• 沒有找到結果。

I Want Products My Own Way, But Which Way? The Effects of Different Product Categories and Cues on Customer Responses to Web-based Customizations

N/A
N/A
Protected

Academic year: 2021

Share "I Want Products My Own Way, But Which Way? The Effects of Different Product Categories and Cues on Customer Responses to Web-based Customizations"

Copied!
9
0
0

加載中.... (立即查看全文)

全文

(1)

© Mary Ann Liebert, Inc. DOI: 10.1089/cpb.2008.0111

I Want Products My Own Way, But Which Way?

The Effects of Different Product Categories and Cues

on Customer Responses to Web-based Customizations

Chia-Chi Chang, Ph.D. and Hui-Yun Chen, M.B.A.

Abstract

Mass customization is a strategy that has been adopted by companies to tailor their products in order to match

customer needs more precisely. Therefore, to fully capture the value of mass customization, it is crucial to

ex-plore how customers react to mass customization. In previous studies, an implied premise has been that

con-sumers are keen to embrace customized products, and this assumption has also been treated by firms as a

pre-requisite for successful mass customization strategies. However, an undesirable complexity may result from

difficult configuration processes that may intimidate and confuse some customers. Hence, this study explores

strategies that marketers can employ to facilitate the customization process. Specifically, this study investigates

how to enhance customer satisfaction and purchase decision toward customized products by providing cues

compatible with the product category. It is hypothesized that for search products, customers rely more on

in-trinsic cues when making configuration decisions. On the other hand, for experience products, customers

per-ceive extrinsic cues to be more valuable in assisting them to make configuration decisions. The results suggest

that consumers tend to respond more favorably toward customized search products when intrinsic cues are

provided than when extrinsic or irrelevant ones are provided. In contrast, when customizing experience

prod-ucts, customers tend to depend more on extrinsic cues than on intrinsic or irrelevant ones.

7 Introduction

M

ASS CUSTOMIZATION, empowered by flexible

computer-aided manufacturing systems, can provide customers tailor-made products and services with near mass-produc-tion efficiency.1 Many companies are increasing their

re-liance on such a strategy in order to match customer needs more precisely.2However, despite the initial enthusiasm

as-sociated with mass customization, consumers can be over-whelmed by its complexity when facing a series of difficult choices. Huffman and Kahn coined the term “mass confu-sion” to delineate a situation in which an excessive variety of potential choices in mass customization results in a high level of perceived complexity and/or information overload, which tends to overwhelm and dissatisfy customers.3

How-ever, only a limited amount of research has focused on how to mitigate the confusion experienced by customers through the employment of decision aids and, in turn, to elevate cus-tomer satisfaction. Therefore, this paper explores the effect of potentially constructive tactics, such as information

pro-vision, that firms can employ to augment customer satisfac-tion and the purchase intensatisfac-tion of customized products.

Mass confusion, mainly triggered by complicated deci-sion-making configurations, can be mitigated if firms can proffer helpful information, such as product specifications and/or samples, to reduce perceived uncertainty and com-plexity. This issue is particularly relevant in online shopping environments where e-tailers are well equipped to provide information in a more efficient and interactive fashion.4,5

Hence, if e-tailers can provide customers with information that facilitates decision making, customers can better handle difficult choices in customization than they can in the tradi-tional settings. However, the effect of information on tomer configuration choices associated with Web-based cus-tomization is an underresearched area, and many questions have remained unanswered. For example, do all kinds of in-formation carry the same weight in the creation of customer satisfaction? Do customers rely on different types of cues/in-formation when assessing products in different categories? Prior studies involving product classification schemes

(2)

(search versus experience products)6,7have found that

di-versity in the types of information can have effects on prod-uct judgments. For example, for experience prodprod-ucts, ex-trinsic cues provided by other consumers (such as users’ ratings, discussions, and comments) tend to carry more weight in customer product evaluations than do intrinsic cues, such as color, size, style, specification, and function.8,9

For search products, however, customers are more likely to rely on information about product specifications from re-tailer Web sites when forming their decisions.8,9Based on

previous research, this study investigates further the effec-tiveness of various cue types in decisions concerning con-figuration across different product categories. It is proposed in this paper that compatible cues can improve customer evaluations of customized offerings to a greater degree than do less compatible ones.

Literature Review and Hypotheses

Perceived complexity

Mass confusion, which creates barriers to customers in customization tasks through difficult decision-making pro-cesses, mainly results from perceived complexity.3For

ex-ample, when customers do not have sufficient knowledge3

or well-structured preferences10,11 to select the “best fit”

product on their own, they may perceive the configuration process to be too complex and difficult. Therefore, strategies reducing perceived complexity need to be explored in order to facilitate the customization process for customers with limited product knowledge. One way to achieve this goal is to facilitate the preference identification process. In a cus-tomizing task, when customers can better compare attributes and identify the most preferred choice for all the customiz-able attributes, the phenomenon of mass confusion can con-ceivably be alleviated or eliminated.3

Factors reducing perceived complexity

Information presentation format.In an effort to reduce the perceived complexity in mass customization,3Huffman and

Kahn have demonstrated that a proper format for the pre-sentation of information (attribute-based rather than alter-native-based) can facilitate the formation of customer pref-erence. When information about a product is presented in the attribute-based format, customers tend to be more will-ing to make configuration choices and feel more satisfied with the customization process. On the other hand, when in-formation about the product options are presented by alter-native, customers perceive more complexity in deciding their preference and feel less satisfied with the customized prod-uct because it is more difficult to process information in-volving multiple attributes simultaneously.3

Provision of information.While previous research efforts have aimed at understanding how information can be orga-nized differently to facilitate the customization process, lit-tle has been documented in terms of how a variety of infor-mation provided by firms can affect customer configuration choices. One exception by Piller and his colleagues proposes that knowledge shared among customers in virtual commu-nities can facilitate customer interaction with the cus-tomization system.12For example, customers in LEGO User

Group Network can create their own works by using the con-figuration system and at the same time can communicate with other customers in the community to exchange opin-ions or even to create joint designs. Therefore, they suggest that by implementing a virtual community for customiza-tion, firms can encourage customers to communicate with each other to share their customization experiences, recom-mend product options, inspire each other with creative ideas, and eventually to feel more confident in customizing their own products.12 Another e-commerce research also

main-tained that word-of-mouth recommendation can signifi-cantly reduce decision difficulty. For example, when choos-ing a restaurant, customers who receive recommendations with high credibility notably reduce their search time and number of searches.13Such findings highlight that the

pro-vision of information and/or decision-making heuristics can make purchase decisions easier for consumers. However, it remains unknown if all information helps customers to spec-ify their preferences in the same way.

Compatibility between types of cues and product categories

Varied reliance on information across product categories. Previous marketing literature6,7has ascertained that the

in-formation sources customers rely on to evaluate product quality prior to purchase can have differential impacts for search and experience products.14,15 Search products are

those for which sufficient information can be acquired through objective product descriptions (such as ingredients and contents appearing on product labels) from firms prior to purchase.7,16 On the other hand, experience products,

which are products whose quality can only be determined after purchase,7 tend to cause a greater uncertainty about

product quality and, hence, usually necessitate further in-formation acquisition.9,14,17The difficulty of evaluating

in-formation derived directly from product-inherent character-istics tends to prompt customers to gather additional information from other reliable sources, such as recommen-dation from friends.9For example, perfume scent is difficult

to communicate using generic terms such as ingredient de-scriptions. Consequently, customers are more likely to draw on subjective comments of existing consumers to facilitate their decision making. This varied perceived importance of information across the search–experience classification scheme has been ascertained by Bei and her colleagues.9

When customers compare search product alternatives, they tend to trust their own judgments of product specifications provided by firms.9In contrast, the increasing difficulty in

assessing objective product descriptions for experience prod-ucts make opinions from other customers a more important source of information.9Therefore, it can be presumed that

differences in the effectiveness of information in assisting customers to make configuration decisions across product categories is rooted in the varied perceived importance of in-formation across search and experience products.a

Intrinsic vs. extrinsic cues and predictive and confidence values.The information that customers can acquire to form quality judgments prior to purchase can be classified into two broad categories: intrinsic and extrinsic cues.18Intrinsic cues

(3)

in-gredients or properties, which cannot be changed without al-tering the inherent characteristics of a product.18By contrast,

extrinsic cues are nonphysical product characteristics, such as brand, country of origin, and corporate image.18,19The

cues (intrinsic versus extrinsic) that are relied on more for quality judgment by consumers are determined by their pre-dictive and confidence values. Prepre-dictive value is defined as the extent to which a consumer perceives or believes that the cue is appropriately representative of product quality. Con-fidence value refers to the extent to which consumers are able to judge a cue precisely.18The attributes of search products

such as color, style, specification, and function, can usually be described in detail objectively. In such a case, customers can make judgments relying more on intrinsic cues, which reflect objective product characteristics.20–22In other words,

intrinsic cues have higher predictive and confidence values for search products. In contrast, many crucial attributes of experience products (for example, scent for perfumes) are dif-ficult to depict using specific and universally quantifiable terms. Therefore, customers prefer to rely more on extrinsic cues such as brand,19country of origin,23expert reviews,24

word of mouth,25and price and warranty.26This preference

can be attributed to two major reasons. First, customers might perceive intrinsic product descriptions to be not well repre-sentative of product quality. For example, for an experience product, such as dinner at a new restaurant, it is difficult to know how the food tastes simply based on the ingredients listed on the menu; therefore, customers may be more reliant on ratings, opinions, and discussions from other customers when forming their own judgments.9,13Second, even when

intrinsic cues are well representative of product quality, cus-tomers may not be confident in their ability or knowledge to judge product quality based on them because jargon is often involved.18,27Take automobiles, for example.7When

assess-ing vehicle reliability, model reputation, and drivassess-ing charac-teristics, customers may rely more on personal evaluations gathered from friends, online third parties, and online chat rooms.28This may be due to customers’ limited confidence

in evaluating the products based on the intrinsic cues provided. In summary, the compatibility between search products and intrinsic cues is postulated to be higher than that with extrinsic cues; on the other hand, extrinsic cues are more compatible with experience products than with intrinsic cues.

Effective cues can provide good reasons to purchase prod-ucts/services for customers whose preferences are too vague or too varied to identify a “best option” from the offered choice set.11Grounded on the assumption that people may

seek reasons to justify their choices to themselves,10effective

reasons are likely to reduce uncertainty perceived by con-sumers, particularly when they are overwhelmed by an enor-mous number of choices, which involve difficult tradeoffs. For example, previous research in cyberspace commerce has averred that the extent of product information provided29

and decision-making supports, such as frequently asked questions (FAQs) and links to product comparison Web sites,30are antecedents of consumer satisfaction. Higher

sat-isfaction can be achieved when customers can make better choices more efficiently with the support of these decision-making aids.30Compatible cues are likely to possess higher

predictive and confidence values and thus serve as better de-cision-making supports.

In Web-based customization configuration processes, the unique customer value is created by a series of configura-tion choices that allow customers to tailor the products. As such, consumers may need more reasons to justify their large number of decisions. Therefore, the impact of compatibility between cues and product categories can be more subtial in the context of Web-based customization than in stan-dard online shopping. Granted, customers are likely to ex-perience greater satisfaction with a product when more helpful reasons are provided in the customization process.10

Conceivably, customization is likely to result in greater sat-isfaction when compatible cues are provided. Hence, the fol-lowing hypotheses can be formulated.

H1: The compatibility between product category and cue type has a greater salutary effect on customer satisfaction with the product in the customization context than in the no-customization context.

H1a: For search products, customization would have a stronger salutary effect on customer satisfaction when in-trinsic cues are provided than when less compatible cues (extrinsic and irrelevant cues) are provided.

H1b: For experience products, customization would have a stronger salutary effect on customer satisfaction when ex-trinsic cues are provided than when less compatible cues (intrinsic and irrelevant cues) are provided.

Based on the attitude–behavior linkage,31customers who

are satisfied with a product are more likely to repurchase it. By the same token, customers who are satisfied with their configuration decisions may demonstrate a higher purchase intention toward customized products. In a study of park and recreation services, both visitor satisfaction and service quality were found to have a direct influence on the inten-tion to revisit.32In an empirical study of a Web-based

gro-cery shop in Finland, the level of consumer satisfaction pos-itively influences the purchases amount.33 Similarly, the

extent to which a customer is willing to purchase a cus-tomized product should also be influenced by the compati-bility between cues and product categories because compat-ibility is a determinant of customer satisfaction and should be highly related to purchase intention.

Although few empirical studies have assessed the influ-ence of the compatibility between cues and product cate-gories on purchase intention in the context of customization, extrinsic cues, price, brand, and store name have been re-ported to increase customer willingness to buy a calculator or stereo headset player in a standard online shopping en-vironment.34 Recent research conducted by Chu et. al

ex-amined the effects of extrinsic cues in the context of online shopping and suggests that manufacturer and retailer brands have positive effects on the purchase intention of computer monitors.35Although there is no substantial amount of

evi-dence, a potential relationship between the compatibility of cues and purchase intention is likely to exist, according to the foregoing discussion. Thus, when more compatible cues are presented along with each product attribute option, cus-tomers are likely to have higher purchase intentions toward the customized products, as they are more likely to be sat-isfied with their choices in various attributes and conse-quently show a higher willingness to buy the final products.

(4)

Moreover, as mentioned in the previous section, the iter-ative decision process of customization highlighted the im-portance of compatibility between cues and product cate-gories. The positive effect of customization on consumer purchase intention is likely to be accentuated when cues pro-vided are compatible with the product categories. Thus, the following hypotheses are advanced:

H2: The compatibility between product category and cue type has a greater salutary effect on customer purchase in-tention of a product in the customization context than in the no-customization context.

H2a: For search products, customization would have a stronger salutary effect on purchase intention when intrin-sic cues are provided than when other less compatible cues (extrinsic and irrelevant cues) are provided.

H2b: For experience products, customization would have a stronger salutary effect on purchase intention when extrin-sic cues are provided than when other less compatible cues (intrinsic and irrelevant cues) are provided.

Methods

The sample was composed of 380 participants, compris-ing both undergraduate and graduate students. Most of the sample was female (67.6%), and most (95.3%) were between 19 and 28 years old. More than 34% of the sample spent 5 to 8 hours a day online and were therefore familiar with online surfing. In terms of the frequency of shopping online, 96.3% of the respondents indicated that they had previous online shopping experience, and 35.3% of them shopped online more than six times each half year. Most respondents were online shoppers and a very suitable target audience for Web-based mass customization products.

A 2 2  3 (customization and no-customization  search and experience products intrinsic, extrinsic and ir-relevant cues) between-participants factorial design was em-ployed in this study. Three hundred eighty Taiwanese were randomly assigned to each of the 12 experimental conditions in this study. Each participant was instructed to explore one of the 12 Web sites constructed specifically for this study. Participants assigned to the customization group were in-structed to browse the product customization Web page and asked to engage in a customizing task of either an MP3 player or a bottle of perfume. Participants in the no-customization group browsed the product Web page in the manner of stan-dard online shopping. Following the paradigm in previous choice studies,36 a yoked control group design was

em-ployed. In the control group, each participant in the no-customization group was yoked to a participant in the ex-perimental condition. That is, each participant in the no-cus-tomization group was given a product that his or her coun-terpart in the experimental group configured. This method was utilized to ensure that the only difference between ex-perimental and control groups was the opportunity to cus-tomize a product. After the customization task, all partici-pants were asked to complete a questionnaire concerning important variables in this study, such as satisfaction and purchase intention.

An online retailing Web site, e-commerce.com, was fabri-cated to test the hypotheses formulated earlier. On the Web

page, participants were provided either an MP3 player or a bottle of perfume to customize. Each product had five at-tributes, and every attribute had three options from which customers could choose, making a total of 243 possible com-binations (3 3  3  3  3  243). The selection of an MP3 player and a perfume (target stimuli) as search and experi-ence products, respectively, was based on a pretest of 32 par-ticipants. The results of pretest showed no significant dif-ference in terms of customer familiarity with the target products, an MP3 player for search product and a perfume for experience product (t 0.658, ns, p  0.521).b

Further-more, from the same set of data, three key attributes were identified for each product. For the MP3 player, earphones, control panel type, and USB plug-in type were selected as the key customizable attributes. In the case of the perfume, the three key attributes were the top, middle, and base notes, which unfold one by one over time, constituting the conso-nant chord of the scent. The manipulation checks for the search/experience products showed expected and desirable results: an MP3 player was rated as a search product (t(189) 7.84; p  0.001), and the perfume was perceived as a product that can only be evaluated after purchase (t(189) 24.48; p 0.001). The intrinsic cues of each product were col-lected from various online shopping malls and also pretested to ensure sufficient realism (t(63) 6.03; p  0.001). With re-gard to extrinsic cues, three Taiwanese celebrities of equal attractiveness (F(2, 12) 0.46; p  0.64) and trustworthiness (F(2, 12) 0.12, p  .89) among the target audience, Rainie Yang (Cheng-Lin Yang), Jasmine Leong (Ching-Ju Leong), and Ariel Lin (Yi-Chen Lin), were chosen as the endorsers for both products in order to control the effect of celebrity influence. For all three attributes, each celebrity endorsed one specific option. The final group of participants, treated as the control group, was presented with irrelevant cues. Such cues were neither intrinsic, containing information re-lated to the attribute performance per se, nor extrinsic, in-cluding information about celebrity endorsements for indi-vidual options of the product attributes. Rather, they comprised information about general product usage, such as MP3 player maintenance instructions or perfume usage tips. This type of information did not contain details that could be utilized to better identify and select configuration choices. The control group was presented with such information to ensure each participant received the same amount of infor-mation so that the only difference among these groups was the type of cues they received and not the amount of infor-mation presented. Additionally, all the inforinfor-mation pre-sented was controlled at a fixed amount (F(2, 374) 0.77; p 0.46) and at a same level of argument strength (F(2, 374) 0.28; p  0.76) across different experimental condi-tions in order to remove the effect information amount.

All the 7-point Likert scale measurement items were adapted from previous studies. The satisfaction measures used by Spreng et al.37were anchored as very satisfied/very

dissatisfied, very pleased/very displeased, contented/frustrated, and delighted/terrible. For purchase intention measures, par-ticipants were asked to rate the extent to which they wanted to buy the final customized product on a three-item scale by Putrevu and Lord.38All the Cronbach’s alphas exceeded the

cutoff point of 0.7. To ensure convergent and discriminant validity of these measures, principal component analyses us-ing varimax rotation were performed. The analyses

(5)

pro-duced reasonable factor structures to support convergent va-lidity with item loadings higher than 0.6 on the appropriate dimensions, and average variance explained greater than 60%. For the test of discriminant validity, the criteria pro-posed by Gaski and Nevin39 were used. The results

sug-gested discriminant validity was proper because the corre-lations among measures were lower than alpha coefficients of themselves.

Results

A multivariate analysis of variance (MANOVA) was used to assess the interactive effects among customization, cue types, and product categories on two dependent variables: satisfaction and purchase intention. There were significant multivariate main effects for customization, F(2, 367) 21.301, Wilks’s   0.896, p  0.001, and cues, F(4, 734)  2.972, Wilks’s   0.968, p  0.05. The customization condi-tions resulted in more favorable customer responses than no-customization conditions, F(2, 377) 19.18, Wilks’s   0.908, p 0.001; satisfaction: MC 4.71 vs. MN 4.14, F(1,

378) 20.83, p  0.001; purchase intention: MC 4.52 vs.

MN 3.70, F(1, 378)  37.43, p  0.001. The main effect of

cues is also significant, which suggest cues had different ef-fects on dependent variables, F(4, 752) 2.41, Wilks’s   0.975, p 0.05. However, the main effects were qualified by higher-order interactions.

A two-way multivariate interaction effect was found for customization and cues, F(4, 734) 3.631, Wilks’s   0.962, p 0.01. The interaction between customization and cues was significant for satisfaction and purchase intention. MANOVA contrast tests revealed that cues induced more fa-vorable responses in the customization group than did no-customization group, F(4, 746) 3.417, Wilks’s   0.964, p 0.01. In the customization condition, participants pressed higher levels of satisfaction when intrinsic or ex-trinsic cues were provided than when irrelevant cues were provided, satisfaction: MINTRINSIC 4.97, MEXTRINSIC 4.92,

MIRRELEVANT 4.23, F(2, 374)  5.50, p  0.01. For

depen-dent variable purchase intention, participants expressed greater purchase intention when intrinsic or extrinsic cues were provided than when irrelevant cues were provided, purchase intention: MINTRINSIC 4.82, MEXTRINSIC 4.79,

MIRRELEVANT 3.94, F(2, 374)  5.52, p  0.01. This two-way

interaction effect, however, was qualified by the three-way interaction, which suggested that the definition of “compat-ible cues” is contingent upon the product category the cues were associated with.

Another two-way interaction between cues and product categories was also found, F(4, 734) 5.04, Wilks’s   0.947, p 0.01. MANOVA contrast tests further revealed that the effect cues was contingent upon product category in both customization and non-customization groups, F(4, 746) 4.593, Wilks’s   0.953, p  0.001. For MP3, regardless whether participants had the opportunity to customize, when they were provided with intrinsic cues, they expressed higher satisfaction and purchase intention than when other cues were provided, satisfaction: MINTRINSIC 5.02, M EX-TRINSIC 4.18, MIRRELEVANT 4.27, F(2, 187)  8.75, p 

0.001; purchase intention: MINTRINSIC 4.69, MEXTRINSIC

4.08, MIRRELEVANT 3.92, F(2, 187)  5.97, p  0.01. On the

other hand, extrinsic cues resulted in higher satisfaction with

perfume than did intrinsic or irrelevant cues, satisfaction: MINTRINSIC 4.22, MEXTRINSIC 4.69, MIRRELEVANT 4.17,

F(2, 187) 3.75, p  0.05. For purchase intention, extrinsic cues was able to achieve a higher level of customer satisfac-tion than irrelevant cues, purchase intensatisfac-tion: MEXTRINSIC

4.26, MIRRELEVANT 3.73, p  0.05. Nevertheless, the

two-way interaction mentioned above was qualified by the three-way interaction, suggesting that the augmented satisfaction due to compatible cues prevailed for customization groups rather than non-customization groups.

A three-way interaction among customization, cues, and product categories was found to be significant, F(4, 734) 2.628, Wilks’s   0.972, p  0.05. As a consequence, the data were analyzed with MANOVA contrasts40to explore the

na-ture of these interactions. H1 and H2 predict that compati-bility between cues and product categories have a stronger effect on customer satisfaction and purchase intention of a product in the customization context than in the no-cus-tomization context. To examine the mean difference of both dependent variables between customization and no-cus-tomization groups, planned contrast tests were performed. Supporting H1 and H2, the results of MANOVA contrasts revealed that for the search product, participants in the cus-tomization group had significantly more favorable ratings of satisfaction and purchase intention, F(4, 182) 4.905, Wilks’s   0.815, p  0.001. In particular, the augmented satisfac-tion with and higher purchase intensatisfac-tion of the product due to customization with compatible cues presented (i.e., in-trinsic cues) were larger than that resulting from cus-tomization with less well-matched cues (i.e., extrinsic and ir-relevant cues), difference in satisfaction: MINTRINSIC 1.45,

MEXTRINSIC 0.53, MIRRELEVANT 0.17, F(2, 92)  7.482,

p 0.001; difference in purchase intention: MINTRINSIC

1.83, MEXTRINSIC 0.55, MIRRELEVANT 0.16, F(2, 92) 

7.732, p 0.001. For the experience product, the same pat-tern was observed, F(4, 182) 4.492, Wilks’s   0.828, p  0.01. Elevated satisfaction with and higher purchase inten-tion of experience products due to customizainten-tion were higher when extrinsic cues were provided than when other less compatible cues were provided, difference in satisfac-tion: MINTRINSIC 0.58, MEXTRINSIC 1.41, MIRRELEVANT

0.20, F(2, 92) 5.277, p  0.01; difference in purchase inten-tion: MINTRINSIC 0.69, MEXTRINSIC 1.92, MIRRELEVANT

0.28, F(2, 92) 9.457, p  0.001. To sum up, all the hypothe-ses formulated in this study section were supported. Conclusions and Discussions

The objective of this study is to contribute empirical evi-dence to mass customization literature, which at present pro-vides little or no insight concerning how customers react within the customization process.12The results showed that

customization can lead to a higher level of customer satis-faction when cues compatible with product categories were provided. More specifically, for search products, customiza-tion achieved the highest augmentacustomiza-tion of purchase inten-tions and satisfaction with the final products when intrinsic cues were provided. On the other hand, for experience prod-ucts, customization led to the most positive responses on both dependent variables when extrinsic cues are proffered. The findings that compatible cues are more effective in en-hancing customer satisfaction and purchase intention are

(6)

consistent with the prediction of cue utilization theory.18

While evaluating search products, customers tend to depend on intrinsic cues as quality indicators because the linkage be-tween quality and intrinsic cues is more salient than for ex-trinsic cues.8For instance, storage size is more important as

a purchasing criterion than an endorsement when a flash drive is involved. Past research regarding customer infor-mation search behaviors in cyberspace9 reveals that when

making purchase decisions for search products, customers tend to depend more on so-called hard data (color, style, specification, size, model, and function) provided by firms. Therefore, firms should include more information of this kind in order to facilitate product searches.16On the other

hand, for experience products, intrinsic cues, such as de-scriptions of attribute performance derived directly from the products, are more difficult for customers to use and have lower predictive and confidence values. Therefore, in this sit-uation, customers tend to rely more on extrinsic cues for as-sistance, such as manufacturer reputation27and

recommen-dations from other customers.9 This is consistent with

previous findings9suggesting that while engaging in online

information searches for experience products, customers perceive so-called soft data (customer opinion, rating, and discussion) to be more important than information provided by firms.9The results also appear to be in accordance with

the findings of Jain and Posavac, which aver that endorse-ments from credible celebrities are particularly effective in augmenting the impact of advertising claims for experience qualities.41 Celebrity endorsements as extrinsic cues create

bonds between themselves and product quality, providing more heuristic evaluations for consumers with which to sup-port their configuration choices. For example, customers may judge a particular scent by the attractiveness of the celebrity endorsing it and thereby infer the scent is more pleasant and more suitable for them.

Although not explicitly stated in a hypothesis, it was found that customization significantly enhanced both cus-tomer satisfaction and purchase intention across all condi-tions of different cues and product categories. In other words, the phenomenon of mass confusion, which describes a situation in which customers are intimidated and dissatis-fied with the complexity of customizing products and would rather choose to buy standard products, did not occur in our experiment. This is conceivable, since the customizing tasks in our study are not as complicated as the ones in previous studies, and mass confusion mainly results from the high level of perceived complexity customers experience in diffi-cult customization processes. In our experiment, only five at-tributes of an MP3 player and a bottle of perfume were de-signed to be customized. Such customizing experiences may not be as complicated as the customizing tasks mentioned in the previous studies, such as customizing 18 different at-tributes of sofas (e.g., back shape, stuffing amount, and cov-ering materials), among which some attributes are fairly dif-ficult to decide on.3 The significant bolstering effect of

compatible cues on customer responses in the absence of mass confusion implies the robustness of cues provision ef-fect. If cues can enhance customer responses when decisions are not too overwhelming, their augmenting effect on posi-tive consumer responses may be greater when customers are in need of assistance.

The significant two-way interaction between cues and cus-tomization suggests cue provision is associated with a higher level of customer satisfaction and purchase intention when customization is involved. This corroborates the notions that the provision of product information can improve positive experiences42and elicit higher customer satisfaction43in

on-line shopping environments, and this enhancement of posi-tive customer responses is contingent upon the complexity of a shopping task. As information provision contributes to save time, enhance customer knowledge, and therefore achieve higher customer satisfaction,43this effect is likely to

be of greater magnitude when it is more difficult for con-sumers to make the purchase decisions. For instance, a study in e-commerce by Swaminathan demonstrates that decision aids such as recommendation agents can enhance decision quality to a greater extent when the number of product at-tributes or alternatives increases.44 Similarly, it can be

ex-pected that the effects of cues would be amplified in a cus-tomization context where task complexity may increase due to a greater number of decisions. Hence, it will be particu-larly beneficial for firms offering customization to provide decision facilitating information because customization tends to create a shopping environment in which consumers are faced with a large number of options and/or difficult configuration choices.

In summary, due to their high predictive and confidence values, compatible cues can better facilitate customer pref-erence identification processes in a customization task, and as a result, reduce perceived complexity to a greater degree3

and achieve a higher level of satisfaction and purchase in-tention.

Managerial implications

Consistent with the existing perspective on mass cus-tomization strategies,1,2,45 the results of this study suggest

that customers would show more favorable responses to cus-tomized products if appropriate assistance is available. Ac-cordingly, marketers could employ strategies more in line with the product category being marketed, while imple-menting mass customization in order to create more positive customer responses. Providing customers with compatible cues during customization processes can perform such a function because cues provide customers who have less well-constructed preferences with a reason to make their deci-sions10and enhance customer evaluations in the

customiza-tion process.11 For example, for a search product, the

interactive systems for mass customization could be built with itemized cues that provide detailed specifications and descriptions of attribute options, thus allowing customers to compare each option effortlessly and then to find the best option. Furthermore, interactive 3D product presentations enabling visual inspection of products online could also be implemented in order to allow customers the opportunity to examine search attributes, such as color and exterior design. However, in the case of experience products, companies would benefit more from the employment of extrinsic cues such as celebrity endorsers, expert recommendations, and a list of best-selling options. Such cues would provide cus-tomers with a customizing experience characterized by greater levels of confidence and ease.

(7)

Limitations and future research directions

While laboratory settings in this study provided the facil-ity to eliminate the influence of individual differences in brand affinity and Web site familiarity, field studies that can observe real customer purchase responses toward cus-tomized offers in a natural setting are needed. Another lim-itation of this study is the issue of external validity. While youths ranging from age 19 to 35 can represent a great po-tential target for Web-based customized products, as a sam-ple group they might not be representative of the whole pop-ulation and, consequently, these results need to be interpreted with caution.

Additional studies can be conducted to extend this re-search by including credence products and testing whether different types of extrinsic cues are differentially effective across different product categories. Moreover, due to the na-ture of experiment, in this study, only a limited number of product attributes (five attributes for every product) and at-tribute options (three options per atat-tribute) were investi-gated. Different levels of the variables may have an impact on customer responses. For example, when the number of product attributes or attribute options increases, would the provision of compatible cues enhance customer satisfaction and purchase intention to a greater degree, or would it de-crease these variables due to customer annoyance with rep-etition? This would be an interesting research question to an-swer.

Notes

a. Credence products, a third class identified by Darby and Karni,46describes products that can be evaluated only

af-ter they have been used for a period of time or evaluated accurately after use because of the lack of consumer ex-pertise. While it can be of value to explore the effective cues for this third product category, no specific predic-tions can be obtained following the logic of cue utiliza-tion theory. According to this theory, for both experience and credence products, customers may similarly rely on extrinsic cues because the intrinsic cues may be low in ei-ther predictive or confidence value or both. Therefore, no significant difference in their reliance on extrinsic cues was expected. While we recognize that customers may de-pend on different kinds of extrinsic cues (e.g., celebrity endorser or expert endorser) for different product cate-gories (e.g., celebrity endorser for experience products and expert endorser for credence products), customers still rely on extrinsic cues in general though different en-dorsers are involved. Since we are interested in the effec-tiveness of different cue types (i.e., intrinsic vs. extrinsic) rather than the finer classification within each type, the credence product was not chosen as our target in this study. Discussions about such possible further research efforts, however, are included in the Conclusions and Dis-cussions section of this paper.

b. Because the confidence value of a cue may fluctuate ac-cording to customer knowledge, customers’ familiarity with both products (MP3 player and perfume) was as-sessed in a pretest. The results of the pretest suggest that participants showed no significant difference in familiar-ity with the products. Moreover, as expert customers such

as engineers and fragrance makers are very rare among our sample, the possibility of this study being biased be-cause of customer familiarity should be negligible. Disclosure Statement

The authors have no conflict of interest. References

1. Kotler P. From mass marketing to mass customization. Plan-ning Review 1989; 17:10–3.

2. Pine BJ, II. Mass customizing products and services. Plan-ning Review 1993; 21:6–13.

3. Huffman C, Kahn BE. Variety for sale: mass customization or mass confusion? Journal of Retailing 1998; 74:491–513. 4. Alba J, Lynch J, Weitz B, et al. Interactive home shopping:

consumer, retailer, and manufacturer incentives to partici-pate in electronic marketplaces. Journal of Marketing 1997; 61:38–53.

5. Gerald H, Valerie T. Consumer decision making in online shopping environments: the effects of interactive decision aids. Marketing Science 2000; 19:4–21.

6. Nelson P. Information and consumer behavior. Journal of Political Economy 1970; 78:311–29.

7. Nelson P. Advertising as information. Journal of Political Economy 1974; 82:729–54.

8. Zeithaml VA. Consumer perceptions of price, quality, and value: a means. Journal of Marketing 1988; 52:2–22. 9. Bei L-T, Chen EYI. Consumers’ online information search

be-havior and the phenomenon of search vs. experience products. Journal of Family & Economic Issues 2004; 25:449–67. 10. Simonson I. Choice based on reasons: the case of attraction

and compromise effects. Journal of Consumer Research 1989; 16:158–74.

11. Simonson I. Determinants of customers’ responses to cus-tomized offers: conceptual framework and research propo-sitions. Journal of Marketing 2005; 69:32–45.

12. Piller F, Schubert P, Koch M. et al. (2004) From mass cus-tomization to collaborative customer co-design. In Proceed-ings of the European Conference on Information Systems, Turku, Finland.

13. Donnavieve S, Satya M, Sivakumar K. Online peer and ed-itorial recommendations, trust, and choice in virtual mar-kets. Journal of Interactive Marketing 2005; 19:15.

14. Hsieh Y-C, Chiu H-C, Chiang M-Y. Maintaining a commit-ted online customer: a study across search-experience-cre-dence products. Journal of Retailing 2005; 81:75–82. 15. Ford GT, Smith DB, Swasy JL. Consumer skepticism of

ad-vertising claims: testing hypotheses from economics of in-formation. Journal of Consumer Research 1990; 16:433–41. 16. Franke GR, Huhmann BA, Mothersbaugh DL. Information content and consumer readership of print ads: a compari-son of search and experience products. Academy of Mar-keting Science Journal 2004; 32:20–31.

17. Mitra K, Reiss MC, Capella LM. An examination of per-ceived risk, information search and behavioral intentions in search, experience and credence services. Journal of Services Marketing 1999; 13:208–28.

18. Olson JC. (1972) Cue utilization in the quality perception process: a cognitive model and an empirical test. In Venkate-san M, ed. Proceedings of the Third Annual Conference of the Association for Consumer Research. Iowa City: Association for Consumer Research, pp. 167–79.

(8)

19. Richardson PS, Dick AS, Jain AK. Extrinsic and intrinsic cue effects on perceptions of store brand quality. Journal of Mar-keting 1994; 58:28–36.

20. Girard T, Silverblatt R, Korgaonkar P. Influence of product class on preference for shopping on the Internet. Journal of Computer-Mediated Communication 2002; 8.

21. Korgaonkar P, Silverblatt R, Girard T. Online retailing, prod-uct classifications, and consumer preferences. Internet Re-search 2006; 16:267–88.

22. Chiang K-P, Dholakia RR. Factors driving consumer inten-tion to shop online: an empirical investigainten-tion. Journal of Consumer Psychology 2003; 13:177–83.

23. Johansson JK, Douglas SP, Nonaka I. Assessing the impact of country of origin on product evaluations: a new method-ological perspective. Journal of Marketing Research 1985; 22:388.

24. Aqueveque C. Extrinsic cues and perceived risk: the influ-ence of consumption situation. Journal of Consumer Mar-keting 2006; 23:237–47.

25. Herr PM, Kardes FR, Kim J. Effects of word-of-mouth and product-attribute information on persuasion: an accessibil-ity-diagnosticity perspective. Journal of Consumer Research 1991; 17:454.

26. Shimp TA, Bearden WO. Warranty and other extrinsic cue effects on consumers’ risk perceptions. Journal of Consumer Research 1982; 9:38–46.

27. Bearden WO, Shimp TA. The use of extrinsic cues to facili-tate product adoption. Journal of Marketing Research 1982; 19:229–39.

28. Klein LR, Ford GT. Consumer search for information in the digital age: an empirical study of prepurchase search for au-tomobiles. Journal of Interactive Marketing 2003;17:29. 29. Szymanski DM, Hise RT. E-satisfaction: an initial

examina-tion. Journal of Retailing 2000; 76:309–22.

30. Kohli R, Devaraj S, Mahmood A. Understanding determi-nants of online consumer satisfaction: a decision process per-spective. Journal of Management Information Systems 2004; 21:115–35.

31. Oliver RL. A cognitive model of the antecedents and conse-quences of satisfaction decisions. Journal of Marketing Re-search 1980; 17:460–9.

32. Tian-Cole S, Crompton JL, Willson VL. An empirical inves-tigation of the relationships between service quality, satis-faction and behavioral intentions among visitors to a wildlife refuge. Journal of Leisure Research 2002; 34:1.

33. Koivumäki T. Customer satisfaction and purchasing behav-iour in a Web-based shopping environment. Electronic Mar-kets 2001; 11:186–92.

34. Dodds WB, Monroe KB, Grewal D. Effects of price, brand, and store information on buyers’ product evaluations. Jour-nal of Marketing Research 1991; 28:307–19.

35. Chu W, Choi B, Song MR. The role of online retailer brand and infomediary reputation in increasing consumer pur-chase intention. International Journal of Electronic Com-merce 2005; 9:115–27.

36. Botti S, Iyengar SS. The psychological pleasure and pain of choosing: when people prefer choosing at the cost of subse-quent outcome satisfaction. Journal of Personality and So-cial Psychology 2004; 87:312–26.

37. Spreng RA, MacKenzie SB, Olshavsky RW. A reexamination of the determinants of consumer satisfaction. Journal of Mar-keting 1996; 60:15–32.

38. Putrevu S, Lord KR. Comparative and noncomparative ad-vertising: attitudinal effects under cognitive and affective in-volvement conditions. Journal of Advertising 1994; 23:77–9. 39. Gaski JF, Nevin JR. The differential effects of exercised and unexercised power sources in a marketing channel. Journal of Marketing Research 1985; 22:130–42.

40. Kirmani A, Sood S, Bridges S. The ownership effect in con-sumer responses to brand line stretches. Journal of Market-ing 1999; 63:88–101.

41. Jain SP, Posavac SS. Prepurchase attribute verifiability, source credibility, and persuasion. Journal of Consumer Psy-chology 2001; 11:169–80.

42. Cook DL, Coupey E. Consumer behavior and unresolved regulatory issues in electronic marketing. Journal of Busi-ness Research 1998; 41:231–8.

43. Ballantine PW. Effects of interactivity and product informa-tion on consumer satisfacinforma-tion in an online retail setting. In-ternational Journal of Retail & Distribution Management 2005; 33:461–71.

44. Swaminathan V. The impact of recommendation agents on consumer evaluation and choice: the moderating role of cat-egory risk, product complexity, and consumer knowledge. Journal of Consumer Psychology 2003; 13:93–101.

45. Wind J, Rangaswamy A. Customerization: the next revolu-tion in mass customizarevolu-tion. Journal of Interactive Marketing 2001; 15:13–32.

46. Darby MR, Karni E. Free competition and the optimal amount of fraud. Journal of Law and Economics 1973; 16:67–88.

Address reprint requests to: Dr. Chia-Chi Chang Department of Management Science National Chiao Tung University 1001 Ta Hsueh Road Hsinchu, Taiwan 300 Republic of China E-mail: chi@mail.nctu.edu.tw

(9)

1. Alessio Trentin, Elisa Perin, Cipriano Forza. 2014. Increasing the consumer-perceived benefits of a mass-customization experience

through sales-configurator capabilities. Computers in Industry 65:4, 693-705. [

CrossRef

]

2. Kim Piew Lai, Siong Choy Chong, Hishamuddin Bin Ismail, David Yoon Kin Tong. 2014. An explorative study of shopper-based

salient e-servicescape attributes: A Means-End Chain approach. International Journal of Information Management . [

CrossRef

]

3. Wei-Tsong Wang, Chia-Cheng Lu. 2013. Determinants of success for online insurance websites: The contributions from system

characteristics, product complexity, and trust. Journal of Organizational Computing and Electronic Commerce 131122223027006.

[

CrossRef

]

4. David M. Hunt, Scott K. Radford, Kenneth R. Evans. 2013. Individual differences in consumer value for mass customized products.

Journal of Consumer Behaviour 12:4, 327-336. [

CrossRef

]

5. Shintaro Okazaki, Natalia Rubio, Sara Campo. 2013. Do Online Gossipers Promote Brands?. Cyberpsychology, Behavior, and Social

Networking 16:2, 100-107. [

Abstract

] [

Full Text HTML

] [

Full Text PDF

] [

Full Text PDF with Links

]

6. Flavio S. Fogliatto, Giovani J.C. da Silveira, Denis Borenstein. 2012. The mass customization decade: An updated review of the

literature. International Journal of Production Economics 138:1, 14-25. [

CrossRef

]

參考文獻

相關文件

This shows that service quality, perceived value, DM advertising, customer satisfaction and loyalty have become important issues on business management.. Therefore, the

The analytic results show that image has positive effect on customer expectation and customer loyalty; customer expectation has positive effect on perceived quality; perceived

(2000), “Assessing the Effects of Quality, Value, and Customer Satisfaction on Consumer Behavioral Intentions in Service Environments,” Journal of Retailing, Vol. (2001),

Our major findings are: (1)The sex of consumers have significant effects on reverse product design but the remaining factors.(2)The mar- riage status of consumers

Research of Cultural and Creative Product’s Effect on Customer Value and Customer Attitude- - -Take - Taiwan “MUJI” Product of RYOHIN KEIKAKU..

Through literatures relevant to service quality, service value, customer satisfaction and customer loyalty, this research conducts study on the five aspects of the theme

The study was based on the ECSI model by Martensen et al., (2000), combined with customer inertia as a mediator in the hope of establishing a customer satisfaction model so as

This study aims to explore whether the service quality and customer satisfaction have a positive impact on the organizational performance of the services and whether the