explicit attempts to frame con- electronic detours in their search for goods and services. sumer decisions (cf. Gardner 1983; Hoch and Deighton *Howdo the contentand presentation of productinformation 1989). Therefore, another irony of IHS could be that the affect consumers'willingness to make choices without the Are there to create technology that enables consumers to make more directlyexperiencing product? ways intelligent vocabularies" Brown,and Hoch in some cases can induce manufacturersto take "consumption (West, 1996) comparisons thatincrease consumers' willingness to inferexperential ben- actions intended to an produce opposite outcome in other efits from descriptive,electronically provided information? cases. As with other determinantsof IHS success, the impor- *Howare consumerconfidence and satisfactionaffected by tance of the brandand the viability of a brand-buildingstrat- searchprocesses that enable efficient screening? The ability to egy will vary as a function of the product class and firms' screenproducts by attributecreates a muchmore manageable individualcompetencies. information environment but simultaneously allows some attractiveoptions to go unnoticed.Do consumersexperience a greaterbut illusory sense of confidence in choices made from Research Opportunities effortfully but incompletely constructedconsideration sets? The advent of IHS raises significant questions pertainingto *How will consumers react to the collection of detailed infor- consumer behaviorand industrystructure. Previous research mation about their needs and purchasebehavior by IHS retail- focuses on heuristics used consumers to make choices ers? The utilization of this informationto tailor merchandise by presentationsprovides a benefit to consumers, but will con- when search and are difficult and comparison relatively sumers be willing to make this personaldata available?What Such a focus has been costly. appropriatebecause the envi- can IHS retailersdo to assure consumers that personal infor- ronment,often aided by the retailer,tends to discouragecon- mation will not be misused?

50 / Journalof Marketing,July 1997

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions *What are the true dynamics of price sensitivity in this envi- formedby retailers.Looking to the future,how will IHS inter- ronment? Although greater amounts of information should act with developments in distributionand flexible manufac- increase sensitivity among comparablegoods and reduce sen- turingto enable manufacturersto mass customize their offer- sitivity for differentiatedgoods, empiricalresearch is required ings and deliver them efficiently to customers? to understandhow this general conclusion is moderated by *Many traditional retailers will find themselves in multiple type of good, branding,and the mannerin which vendors pre- channels-maintaining their bricks-and-mortaroperations sent information. while also creating an electronic presence. What are the eco- *How will the nature of the relationshipsamong manufactur- nomics of such dual systems and how sustainableare existing ers, retailers, and consumers evolve as a function of technol- stores if electronic sales grow to significant levels? In other ogy-based reductionsin search costs (cf. Zettelmeyer 1996)? words, if total sales do not increase, at what point does canni- *Towhat extent will vendorsbe able to control the search envi- balizationreduce the viability of stores? ronment?In part, technological developments can determine These questions are a mere sample of a much larger set the of vendors to inhibit search and comparison. At ability both within and the of our present, Internetvendors can prevententry by search agents. beyond scope analysis. Clearly, form of IHS are However, irrespectiveof technology, to what extent will mar- predictions about the ultimate fate and risky. ket forces determinenot only control of entry but also search However, it is equally clear that this emerging channel pro- proceduresallowed by vendors? vides marketing researchers and practitioners with much *We argue that disintermediationwill not blossom in the pre- opportunity to test their theories and apply their tools. sent environment because of the critical functions now per-

REFERENCES Aaker, David (1989), "ManagingAssets and Skills: The Key to a Research,6 (September), 141-53. Sustainable Advantage," California Management Review, 31 Biehal, Gabriel and Dipankar Chakravarti(1983), "Information (Winter),91-106. Accessibility as a Moderatorof ConsumerChoice," Journal of Agins, Terry (1994), "Go Figure: Same Shopper Wears Size 6, 8, ConsumerResearch, 10 (June), 1-14. 10, 12," The WallStreet Journal, (November 11), B 1. Blattberg,Robert C. and John Deighton (1991), "InteractiveMar- Alba, Joseph W. and Amitava Chattopadhyay(1985), "Effects of keting: Exploiting the Age of Addressability,"Sloan Manage- Context and Part-Category Cues on Recall of Competing ment Review, 33 (Fall), 5-15. Brands,"Journal of MarketingResearch, 22 (August), 340-49. Broniarczyk,Susan and Joseph W. Alba (1994) "The Role of-Con- and J. Wesley Hutchinson (1987), "Dimensions of Con- sumers' Intuitionsin Inference Making,"Journal of Consumer sumer Expertise,"Journal of ConsumerResearch, 13 (March), Research, 21 (December), 393-407. 411-54. Carlton, Jim (1996), "Think Big," The Wall Street Journal, (June , and John G. Lynch (1991), "Memory and Deci- 17), R27. sion Making,"in Handbookof ConsumerTheory and Research, Copeland, Melvin T. (1923), "Relation of Consumer's Buying Harold H. Kassarjianand Thomas S. Robertson, eds. Engle- Habits to Marketing Methods," Harvard Business Review, I wood Cliffs, NJ: Prentice Hall, 1-49. (April), 282-89. , Howard Marmorstein, and Amitava Chattopadhyay Cortese, Amy (1996), "Software's Holy Grail," Business Week, (1992), "Transitionsin Preference Over Time: The Effects of (June 24), 83-92. Memory on Message Persuasiveness,"Journal of Marketing Darby, Michael R. and Edi Karni (1973), "Free Competition and Research, 29 (November), 406-16. the OptimalAmount of Fraud,"Journal of Law and Economics, Armstrong,Arthur and John Hagel III (1996), "The Real Value of 16 (April), 66-86. On-Line Communities," Harvard Business Review, 74 Day, George S., Allan D. Shocker, and Rajendra K. Srivastava (May/June), 134-41. (1979), "Customer-OrientedApproaches to IdentifyingProduct Anderson, Christopher(1995), "The Accidental Superhighway:A Markets,"Journal of Marketing,43 (Fall), 8-19. Survey of the Internet,"The Economist, (July 1), 50-68. Dickson, PeterR. (1982), "Person-Situation:Segmentation's Miss- Bakos, J. Yannis (1991), "A StrategicAnalysis of Electronic Mar- ing Link,"Journal of Marketing,46 (Fall), 56-64. ketplaces,"MIS Quarterly, 15 (September),295-310. Donegan, Priscilla(1996), "HighTech in a High TouchWay," Pro- Barrett,Paul M. (1991), "Anti-DiscountPolicies of Manufacturers gressive Grocer,(December), 133. Are Penalizing CertainCut-Price Stores," The WallStreet Jour- Edell, Julie A. and RichardStaelin (1983), "The InformationPro- nal, (February27), B -B5. cessing of Pictures in Print Advertisements,"Journal of Con- Benjamin, Robert and Rold Wigand (1995), "Electronic Markets sumer Research, 10 (June), 45-61. and Virtual Value Chains on the InformationSuperhighway," Ehrlich, Isaac and LawrenceFisher (1982), "The Derived Demand Sloan ManagementReview, 37 (Winter),62-72. for Advertising: A Theoretical and Empirical Investigation," Bergen, Mark, Shantanu Dutta, and Steven M. Shugan (1996), American Economic Review, 72 (June), 366-88. "BrandedVariants: A Retail Perspective,"Journal of Marketing Feinberg, Fred M. and Joel Huber, (1996), "A Theory of Cutoff Research, 33 (February),9-19. Formation Under Imperfect Information,"Management Sci- Bettman, James R. and PradeepKakkar (1977), "Effects of Infor- ence, 42 (January),65-84. mation PresentationFormat on ConsumerInformation Acquisi- Gardner,Meryl Paula (1983), "AdvertisingEffects on Attributes tion Strategies," Journal of Consumer Research, 3 (March), Recalled and CriteriaUsed for Brand Evaluations,"Journal of 233-40. ConsumerResearch, 10 (December), 310-19. and C. Whan Park (1980), "Effects of Prior Knowledge, Greenleaf, Eric A. and Donald R. Lehmann(1995), "Reasons for Exposure, and Phase of the Choice Process on ConsumerDeci- SubstantialDelay in Consumer Decision Making,"Journal of sion Processes: A Protocol Analysis," Journal of Consumer ConsumerResearch, 22 (September), 186-99. Research, 7 (December), 234-48. Gilovich, Thomas and VictoriaHusted Medvec (1995), "TheExpe- and Michel A. Zins (1979), "InformationFormat and rience of Regret: What, When, and Why," Psychological Choice Task Effects in Decision Making,"Journal of Consumer Review, 102 (April), 379-95.

InteractiveHome Shopping /51

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions Hauser,John R. and Steven Shugan (1983), " Defensive Marketing Newman, Joseph W. and Richard Staelin (1972), "Prepurchase Strategies,"Management Science, 3 (Fall), 327-51. InformationSeeking for New Cars and MajorHousehold Appli- , Glen L. Urban, and Bruce D. Weinberg (1993), "How ances,"Journal of MarketingResearch, 9 (August), 249-57. Consumers Allocate Their Time when Searching for Informa- Park,C. Whan, Easwar S. lyer, and Daniel C. Smith (1989), "The tion," Journal of Marketing Research, 30 (November), Effects of Situational Factors on In-Store Grocery Shopping 452-66. Behavior:The Role of Store Environmentand Time Available and Birger Wernerfelt(1990), "An EvaluationCost Model for Shopping," Journal of Consumer Research, 15 (March), of Consideration Sets," Journal of Consumer Research, 16 422-33. (March), 393-408. Peppers,Don and MarthaRodgers (1993). The One to One Future. Hill, G. Christian(1996), "Cyberslaves,"The WallStreet Journal, New York:Currency Doubleday. (June 17), R20. Pine, B. Joseph II, Don Peppers, and MarthaRogers (1995), "Do Hoch, Stephen J. and John Deighton (1989), "Managing What You Wantto Keep YourCustomers Forever?" Harvard Business Consumers Learn from Experience,"Journal of Marketing,53 Review, 73 (March/April),103-114. (April), 1-20. Quelch, John and Lisa Klein (1996), "The Internetand Interna- Hoffman, Donna L. and Thomas P. Novak (1996), "Marketingin tional Marketing,"Sloan Management Review, 38 (Spring), Hypermedia Computer-Mediated Environments: Conceptual 60-75. Foundations,"Journal of Marketing,60 (Winter),50-68. Ratchford,Brian T. (1980), "TheValue of Informationfor Selected Hutchinson, J. Wesley, Kalyan Raman, and Murali K. Mantrala Appliances," Journal of Marketing Research, 17 (February), (1994), "Finding Choice Alternatives in Memory: Probability 14-25. Models of Brand Name Recall," Journal of Marketing (1982), "Cost-Benefit Models for Explaining Consumer Research, 31 (November), 441-61. Choice and InformationSeeking Behavior,"Management Sci- Jarvenpaa,Sirkka L. (1989), "The Effect of Task Demands and ence, 28 (February),197-212. GraphicalFormat on InformationProcessing Strategies,"Man- Ratneshwar,S. and Allan D. Shocker (1991), "Substitutionin Use agement Science, 35 (March), 285-303. and the Role of Usage Contextin ProductCategory Structures," (1990), "Graphical Displays in Decision Making-The Journal of MarketingResearch, 28 (August), 281-95. Visual Salience Effect," Journal of Behavioral Decision Mak- Reda, Susan (1995), "InteractiveHome Shopping:Will Consumers ing, 3 (3), 247-62. Catch Up with Technology?"Stores, (March),20-24. Kardes, FrankR. and GurumurthyKalyanaraman (1992), "Order- Reicheld, FrederickF. (1993), "Loyalty-BasedManagement," Har- of-Entry Effects on Consumer Memory and Judgment: An vard Business Review, 71 (March/April),64-73. Information Integration Perspective," Journal of Marketing Rennie, W. R. (1993), "Global Competitiveness: Born Global," Research, 29 (August), 343-57. McKinseyQuarterly, (September 22), 45-52. , Gurumurthy Kalyanaraman, Murali Chandrashekaran, Rigdon, Joan (1996), "Caughtin the Web,"The WallStreet Jour- and Ronald J. Dornoff (1993), "BrandRetrieval, Consideration nal, (June 17), R14. Set Composition,Consumer Choice, and the PioneeringAdvan- Roberts, John H. and James M. Lattin (1991), "Developmentand tage,"Journal of ConsumerResearch, 20 (June), 62-75. Testing of a Model of ConsiderationSet Composition,"Journal Karni, E and A. Schwarz (1977), "Search Theory: The Case of of MarketingResearch, 28 (November), 429-40. Search with UncertainRecall," Journal of Economic Theory, 16 Rosen, Sherwin (1978), "Advertising,Information, and Product (October), 38-52. Differentiation,"in Issues in Advertising, David Tuerck, ed. Leonard-Barton,D. (1995) Wellspringsof Knowledge.Cambridge, Washington,DC: American EnterpriseInstitute, 161-91. MA: HarvardBusiness School Press. Rouet, Jean-Francois,Jarmo J. Levonen,Andrew Dillon, and Rand Levitt,Theodore (1981), "MarketingIntangible Products and Prod- J. Spiro (1996), Hypertext and Cognition. Mahwah, NJ: uct Intangibles," Harvard Business Review 59 (May/June), Lawrence ErlbaumAssociates. 94-102. Russo, J. Edward(1977), "The Value of Unit Price Information," Levy, Michael and BartonWeitz (1995), RetailingManagement, 2d Journal of MarketingResearch, 14 (May), 193-201. ed. Burr Ridge, IL: RichardD. Irwin, Inc. Sarkar,Mira, Brian Butler, and Charles Steinfield (1996), "Inter- John Lynch, G., Jr., HowardMarmorstein, and Michael F. Weigold medaries and Cyberintermediaries:A Continuing Role for (1988), "Choices from Sets Including Remembered Brands: Mediating Players in the Electronic Marketplace,"Journal of Use of Recalled Attributesand Overall Evaluations,"Journal of Computer-Mediated Communications, 1 (3) (http://jcmc. ConsumerResearch, 15 (September), 169-84. huji.ac.il/voll/issue3/). Maes, Patricia(1994), "Agents that Reduce Workand Information Schiesel, Seth (1997), "Payoff Still Elusive on Internet Gold Overload,"Communications of the ACM, 37 (July), 31-40. Rush,"The New YorkTimes, (January 2), C17. (1995), "IntelligentSoftware," Scientific American, (Sep- Sheffet, Mary Jane and Debra L. Scammon (1985), "Resale Price 84-86. tember), Maintenance:Is it Safe to Suggest Retail Prices?"Journal of Mitchell, Andrew A. and Peter A. Dacin (1996), "The Assessment Marketing,49 (Fall), 82-91. of Alternative Measures of Consumer Expertise,"Journal of Simonson, Itamar,Joel Huber,and John Payne (1988), "The Rela- ConsumerResearch, 23 (December), 219-39. tionship between Prior Brand Knowledge and Information Mitra, Anusree and John G. Lynch, Jr. (1995), "Towarda Recon- Acquisition Order," Journal of Consumer Research, 14 ciliation of Market Power and InformationTheories of Adver- (March), 566-78. tising Effects on Price Elasticity," Journal of Consumer Stigler, George (1961), "The Economics of Information,"Journal Research, 21 (March), 644-59. of Political Economy, 69 (January/February),213-25. and (1996), "Advertising Effects on Consumer Tauber,Edward (1972), "Why Do People Shop?"Journal of Mar- Welfare:Prices Paid and Liking for Brands Selected,"Market- keting, 36 (October),42-49. 7 644-59. ing Letters, (March), Templin, Neal (1996), "VeteranPC Customers Spur Mail-Order Nedungadi, Prakash(1990), "Recall and ConsumerConsideration Boom," The WallStreet Journal, (July 17), B 1. Sets: Influencing Choice without Altering Brand Evaluations," Todd, Peter and Izak Benbasat(1994), "The Influenceof Decision Journal Consumer 17 of Research, (December), 263-76. Aids on Choice Strategies: An ExperimentalAnalysis of the Negroponte, Nicholas (1995), Being Digital. New York:Alfred A. Role of Cognitive Effort,"Organizational Behavior and Human Knopf. Decision Processes, 60, 36-74.

52 / Journal of Marketing, July 1997

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions Tversky,Amos and EldarShafir (1992), "Decision UnderConflict: puter-Assisted Formats in Multiple Decision Environments," An Analysis of Choice Aversion," Psychological Science, 6 Journal of MarketingResearch, 30 (May), 125-4 1. (November), 358-361. Wilensky, Dawn (1995), "The Internet,The Next Retailing Fron- Verity, John and Robert Hof (1994), "The Internet:How Will It tier,"Discount Store News, (December 4), 6-7. Change the Way You Do Business," BusinessWeek,(November Wilkie, William L. and Peter R. Dickson (1985), "ConsumerInfor- 14), 80-86, 88. mation Search and Shopping Behavior,"Management Science Wernerfelt,Birger (1994), "An Efficiency Criterionfor Marketing Institutepaper series, Cambridge,MA. Design," Journal of Marketing Research, 31 (November), Wright, Alice and John G. Lynch, Jr. (1995), "Communication 462-70. Effects of Advertising Versus Direct Experience When Both West, PatriciaM., ChristinaL. Brown, and StephenJ. Hoch (1996), Search and ExperienceAttributes are Present,"Journal of Con- "ConsumptionVocabulary and PreferenceFormation," Journal sumer Research,21 (March),708-18. of ConsumerResearch, 23 (September), 120-35. Zettelmeyer, Florian (1996), "The Strategic Use of Consumer Widing, Robert E. II and W. Wayne Talarzyk(1993), "Electronic Search Cost," working paper, Massachusetts Institute of InformationSystems for Consumers:An Evaluation of Com- Technology.

Want to do business with senior-level marketing managers and educators?

Meet them in the pages of the Journal of Marketing, the premier journal in its field. You'll find that 50% of Journal of Marketing subscribershold titles such as manager, director,vice president, president, CEO and partner.Another 39% are educators.Next issue, this space can feature an advertisementfor your organization's products or services.

Contact Kathy Hays for details on advertisingin the Journal of Marketing, Marketing News and four other fine American Marketing Association publications.

Phone:312-831-2729* Fax:312-648-0103 * Email: [email protected]

InteractiveHome Shopping / 53

This content downloaded from 152.3.152.120 on Wed, 27 Nov 2013 12:51:55 PM All use subject to JSTOR Terms and Conditions