Special Session Summary Understanding, Measuring and Engineering the Consumption Experience in the Online Environment

ABSTRACT - Recently, there has been extensive interest in engineering the consumer experience, i.e., the design and management of objective features of a firm’s offering and communications, in a way that induces a positive subjective state in the consumer, engages the consumer in a personal way, and ultimately shapes positive responses such as purchase behavior, satisfaction and loyalty. The three papers in this session use complementary research approaches to understand different dimensions of the consumption experience in the online environment. The first paper by Annamma Joy and John Sherry builds upon the ethnographic approach to identify the key metaphors and metonymies used by consumers to make sense of the internet. When metaphorical property mapping occurs, the focus is on terms such as Avirtual worlds,@ Afantasy experiences,@ Aaddiction,@ Abeing lost in space@, Abuilding bridges@ Asuperhighways.@ In metonymy, the text reveals phrases and terms such as Abetrayal@ [being on guard against viruses], Aoverwhelmed@ [due to information overload], Afrustration@ [because of anonymity of the web], Atransformation@ [continuously re-invent themselves], and Ainformality@ [at home -while the fingers do the moving]. The authors discuss how their results provide insights to web-designers to create specific types of online experiences, much the same way an architect would design a space to create a certain subjective experience in visitors to a building. The second paper by Laurette Dube, Demetrios Vaskatras, and Jianying Zhao focuses on the structure and dynamics of emotional experiences when consumers use functional e-services such as web-search services (e.g., Yahoo, Farefinder). The authors present a longitudinal study in which consumers of web-search services were asked to report on their emotions and their perceptions of service attributes, as they perform a sequence of 15 calibrated search tasks on as many different days. Analyses revealed a 4-factor structure for emotions, organized in two dominant high-arousal factors (frustration; enthusiasm) and two less prevalent, low-arousal factors (calm and sadness). Perceptual features of the product (i.e., web-search service) were organized into three factors: processing effectiveness, speed, and graphic quality. Of these, perceived effectiveness was the most powerful and universal attribute antecedent of consumption emotions. Further, while gender, familiarity, objective knowledge and level of usage were all found to be significant moderators of emotional experiences, the effects of some of these moderators were found to vary by the type of consumption emotions. The third paper by J. J. Brakus and Bernd Schmitt investigates five types of consumption experiences and their underlying processing, in the context of static web banners and animated web advertisements. This research builds on a theoretical framework that distinguishes consumer experiences into five types ( i.e., sensory, affective, intellectual, bodily and social), and posits that these experiences occur on two separate levels. At a primary level, responses are fairly instinctual and automatic, while at a secondary level, responses are learned and acquired. Predictions derived from the framework are supported in two experimental studies. The discussant, Gilles Laurent, commented on ways in which the portfolio of methodological approaches presented in this session could be profitably integrated to develop innovative research on ways to engineer online emotional experiences to be friendlier to consumers and more effective for marketers.

Citation:

Laurette Dube and Ashesh Mukherjee (2001) ,"Special Session Summary Understanding, Measuring and Engineering the Consumption Experience in the Online Environment", in E - European Advances in Consumer Research Volume 5, eds. Andrea Groeppel-Klien and Frank-Rudolf Esch, Provo, UT : Association for Consumer Research, Pages: 1-3.

European Advances in Consumer Research Volume 5, 2001      Pages 1-3

SPECIAL SESSION SUMMARY

UNDERSTANDING, MEASURING AND ENGINEERING THE CONSUMPTION EXPERIENCE IN THE ONLINE ENVIRONMENT

Laurette Dube, McGill University, Canada

Ashesh Mukherjee, McGill University, Canada

ABSTRACT -

Recently, there has been extensive interest in engineering the consumer experience, i.e., the design and management of objective features of a firm’s offering and communications, in a way that induces a positive subjective state in the consumer, engages the consumer in a personal way, and ultimately shapes positive responses such as purchase behavior, satisfaction and loyalty. The three papers in this session use complementary research approaches to understand different dimensions of the consumption experience in the online environment. The first paper by Annamma Joy and John Sherry builds upon the ethnographic approach to identify the key metaphors and metonymies used by consumers to make sense of the internet. When metaphorical property mapping occurs, the focus is on terms such as "virtual worlds," "fantasy experiences," "addiction," "being lost in space", "building bridges" "superhighways." In metonymy, the text reveals phrases and terms such as "betrayal" [being on guard against viruses], "overwhelmed" [due to information overload], "frustration" [because of anonymity of the web], "transformation" [continuously re-invent themselves], and "informality" [at home -while the fingers do the moving]. The authors discuss how their results provide insights to web-designers to create specific types of online experiences, much the same way an architect would design a space to create a certain subjective experience in visitors to a building. The second paper by Laurette Dube, Demetrios Vaskatras, and Jianying Zhao focuses on the structure and dynamics of emotional experiences when consumers use functional e-services such as web-search services (e.g., Yahoo, Farefinder). The authors present a longitudinal study in which consumers of web-search services were asked to report on their emotions and their perceptions of service attributes, as they perform a sequence of 15 calibrated search tasks on as many different days. Analyses revealed a 4-factor structure for emotions, organized in two dominant high-arousal factors (frustration; enthusiasm) and two less prevalent, low-arousal factors (calm and sadness). Perceptual features of the product (i.e., web-search service) were organized into three factors: processing effectiveness, speed, and graphic quality. Of these, perceived effectiveness was the most powerful and universal attribute antecedent of consumption emotions. Further, while gender, familiarity, objective knowledge and level of usage were all found to be significant moderators of emotional experiences, the effects of some of these moderators were found to vary by the type of consumption emotions. The third paper by J. J. Brakus and Bernd Schmitt investigates five types of consumption experiences and their underlying processing, in the context of static web banners and animated web advertisements. This research builds on a theoretical framework that distinguishes consumer experiences into five types ( i.e., sensory, affective, intellectual, bodily and social), and posits that these experiences occur on two separate levels. At a primary level, responses are fairly instinctual and automatic, while at a secondary level, responses are learned and acquired. Predictions derived from the framework are supported in two experimental studies. The discussant, Gilles Laurent, commented on ways in which the portfolio of methodological approaches presented in this session could be profitably integrated to develop innovative research on ways to engineer online emotional experiences to be friendlier to consumers and more effective for marketers.

 

"EXPERIENCING CYBERSPACE: CONSUMER RESPONSES TO NEW TECHNOLOGY"

Annamma Joy, Concordia University

John F Sherry Jr., Northwestern University

Cyberspace is a symbolic system supported by computers and other machines and related technologies. The metaphor of #space’ is a useful one in contemplating cyberspace. In opposition to #place’, that often requires some form of physical locality and a body rooted in gravity, #space’ is more abstract. There is a sense of vastness and openness about cyberspace and it suggests a new frontier that can be annexed. Cyber space can be experienced in many ways; the only condition is that the flesh cannot enter these spaces. The metaphors that we use guide the ways in which we think of cyberspace. If we view such spaces as navigable spaces, then we can consider whether it is sculptural space [palpable] or architectural space [livable]. If we use other metaphors then the design of such systems would also differ.

This study examines the meanings of cyberspace as part of a larger study on experiencing art on the Internet. First, however, we wished to know how the participants viewed and made sense of the Internet. A version of the Zaltman metaphor elicitation technique was used for this study. Thirty volunteers who wished to participate in this study were asked to choose ten pictures [any form of visual presentation] that represented in their minds the meanings they associated with the Internet. These pictures are projective in terms of representing what the Internet and cyberspace meant to each of them. They were then asked to write a two-page report explaining how and why each of these pictures represented their current thoughts and meanings associated with the use of this technology. They were also asked to visually manipulate the ten images [scanned or print media] and create a collage. A final report summarized their overall reactions to the Internet and provided a description of the collage that they created. Our understanding of the many meanings of cyberspace is based on thick descriptions provided by subjects about the images, that make choate their inchoate thoughts. These articulated meanings in turn, contribute towards an understanding of the larger culture that is evolving around the use of the internet and related technologies

Based on participant responses we were able to identify certain themes. For instance, when metaphorical property mapping occurs, the focus in the text is on terms such as "virtual worlds," "fantasy experiences," "addiction," "being lost in space", "building bridges" "superhighways." In the second context of metonymy, the text focuses on phrases and terms such as "instant and frequent communication [often cross-cultural]", "betrayal" [being on guard against viruses], "overwhelmed" [due to information overload and because they are exchanging technical skills for thought processes], "frustration" [because of anonymity of the web], "transformation" [continuously re-invent themselves], and "informality" [at home -while the fingers do the moving]. We discuss how a better knowledge of the metaphorical language used by consumers to describe how they see and experience the virtual space, can help web-designers to create contexts within which communication and persuasion can quickly occur, and not simply create a series of screen designs. In other words, web-sites can be built, engineered or designed to emulate the three dimensional world that we live in, by using metaphorical connections. This paper helps map the metaphors and associations that help make effective connections, interactions and dialogues between the everyday world of consumers and cyberspace.

 

"CONSUMER EMOTIONS IN VIRTUAL ENVIRONMENTS: STRUCTURE, ANTECEDENTS, MODERATORS AND DYNAMIC MODELING OVER TIME"

Laurette Dube, Demetrios Vakratsas, and Jianying Zhao, McGill University

Research on consumption emotions in the context of e-services or, more generally, as people travel in cyberspace has thus far primarily focused on "flow," i.e., intense emotional experience in which one loses a sense of oneself and a sense of time. Yet, as e-services or product and services marketed through internet are transformed into commodities, relatively few consumers are likely to experience flow. In the present research, we explore consumption emotions beyond flow, while using an e-service on repeated episodes (namely Yahoo websearch). Specifically, we empirically explore the various dimensions of consumption emotions in virtual environments, their perceptual antecedents in terms of service features, and individual-level moderators of the effect of features on emotional experience. We also explore the dynamics of emotions in search episodes over time.

In a controlled experiment, 82 consumers (average age 20 years; 41 males, 41 females) of web-search services were asked to report on their emotions while using the Yahoo search engine to perform 15 tasks in sequence on as many different days. Each of the tasks was designed and pre-tested to take between 10-20 minutes to perform, and each participant completed the same 15 tasks in a different random order under one of two judgment instruction manipulations after each search. Participants provided self-report (11-point scales) of emotions (positive: enthusiastic, happy, stimulated, content, elated, calm and relaxed; negative: frustrated, angry, irritated, anxious, tense, impatient, sad, depressed) and their perceptions of various service features (processing speed, graphic quality, ease of use, search process efficiency, search process pleasantness, search outcome quality). For each episode, they also indicated how much they liked the task (single item) as well as the times the task started and ended. Individual moderators considered were gender, self-reported familiarity (11-point scales, familiar, knowledgeable, interested), objective knowledge (6-item scale), and usage level (# searches per week) of web-search services.

Factor analyses (exploratory followed by confirmatory) revealed a 4-factor structure for consumption emotions, organized in terms of valence and arousal. The two high-arousal factors of frustration (frustration, angry, irritated) and enthusiasm (enthusiastic, happy and stimulated) each accounted for about one quarter of the shared variance with two less prevalent, low arousal factors (calm: calm, relaxed; sad: sad depressed). The perceptual features of the search engine were organized into three factors capturing processing effectiveness (easy of use, efficiency, enjoyment, outcome quality), speed (time to completion, processing speed), and graphic quality (single item). The impact of perceptual antecedents and moderators on consumption emotions was assessed by a random coefficient approach. In this approach, the three perceptual antecedents and liking for the task were taken as random effects, while gender, familiarity, knowledge, and usage level were considered fixed effects in distinct predictive models for each dimension of emotional experience. Perceived effectiveness was found to be the most powerful and universal antecedent of consumption emotions with strong effect on each dimension, being the unique predictor for enthusiasm and sadness. Perceived speed was a significant antecedent of frustration (-) and calm (+). A puzzling positive relationship emerged between graphic quality and frustration. Liking for the task influenced frustration (-) and enthusiasm (+) only.

In terms of moderators, all four moderators had significant effects of some of the emotional dimensions but the impact varied by emotions. Gender moderated the degree of frustration (p=0.10), enthusiasm (p<.05) and calm (p<.01) with females, compared to males, reporting more intense frustration and enthusiasm and less calmness. Self-reported familiarity had a moderating effect on three dimensions of emotions, with less frustration (p<0.10), more enthusiasm (p<0.01) and more calm (p<0.01). More knowledgeable consumers reported less enthusiasm (p<.01), more calm (p<.01) and more sadness (p=.05). As for the level of usage, it was found that heavier user of web-search services seem to have more intense emotional responses on the dimensions of frustration (p<0.01), enthusiasm (p<0.10), and sadness (p<0.01).

Finally, we modeled the dynamics of each emotional dimension over the course of a single search episode. We used a multiplicative effects model to break down the cumulative effect of the order of the task and its duration. Results show that, for all four emotions, intensity fades over time as consumers move along the task sequence. Within each task, the experience of frustration and enthusiasm (and marginally so for calm) reflects a non-monotonic dynamic. More specifically, the effect of time on frustration exhibits an inverted U-shape with a late peak, whereas the effect on enthusiasm initially has as an inverted-U shape which later turns into a U-shape (in other words, first exhibits a dip, then a peak and later decreases). No dynamic pattern emerged for sadness. Theoretical and managerial implications of the results are discussed.

 

"VARIETIES OF CONSUMER EXPERIENCE: AN EMPIRICAL TEST IN THE ONLINE ENVIRONMENT"

J.J. Brakus and Bernd Schmitt, Columbia Business School

This paper builds on a theoretical framework that distinguishes consumer experiences into five types, i.e., sensory, affective, intellectual, bodily and social. These experiences occur on two separate response levels-a primary level at which responses are fairly instinctual and automatic, and a secondary level at which responses are learned and acquired. At the primary level, the process entails evolutionary response to specific environmental cues, called "affordances," that offer perceptual clues to the organism by mere virtue of the specific organism-environment interaction. The secondary level consists of a conscious step-by-step process of perception, encoding, storage, and retrieval that results in the respective experiences. Predictions derived from this framework are tested in an online environment using animated web advertisements and static web banners.

We propose that consumers engage in different processes (direct perception vs. inference making) as they engage in experiences of different types. The five types of consumer experiences are expected to operate in the following way. At the primary level, visual attributes of fundamental chromatic colors and basic shapes trigger the sensory module; facial displays of universal primary emotions (e.g., a smile) trigger the affective module; syllogisms, thought schemas and violations of expectations trigger the intellectual module; stimuli that result in arousal and/or pain trigger bodily modules; and references to kin (mother, father, sister, brother) trigger the social module. These propositions were tested in two experiments using web-advertisements and web-banners as experimental stimuli.

In Study 1, three animated web advertisements for a new product Ban MP3 playerBwere developed and pre-tested to create each of three types of consumer experience (sensory, affective and intellectual). The sensory and affective ads contained specific affordances intended to provide the respective experiences, while the intention of the intellectual ad was to simply present features of the product. The number of presented claims across the ads was the same. Furthermore, for each ad we made two versionsBthe "normal tempo" version and the "fast tempo" version. We made sure that the subjects who saw the fast version did not have enough time to read all the presented information, unlike the subjects who saw the normal version. Subjects provided product evaluation and measures of the consumption experience. We expected that sensory and affective information would be processed faster and more consistently across subjects than the intellectual information. Hence, even if the time to process all the information contained in an advertisement is cut short, subjects might still evaluate the advertised product highly if it is presented in affective and sensory rather than in an intellectual way. However, if there is not enough time to process the functional information, consumers become insecure about the product performance. This insecurity could drive down their evaluation of the product. These predictions were supported by the results.

In Study 2, we wanted to test further whether primary and secondary level processing could exist within each type of consumer experience. Additional pre-tests were performed to formulate web-banners that manipulated processing levels within sensory (color; red-green/orange-olive), affective (facial expressions; happy-sad/surprise-angry), and intellectual (yin and yang symbol; two variants) types of consumption experiences. Results from recognition and product evaluation data showed that the two levels of processing could be empirically distinguished within each type of experience. Subjects recognized web-banners more easily, consistently and faster at the primary level than on the secondary level. Moreover, subjects’ responses were affected less by exposure time at the primary level compared to the secondary level of processing.

----------------------------------------