© Mary Ann Liebert, Inc. DOI: 10.1089/cpb.2008.0170
The Interplay between Customer Participation
and Difficulty of Design Examples in the Online Designing
Process and Its Effect on Customer Satisfaction:
Mediational Analyses
Chia-Chi Chang, Ph.D., Hui-Yun Chen, M.B.A., and I-Chiang Huang, M.B.A.
Abstract
In the current consumer-centric economy, consumers increasingly desire the opportunity to design their own
products in order to express more effectively their self-image. Mass customization, based on efficient and
flex-ible modulization designs, has provided individualized products to satisfy this desire. This work presents an
experiment employed to demonstrate that customer participation leads to higher satisfaction. Specifically, the
increment in customer satisfaction due to participation is greater when an easy example is provided than when
either no example or a difficult one is provided. Additionally, self-congruity plays a mediating role on the
cus-tomer participation–satisfaction relationship, and this mediating effect varies across different levels of the
de-sign example provided in the dede-sign process. When an easy dede-sign example is present, customer participation
has a direct effect on satisfaction, in addition to the indirect effect of self-congruity. When a difficult example
is provided, customer participation does not have incremental effects on either self-congruity or customer
sat-isfaction. Finally, when no design example is shown to customers, contrary to our expectation, participation
still enhances customer satisfaction due to an increased sense of self-congruity.
147 Introduction
I
N THE AGE OF E-COMMERCE, technological advances offerfirms the means with which to strike a balance between opportunities for customer selection and product cus-tomization and the need to keep production costs down.1An
increasing number of companies in different industries, such as Dell and Capital One, have successfully implemented cus-tomization. By allowing customers to configure product at-tributes such as computer components and interest rates, firms proffer customers with greater economic and func-tional benefits.2,3Besides utilitarian benefits, customers
re-ceive hedonic benefits from both interactive design tools and customized products. Interactive design tools offer cus-tomers opportunities to enjoy interactive designing experi-ences, and customized products also provide customers with opportunities to express their ideas and tastes to greater de-grees.4For example, the project Mi Adidas allows consumers
to order unique footwear by specifying their preferences for colors and styles as well as the opportunity to enjoy the novel experience of interacting with virtual 3D technologies. How-ever, little effort has been made to investigate empirically
whether or not online interactive design tools can effectively assist customers to create satisfactory and individual design patterns. Among the few exceptions, one study focuses on the extent to which customers participate in using online in-teractive design tools and demonstrates that for customized T-shirts, the greater the extent of customer participation in the designing process, the greater the satisfaction achieved.5
Another related empirical study suggests that the interactive design tool has the advantage of assisting customers in se-lecting components of watches and of giving visual feedback on watch design offers.6 Compared to standard watches,
such self-designed watches through the interactive design tool can increase customer willingness to pay by up to 100%.6
Despite the positive findings, some practitioners and re-searchers have maintained that the effect of customer par-ticipation may be contingent upon perceived complexity or difficulty.8–11At times, the large number of product features
and options could increase the perceived difficulty of tomizing a product or service, particularly for those cus-tomers who lack the necessary expertise or well-defined pref-erences.7 In order to increase customer evaluation and
acceptance of customized products, it is necessary to
ically study strategies marketers can employ to reduce the difficulty of the customizing process and to attain more pos-itive customer responses.
In addition, the increased satisfaction resulted from tomer participation can be attributed to the fact that cus-tomers can gain a higher sense of self-congruity by design-ing their own individual products. The investigation of self-congruity in this study is of particular relevance because whether a product can reflect an individual’s self-image usu-ally influences his or her purchasing behavior.8Moreover, it
has been suggested that customers consider the enhance-ment of individuality an important factor that motivates them to participate in apparel customization.4Customized
products can better enhance the self-congruity between a product and customer self-image, which may result in higher customer satisfaction due to the opportunity to design prod-ucts that reflect personal image and style.4
This study conducted an experiment to address the fol-lowing research questions: (a) Is customer satisfaction en-hanced when customers are allowed to participate in the de-signing process by using online interactive design tools? (b) If customer participation is found to increase customer sat-isfaction, does the increment vary with levels of difficulty of the design example provided? (c) Does customer participa-tion increase satisfacparticipa-tion via augmented self-congruity? (d) Is this mediating effect of self-congruity the same for design examples of varying levels of difficulty?
Most previous studies concerning customer participation focused on discovering the augmenting effect of customer participation on satisfaction;5,9,10however, evidence in
sup-port of the mediating effect of self-congruity has not been documented. Therefore, this study proceeds from the ex-pectation that variations in the design example difficulty may affect the customer participation–satisfaction relation-ship. In addition, the mediating effect of self-congruity may also be contingent upon the difficulty of design examples. For this purpose, customer satisfaction and perceived self-congruity with the customized T-shirts were measured across levels of design example difficulty in order to address the questions posited above.
Literature Review
Benefits associated with customer participation
Several empirical studies of mass customization revealed that customers who participate in the design processes show a higher level of satisfaction and willingness to pay premium prices.3,5,6Through the contingent valuation of method and
a Vickrey auction, Franke and Piller assessed the price pre-mium in mass customization and discovered that watches designed by customers themselves surpass standard watches in terms of customer willingness to pay by almost 100%.6
Similar findings were obtained in Schreier’s study, which shows that by using design toolkits that allow instant visual feedback on customized products, such as cell phone covers, T-shirts, and scarves, customers show a greater willingness to pay than they show for standard ones, ranging from 106% (scarves), to 113% (T-shirts), to 207% (cell phone covers).3
The value increment of self-designed products may be at-tributed to both the enhanced utilitarian and hedonic bene-fits associated with the final products.3Additionally, the
con-trol customers obtain from being able to master a task may
contribute to the elevated product value perceived by them. Utilitarian benefits are derived from the performance, func-tionality, convenience, or efficiency of a product.11Because
customized products are at least partially designed by cus-tomers, they are expected to meet customer needs more pre-cisely. As a result, customers can derive more utilitarian ben-efits from customized products than from standard ones.3
For example, customers can configure a Dell computer by selecting processor, memory, storage, and other equipment to meet their needs more precisely.
In addition to the utilitarian benefits, another drive for cus-tomers to participate in the designing process is the height-ened hedonic benefit from the individuality of the cus-tomized product and the excitement gained from the designing process.4Hedonic benefits are defined as the
plea-sure-oriented aspects of shopping, such as fun, enjoyment, fantasy, sensory stimulation, and festive experiences.12,13
Customized products can deliver more symbolic values to customers and better satisfy customer needs, such as the ex-pression of beliefs, demonstration of social status, and sense of uniqueness.11,14And the designing process itself entails
exciting experiences for customers, such as fulfilling a fan-tasy to become a designer.4
In addition to the utilitarian and hedonic benefits associ-ated with customized products, the sense of ownership and responsibility15 is enhanced because customers gain more
control in the customization process. Behavioral control, de-fined as “availability of a response which may directly in-fluence or modify the objective characteristics of an event,”16
over the interactive design tools is likely to result in higher evaluations of outcomes17 and greater behavioral
inten-tions.18An experiment on pizza customization also
demon-strates that customers who participate in the production of pizzas through their choice of ingredients demonstrate greater behavioral control, which leads to higher product evaluations.17Similarly, Kamali and Loker found that
cus-tomers who are allowed to mix and match T-shirt compo-nents such as neckline, sleeve, color, and patterns in an on-line store experience higher levels of satisfaction.5 It is
reasonable to expect that when customers participate in the design process and have a direct influence on final products, they are likely to experience higher levels of satisfaction.
H1: In general, consumers who participate in designing their own products will be more satisfied with the product than those who do not participate in the designing process.
The moderating effect of the difficulty of a design example in customization processes
Although, generally speaking, customers who can partic-ipate in the design process demonstrate higher levels of sat-isfaction, at times customers can also be confused by the di-versity and complexity of products or elements offered in customization processes. While some customers enjoy the process of interactive customization, others may experience impatience, frustration, and confusion due to either low need for cognition,19 deficient involvement,20 unclear
prefer-ences,7or inability to perform the customizing task. As a
con-sequence, the potentially positive effect of customer partici-pation may be limited if customers perceive the customization task to be too complex for them to perform successfully.
Huffman and Kahn use the term mass confusion to delin-eate the phenomenon mentioned above.7 They assert that
mass confusion results mainly from perceived complexity and can be effectively reduced if tactics in dealing with complicated customization processes are proffered.7For
ex-ample, presenting information about hotels or sofas in an at-tribute-based format can assist customers in learning within-attribute preferences, and thus their perceived complexity of choice sets can be reduced.7Providing materials that can
as-sist participants in walking through difficult customization processes may also be an effective way to alleviate this prob-lem and to increase customer enjoyment. For example, in a case study of a collaborative residential development proj-ect in which community residents were encouraged to par-ticipate in the design process through a Web site, a system offering timely feedback from the architect about the feasi-bility of construction details (e.g., layout, safety, and public spaces) assisted residents without professional knowledge to still enjoy the designing process.21As a result, a higher level
of community satisfaction was achieved.21In summary, it is
expected that when customers are given assistance in rec-ognizing their affinities for individualized designs, the pos-itive effects of customer participation are more likely to be brought into full play.
Sometimes, firms present design examples to customers in order to better facilitate the process of customer partici-pation. Examples may help customers to identify their pref-erences in designing styles more effectively, mitigate the frustration induced by complicated designing processes, and ultimately enhance customer satisfaction with the outcome. However, many companies tend to provide attractive but rather complicated design examples to arrest the attention of customers, which might result in customer frustration and lowered purchase intension. In accordance with social cog-nitive theory, people who believe that they possess the nec-essary knowledge and skills to perform a particular task will feel more comfortable to engage in that task22and are more
likely to take part in those tasks, since they may be afforded with mastery experiences. On the other hand, those who ceive themselves as less efficacious will be less willing to per-form the task and feel less satisfied with any outcome. For example, if a student is shown an average student record, he is more likely to be satisfied or confident with his own run-ning time. In contrast, if a seemingly unachievable standard has been set (e.g., a race record set by an Olympic gold medalist), the same person’s satisfaction may be reduced. Hence, when customers are provided with design examples exceeding their design ability, they are more likely to become dissatisfied with their own work because they feel they are less confident in creating equally attractive designs. When no design examples are presented, customers may feel there is no direction and perceive the design task as too difficult. As a result, in situations with no design examples provided, customers are less likely to feel satisfied with their own de-signs.
In contrast, easy design examples may provide customers with direction about what to do and, more importantly, con-vey the message that the creation of designs is realistically achievable. Thus, in contrast to customers who receive dif-ficult or no design examples, those who are faced with easy design examples are more likely to enjoy the customization process and show higher levels of satisfaction with their
fi-nal products. Accordingly, in this study it is argued that not all design examples are equally effective in augmenting con-sumer satisfaction.
Based on the preceding discussion, the second hypothesis can be formulated as follows:
H2: The positive effect of customer participation on satis-faction will be enhanced when a customer is provided with an easy design example than when provided with a diffi-cult design example or no design example.
The role of self-congruity
Self-congruity has been defined as the match between product personality images and customer perceived self-im-ages.8Products considered to have “personality images” can
be described as lively, youthful, faithful, or conservative. In order to maintain their self-consistency, customers are more likely to choose products of high self-congruity because these products reflect their self-images.8Previous research on
ad-vertising has also maintained that for value-expressive prod-ucts, advertising should emphasize the self-congruity that exists between customers and products in order to encour-age positive attitudes in potential customers.23In the context
of customization, self-congruity may have an even greater impact on customer responses, since the purpose of cus-tomization for many customers is to express their values, tastes, and styles through their individualized products.
When customers participate in the design process, the op-portunity to be involved in creating individualized designs is likely to increase customer perception of self-congruity. For instance, customers are able to enhance their individu-ality by creating unique fashion products for self-expression such as shoes, pants, dresses, and skirts.4This augmented
self-congruity then increases customer satisfaction with the customized products. Hence, it is reasonable to conjecture that customer participation may result in higher levels of sat-isfaction as a result of increased self-congruity. In other words, it is expected that self-congruity will mediate the re-lationship between customer participation and satisfaction. Hence, the following hypothesis can be formulated.
H3: The relationship between customer participation and satisfaction is mediated by the self-congruity perceived by customers.
Our second hypothesis states that the augmented satis-faction resulting from customer participation is likely to be contingent upon the level of difficulty in the design exam-ple. Since self-congruity mediates the participation–satisfac-tion link, it is reasonable to postulate that the mediating ef-fect of self-congruity is also likely to vary due to the difficulty of design examples. According to social cognitive theory,24
consumers prefer and enjoy more behaviors they are capa-ble of performing. Compared with a situation in which a dif-ficult example or no example is present, when an easy ex-ample is provided, it is easier for consumers to perceive that they have mastered the task. Under such circumstances, self-serving bias will tend to lead customers to take credit for the relatively more successful final customized products. Fur-thermore, they perceive such products as more self-congru-ent due to the biased attribution of the credit for success to themselves and of failure to external causes.20 Bendapudi
and Leone empirically demonstrated that when the out-comes are better than expected, customers who participate in the production processes of products or services will at-tribute to themselves greater credit for successful out-comes.20On the other hand, when the outcomes are worse
than expected, even though customers have participated to create the products, they are less likely to take the responsi-bility for the outcomes and to consider the final products as unrepresentative of themselves. In other words, when cus-tomers are provided with an easy design example, they are more likely to perceive themselves as capable of creating sat-isfactory designs and are more likely to consider the designs as self-expressive. In such a case, they will be more likely to sense the final customized product as self-congruent, and this will eventually lead to a higher level of satisfaction. Thus, the following hypotheses have been developed:
H4: The mediating effect of self-congruity on the customer participation–satisfaction relationship will vary contingent upon the difficulty of the design example provided. H4a: Self-congruity will mediate the customer participation and satisfaction link only when an easy design example is provided.
Methods Participants
One hundred eighty students (115 male, 65 female) par-ticipated in this study. Their ages ranged from 18 to 25 years. Respondents were provided with the chance to win ap-proximately US$30 as an incentive for their participation.
Experimental design
A 2 3 between-participants factorial experiment (n 30 for each cell) was conducted, including two levels of cus-tomer participation (participation and no participation), and three levels of perceived difficulty (no design example, easy design example, and difficult design example).
Stimulus and procedure
Previous researchers studied mass customization in the fashion industry,5,25an industry in which it is flourishing.
Therefore, the T-shirt was selected as the target stimulus for customization in this study. In order to select the proper tar-get stimulus, a pretest of 115 participants was conducted to choose two design examples (easy vs. difficult) out of 12 de-signs from an online gallery. The results of the pretest showed no significant difference in terms of attractiveness (t 0.271, ns, p 0.506), but a significant difference in per-ceived difficulty (t 6.761, p 0.001) between the two se-lected design examples was found. The results of this pretest indicated that the manipulations were effective.26
Participants were first asked to self-report their ability to design a T-shirt, and then they were randomly assigned to the experimental group (participation) and the control group (no participation). Participants in the experimental group were told that they could use the Flash drawing tool1that
had been recently launched by an online T-shirt store to de-sign a T-shirt of their own. Those in the control group were told that an online T-shirt store recently introduced a design
tool, which they were welcome to try out. However, after their trial use, their input was not incorporated into the fi-nal products. Rather, they were presented with a T-shirt al-legedly designed by another person using the same tool.2In
each group, one-third of participants were provided with no example, one third with an easy example, and the final third with a difficult example. After they finished the tasks de-scribed above, participants were asked to fill out a ques-tionnaire regarding their satisfaction3and perception of
self-congruity for the T-shirt either designed by themselves (participation) or by their yoked counterpart (no participa-tion).
Measures
As customers’ self-perception of their designing ability may have a significant impact on their satisfaction ratings, their self-assessed designing ability was measured to isolate its effect. It was measured using subjective perceptions of how customers evaluate their ability to manage the design-ing tasks. These measures were modified from the scale used by Park and Moon.27One item, for example, to which they
responded was, “Compared to other people, I think I have superior art abilities.”
The satisfaction scale was modified from the scale used in the study of Spreng et al,28which includes five semantic
dif-ferential items anchored as very dissatisfied/very satisfied, very
displeased/very pleased, very uncomfortable/very comfortable, very dislike/very like, and very frustrated/very contented.
Self-congruity was assessed following the steps proposed by Sirgy et al.29Participants were first instructed to take a
moment to think about the T-shirt pattern and the kind of person who would typically wear it. Then they were asked to describe this person using one or more adjectives, such as “stylish” and “sexy.” After they wrote down the adjectives, participants were asked to rate the extent to which they dis-agreed or dis-agreed with the statements adapted from the study.29One statement, for example, was: “Wearing this
T-shirt is consistent with how I see myself.”
To satisfy the demands of reliability, the Cronbach alpha was calculated, and all measurements exceeded the cutoff point of 0.7, indicating a high internal consistency for each scale.
Validity
To satisfy the requirements of construct validity, the prin-cipal axis factor analysis using the direct oblimin rotation was performed due to the correlation between satisfaction and self-congruity. A reasonable factor structure was pro-duced to support the convergent validity with all item load-ings higher than 0.6 on the appropriate dimensions. Finally, with respect to discriminant validity, the correlations among the measures taken were lower than the alpha coefficients of themselves.
Results
Manipulation checks
A manipulation check was employed to assess the effec-tiveness of the perceived difficulty for each design example. The results show that when participants were provided with a difficult example, their perceived difficulty was
signifi-cantly higher than when an easy example was provided, F(1, 118) 45.71, p 0.01. The results also demonstrate that the attractiveness of these two design examples was not per-ceived to be significantly different, F(1, 118) 0.07, ns, p 0.79.
Main effect of customer participation and moderating effects of design examples
An ANCOVA was conducted to examine the first and sec-ond hypotheses. In this analysis, self-assessed ability in de-signing was taken as a covariate to control the possible ef-fects caused by individual differences in designing capability, and the dependent variable was customer satis-faction with the customized product.
Table 1 summarized the ANCOVA test. The results reveal a significant effect of customer participation, F(1, 173) 41.525, p 0.001, and a significant interaction between cus-tomer participation and design examples, F(2, 173) 4.012,
p 0.05. To further investigate the main effect of customer
participation, the group means were examined. The means suggest that respondents who participated in the designing process had a higher level of satisfaction than respondents in the no-participation condition (XP 4.67, XNP 3.56, p
0.001). Thus, these results provide support for H1.
The significant interaction supports H2, which states that the positive effect of customer participation on satisfaction will be stronger when an easy design example is provided than when a difficult one or no example is provided. The re-sults of post hoc tests show that mean differences in satis-faction between participation and no-participation groups were greater when an easy design example was provided than when a difficult one or when no example was provided (difference in satisfaction: XEASY 1.751, XDIFFICULT 0.597, XNO 0.872, least significant difference post hoc test, p
0.05).
Mediation analysis
Mediation analyses following the procedures suggested by Baron and Kenny30were performed to demonstrate that
(a) the independent variable (customer participation) has a significant influence on the proposed mediator (self-congru-ity) by regressing the mediator on the independent variable; (b) the independent variable is shown to affect the depen-dent variable (satisfaction level) significantly by regressing the dependent variable on the independent variable; (c) when both the independent variable and the mediator are in
the regression model, the mediator must significantly affect the dependent variable, and the effect of the independent variable on the dependent variable must be less than that in the second regression model. The whole sample was first an-alyzed with the above procedures. Additionally, for each level of difficulty of the design examples provided, the same analysis was conducted.
To examine the degree of mediation for the whole sam-ple, first, self-congruity was taken as a dependent variable and regressed on customer participation. Then the depen-dent variable satisfaction was regressed on self-congruity, on customer participation, and finally on both customer partic-ipation and self-congruity. The impact of customer partici-pation on self-congruity was significant (t 5.97, p 0.001,
0.408). Moreover, it was found that customer
participa-tion significantly influenced satisfacparticipa-tion (t 6.083, p 0.001,
0.415). When satisfaction was regressed on both
cus-tomer participation and self-congruity, both cuscus-tomer par-ticipation (t 2.523, p 0.05, 0.139) and self-congruity (t 12.253, p 0.001, 0.675) were significant. Thus, in general, self-congruity was corroborated to be a significant mediator in the participation–satisfaction relationship. Therefore, the above provides support for H3.
Moderated mediation analyses
When an easy design example was provided, the impact of customer participation on self-congruity was significant (t 5.69, p 0.01, 0.599). Furthermore, the influence of customer participation on satisfaction was significant (t 6.78, p 0.01, 0.665). When satisfaction was regressed on both customer participation and self-congruity, both cus-tomer participation (t 3.30, p 0.01, 0.322) and self-congruity (t 5.88, p 0.01, 0.573) remained signifi-cant. The results of the mediation analyses indicate that self-congruity acts as a partial mediator between customer participation and satisfaction when easy design examples are provided. These findings were consistent with the prediction of H4a.
In the case of a difficult design example, the impact of cus-tomer participation on satisfaction was significant (t 2.01,
p 0.05, 0.255); however, the impact of customer par-ticipation on self-congruity was not (t 1.61, p 0.113, 0.207). When satisfaction was regressed on both customer participation and self-congruity, the impact of self-congruity on satisfaction remained significant (t 7.342, p 0.001, 0.689), whereas the influence of customer participation was insignificant (t 0.235, p 0.235, 0.113). These results TABLE 1. SUMMARY OFANCOVA
Type III sum of
Source squares df MS F Sig.
Self-assessed ability 9.229 1 9.229 7.007 0.009** Participation 54.696 1 54.696 41.525 0.000*** Sample provided 1.129 2 0.564 0.428 0.652 Interaction 10.568 2 5.284 4.012 0.020** Error 227.874 173 1.317 Total 3351.640 180
Dependent variable: satisfaction. df, degrees of freedom; Sig, Significance. **p 0.05; ***p 0.001.
indicate the absence of any mediation effect of self-congruity on the relationship between customer participation and sat-isfaction.
When no design example was provided, the effects of cus-tomer participation on both self-congruity (t 3.68, p 0.01,
0.435) and satisfaction (t 2.55, p 0.05, 0.317)
were significant. In addition, when satisfaction was re-gressed on both customer participation and self-congruity, the impact of customer participation became insignificant (t 0.07, p 0.948, 0.007), while the impact of self-con-gruity remained significant (t 6.96, p 0.01, 0.714). Thus, a perfect mediation of self-congruity was evident.
The mediation analyses provide support for H4, the me-diating effect of self-congruity on the relationship between customer participation and satisfaction being contingent upon design examples. However, H4a was only partially supported. When an easy design example was provided, a mediating effect of self-congruity on the relationship be-tween customer participation and satisfaction was observed. In the instance involving a difficult example, no mediation effect was found. However, surprisingly, when no design ex-ample was provided, self-congruity also mediated the rela-tionship between customer participation and satisfaction. These relationships are discussed in detail in the next sec-tion.
Discussion Conclusions
Based on the ANCOVA results, the main effect of partic-ipation suggests that encouraging customer particpartic-ipation tends to raise their satisfaction with a customized product. From a theoretical perspective, this supports the link be-tween customer participation and satisfaction. This phe-nomenon of participation may be explained by greater be-havioral control over the final outcome. The heightened customer satisfaction due to more perceived control is rooted in the personal responsibility15 that customers are more
likely to assume in a designing process. This assumed re-sponsibility, in turn, increases customer ownership of the product and therefore raises customer evaluations of their creations. This finding is consistent with the idea that cus-tomers who have the opportunity to participate in a certain task perceive more control and tend to feel more responsi-ble for as well as more satisfied with the outcome.15For
ex-ample, users of iGoogle are allowed to select different fea-tures, such as themes, gadgets, and layouts, for their Google home pages; thus, they obtain a higher sense of control as well as more positive experiences of iGoogle.
According to prior research, control affects outcome sat-isfaction when customers believe that they have the capa-bility to bring about desirable outcomes.31When customers
are not very certain whether they can accomplish a task suc-cessfully, such as when facing difficult examples during the designing process and becoming less confident in mastering the designing task, the effect of control might be reduced. The results of our study concerning H2 suggest this ratio-nale has some basis. Even when self-assessed ability is taken into account, when an easy example is provided to cus-tomers, customer participation can increase satisfaction to the highest degree. However, when a difficult example or no example is present, customer satisfaction augmentation fails to reach the same level.
The meditation analyses further suggest that the aug-menting effect of customer participation on satisfaction is mediated by consumer perception of self-congruity. When customers have the opportunity to customize, they are more likely to find the final product to be self-congruent. Fur-thermore, the more consumers perceive the final product they design as a reflection of who they are, the greater the satisfaction. However, the mediating effect of self-congruity is not the same across different levels of design examples. The results concerning H4 further suggest that the mediat-ing effect of self-congruity is contmediat-ingent upon the level of difficulty of the design examples with which customers are provided.
When provided with an easy design example, self-con-gruity is a partial mediator, which indicates that customer participation has a direct effect on customer satisfaction in addition to the indirect effect via self-congruity. The indirect effect suggests that when an easy example is present, cus-tomers are more likely to consider themselves as being ca-pable of translating their personal characteristics into de-signs. This incremented perception of self-congruity due to participation in the designing process then enhances cus-tomer satisfaction with the final products. In such a process, they are more likely to enjoy their participation in the de-sign process and able to create dede-signs more congruent with their perceptions of themselves that express their personal tastes, faith, and social status,29and thus they are more likely
to feel satisfied with their designs. Furthermore, above and beyond this indirect effect, participation alone has a direct positive impact on customer satisfaction when an easy ex-ample is provided. That is, simply being able to participate in the design process can have the additional effect of en-hancing customer satisfaction. While the sense of control is highly contingent upon individuals’ perception of their ca-pability to influence the outcome of an event, an easy design example is more likely to lead customers to believe that they can successfully accomplish such designing tasks. This be-lief in turn raises their level of satisfaction with their final products.
When a difficult design example was provided, the pat-tern of mediation was different. Customer participation did not lead to the enhancement of either self-congruity or cus-tomer satisfaction. This corresponds to social cognitive the-ory. When individuals do not deem themselves as efficacious in the satisfactory accomplishment of a task, they tend to be-come less willing to perform the task and are more likely to feel dissatisfied with their situation.22,24For example, a
stu-dent who is not good at mathematics is likely to feel frus-trated and dissatisfied with mathematics study. In the con-text of T-shirt customization, a design example that exceeds the abilities of customers may cause them to think that they are incapable of creating a T-shirt pattern that is equally com-plex and attractive. Under such circumstances, owing to self-serving bias, customers are less likely to consider that their designs are self-relevant. As a result, participation did not have an indirect effect on satisfaction via the sense of self-congruity. Additionally, in such a situation, customers are less likely to feel competent in creating satisfactory results, and this reduces the direct impact of participation on the level of satisfaction with the final outcome. This might help explain why there is no direct effect of participation on cus-tomer satisfaction.
con-dition somewhat contradict our prediction that only easy de-sign examples can help customers gain a higher sense of self-congruity and thereby increase the level of satisfaction with the customized T-shirt. Without the presence of design ex-amples, the meditation analyses show that self-congruity is a perfect mediator of the participation–satisfaction relation-ship. In other words, although no direct effect of participa-tion on customer satisfacparticipa-tion is observed, participaparticipa-tion can increase customer satisfaction via self-congruity. That is to say, whether or not customers are satisfied with the final cus-tomized products depends on the sense of self-congruity rooted in the designs that they create. While speculative in nature, this may be because in the absence of an example, some customers were able to create a design congruent with their self-images while others were not. Therefore, partici-pation alone did not have a direct effect on satisfaction. Only when participation increases self-congruity does customer satisfaction with the product also increase. The mechanism of this phenomenon, however, is yet to be discovered by fur-ther research.
Implications
Some directions for marketing practitioners can also be drawn from these results. The main effect of customer par-ticipation indicates that marketers should encourage tomers to participate in the designing process to enhance cus-tomer satisfaction. For example, cuscus-tomers can experience the pleasure and satisfaction associated with designing in-dividual T-shirts that meet their personal and symbolic needs.
While consumer participation is associated with aug-mented satisfaction, this effect is contingent upon the pres-ence of design examples that do not exceed customers’ design abilities. Provision of attractive but inherently com-plicated design examples or no example may not be an ef-fective strategy for creating customer satisfaction with cus-tomization. Customers may become frustrated by their inability to produce equally attractive and complicated items when an intricate example is proffered. On the other hand, when no example is at hand in the designing process, some customers might feel lost about what they can do and how to do it. Therefore, marketers should conduct general sur-veys to identify which level of difficulty suits most customers in order to find design examples to meet general customers’ design abilities. Firms can then provide design examples that are not overwhelmingly complicated to facilitate the design process in order to increase customer satisfaction. Design ex-amples with proper levels of difficulty will allow customers to experience greater satisfaction through their participation in the customizing process. In this way, the effect of cus-tomer participation can work to its full capacity.
Limitations and future research direction
While the sample size is sufficient for an experimental de-sign, the ability to transfer these results to the general pop-ulation is limited by its base sample, which mostly com-prised university students ranging in age from 18 to 25 years. However, a mostly student sample may be useful and justi-fied based on the findings of a survey of online shopping, which reports that customers aged from 20 to 29 compose the largest segment (approximately one-third) of online
shoppers in Taiwan.32Therefore, the importance of this
seg-ment itself warrants reasonable research attention.
Since the positive effects of customer participation may be explained by increased control, it would be useful to estab-lish whether or not perceived control also serves as a medi-ator to explain the effects of customer participation on these variables. Furthermore, the effect of customer participation may vary across cultures, since the degree to which indi-viduals want to express themselves by customizing products can differ. The need to express one’s personality via cus-tomization can be potentially stronger in individualistic cul-tures than in collective culcul-tures.33Thus, the enhanced
satis-faction as a result of participation can be strengthened in the individualistic cultures.
In addition, the present study examines customer partici-pation for a specific product category, T-shirts, a highly he-donic and symbolic product. However, it is possible that these results cannot always be generalized to other product categories. For instance, whether or not the customization of a utilitarian product, such as a computer, that allows cus-tomers to configure hardware and settings to meet their spe-cific functional needs provides the same result warrants fur-ther research.
Notes
1. Participants in each group were asked to use an online Flash drawing tool, Mr. Picassohead (www.mrpicasso-head.com). None of the respondents reported previous ex-perience in using such a Flash drawing tool; therefore, the experience with this specific tool was not a confounding variable.
2. In order to ensure that each participant received a T-shirt with an identical degree of attractiveness and that the only difference between the participation and no-participation groups was the opportunity to create their own cus-tomized T-shirt, the yoked-control technique employed by many prior choice studies was adopted. Using this technique, participants in the experimental group de-signed their own T-shirt and rated their satisfaction with their own work, whereas their counterparts in the control group were assigned a T-shirt designed by their yoked counterpart and were asked to rate their satisfaction with that T-shirt.
3. The participants rated their satisfaction with the T-shirts before receiving them. Therefore, it ruled out the possi-bility of the influences of cognitive dissonance and buyer’s remorse after the purchase.
Disclosure Statement
The authors have no conflict of interest.
References
1. Dellaert BGC, Stremersch S. Marketing mass-customized products: striking a balance between utility and complexity. Journal of Marketing Research 2005; 42:219–27.
2. Kramer T. (2003) The effect of preference measurement on preference construction and responses to customized offers. Stanford, CA: Stanford University.
3. Schreier M. The value increment of mass-customized prod-ucts: an empirical assessment. Journal of Consumer Behav-iour 2006; 5:317–27.
4. Fiore AM, Lee S-E, Kunz G. Individual differences, motiva-tions, and willingness to use a mass customization option for fashion products. European Journal of Marketing 2004; 38:835–49.
5. Kamali N, Loker S. Mass customization: online consumer in-volvement in product design. Journal of Computer-Medi-ated Communication 2002; 7.
6. Franke N, Piller F. Value creation by toolkits for user inno-vation and design: the case of the watch market. Journal of Product Innovation Management 2004; 21:401–15.
7. Huffman C, Kahn BE. Variety for sale: mass customization or mass confusion? Journal of Retailing 1998; 74:491–513. 8. Sirgy MJ. Using self-congruity and ideal congruity to
pre-dict purchase motivation. Journal of Business Research 1985; 13:195–206.
9. Hsieh A-T, Yen C-H, Chin K-C. Participative customers as partial employees and service provider workload. Interna-tional Journal of Service Industry Management 2004; 15:187. 10. Bitner MJ, Faranda WT, Hubbert AR, et al. Customer con-tributions and roles in service delivery. International Jour-nal of Service Industry Management 1997; 8:193.
11. Allen MW, Ng SH. The direct and indirect influences of hu-man values on product ownership. Journal of Economic Psy-chology 1999; 20:5–39.
12. Holbrook MB, Hirschman EC. The experiential aspects of consumption: consumer fantasies, feelings, and fun. Journal of Consumer Research 1982; 9:132–40.
13. Babin BJ, Darden WR, Griffin M. Work and/or fun: mea-suring hedonic and utilitarian shopping. Journal of Con-sumer Research 1994; 20:644–56.
14. Hirschman EC, Holbrook MB. Hedonic consumption: emerging concepts, methods and propositions. Journal of Marketing 1982; 46:92–101.
15. Van Raaij WF, Pruyn ATH. Customer control and evalua-tion of service validity and reliability. Psychology & Mar-keting 1998; 15:811.
16. Averill JR. Personal control over aversive stimuli and its re-lationship to stress. Psychological Bulletin 1973; 80:286–303. 17. Godek J, Yates JF, Yoon Y. Customization and personaliza-tion: the influence of perceived control and perceived capa-bility on product evaluations. Advances in Consumer Re-search 2002; 29:157.
18. Lin H-F. Understanding behavioral intention to participate in virtual communities. CyberPsychology & Behavior 2006; 9.
19. Tam KY, Ho SY. Web personalization as a persuasion strat-egy: an elaboration likelihood model perspective. Informa-tion Systems Research 2005; 16:271.
20. Bendapudi N, Leone RP. Psychological implications of cus-tomer participation in co-production. Journal of Marketing 2003; 67:14–28.
21. Stouffs R, Tunçer B, Sariyildiz S. (2002) Examples of Web-based custom design in Dutch residential developments. In
Proceedings of International Conference of Distributing Knowl-edge in Building, International Council for Research and In-novation in Building and Construction, Aarhus School of Architecture.
22. Bandura A. Self-efficacy mechanism in human agency. American Psychologist 1982; 37:122–47.
23. Johar JS, Sirgy MJ. Value-expressive versus utilitarian ad-vertising appeals: when and why to use which appeal. Jour-nal of Advertising 1991; 20:23.
24. Bandura A. Self-efficacy: toward a unifying theory of be-havior change. Psychological Review 1977; 84:191.
25. Ulrich PV, Anderson-Connell LJ, Wu W. Consumer co-de-sign of apparel for mass customization. Journal of Fashion Marketing & Management 2003; 7:398–412.
26. Perdue BC, Summers JO. Checking the success of manipu-lations in marketing experiments. Journal of Marketing Re-search 1986; 23:317–26.
27. Park C-W, Moon B-J. The relationship between product in-volvement and product knowledge: moderating roles of product type and product knowledge type. Psychology & Marketing 2003; 20:977.
28. Spreng RA, MacKenzie SB, Olshavsky RW. A reexamination of the determinants of consumer satisfaction. Journal of Mar-keting 1996; 60:15.
29. Sirgy MJ, Dhruv G, Tamara FM, et al. Assessing the predic-tive validity of two methods of measuring self-image con-gruence. Academy of Marketing Science Journal 1997; 25:229.
30. Baron RM, Kenny DA. The moderator-mediator variable dis-tinction in social psychological research: conceptual, strate-gic, and statistical considerations. Journal of Personality & Social Psychology 1986; 51.
31. Namasivayam K. Action control, proxy control, and con-sumers’ evaluations of the service exchange. Psychology & Marketing 2004; 21:463–80.
32. Yang CH. Online shopping behavior: the features and per-spective of b2c development in Taiwan. Market Intelligence Center 2007.
33. Hofstede G, McCrae RR. Personality and culture revisited: linking traits and dimensions of culture. Cross-Cultural Re-search 2004; 38:52–88.
Address reprint requests to:
Dr. Chia-Chi Chang Department of Management Science National Chiao Tung University 1001 Ta Hsueh Road, Hsinchu Taiwan 300, Taiwan Republic of China E-mail: chi@mail.nctu.edu.tw
