Consumers nowadays purchase a variety of products in online shops for different reasons. Certain products involve high involvement decision-making with low purchase frequencies in general. At the same time, virtual shelf space is unlimited and consumers face a variety of products, which exceeds their rational capabilities. This condition requires online shop operators to implement search tools in their web sites that allow consumers to structure and reduce complexity, both on a catalogue and a product level. Consumers in general do not always possess product expertise, especially in the case of low frequency purchases such as digital cameras. Virtual product advisors intend to fill this gap.
The primary objective of the thesis is to investigate the interaction effect between different levels of consumer knowledge and a chosen product search approach. A special focus is put on a virtual product advisor and a facet search as a structuring tool. Based on theoretical work in marketing, psychology, information system management a set of hypotheses was developed pertaining to the interaction effect and how it affects the perceived quality of the online feature of
a product search interface from a consumer perspective.
A randomized experiment with a control group design in a live Online Shop was conducted to test the hypotheses. In sum, the findings suggest a contingency between the consumer
knowledge and a product search interface in regard of the impact on antecedents of esatisfaction. The results provide two different angles from a marketing perspective in terms of
usefulness and from an information system management point of view in terms of usability.
Table of Contents
ABSTRACT
ZUSAMMENFASSUNG
RÉSUMÉ
List of Tables
List of Illustrations
List of Abbreviations
1. Introduction
I. THEORETICAL FRAMEWORK
2. The Net Economy
2.1. The Concept of an Online Shop
2.2. Electronic Sale of Products
2.3. Search Quality of a Camera
3. Definitions
3.1. General Terms
3.2. Complexity
3.2.1. Catalogue Complexity
3.2.2. Product Complexity
3.3. Product Search Approaches
4. Consumer Behavior
4.1. Information Processing Strategies
4.2. Expertise Knowledge
4.3. Online Shopping Strategies
5. Consumer E-Satisfaction
5.1. Understanding of E-satisfaction
5.2. Web Site Quality from a Marketing Perspective
5.3. Usability from an Information System Perspective
5.4. Application of the Two Perspectives
6. Problem Statement
6.1. Research Questions
6.2. Hypotheses
6.2.1. Usefulness:
6.2.2. Usability (Ease of Use)
II. EMPIRICAL PART
7. Research Methodology
7.1. Experimental Design
7.2. Questionnaire
7.3. Key Items of the Questionnaire
7.3.1. Usefulness
7.3.2. Usability (Ease of Use)
7.3.3. Manipulation Check
7.3.4. Control Variable
7.4. Socio-Demographic Data
7.5. Distribution
8. Preliminary Data Analysis
8.1. Data Screening
8.2. Assessment of Normality
9. Preparation of Data
9.1. Outliers
9.2. Transformation of variables for further analysis
9.3. Reliability
10. Data Analysis
10.1.Chosen statistical methods
10.2.Review of Assumptions for the ANOVA
11. Test Results
11.1.Test Results for Usefulness of the Product Search Approach .
11.2.Test Results for Usability of the Product Search Approach
11.3.Hypotheses Validity
III. DISCUSSION
12. Managerial Implications
12.1.Limitations of the Study
12.2.Future Research
Bibliography
Literature:
Internet Sources:
Appendixes
Appendix A
Appendix B
Appendix C
Appendix D
Appendix E
Appendix F
Appendix G
Appendix H
Appendix I
Appendix J
Appendix K
Appendix L
Appendix M
Appendix N
Appendix O
Appendix P
Appendix Q
Appendix R
Appendix S
ABSTRACT
Consumers nowadays purchase a variety of products in an online shops for different reasons. Certain products involve high involvement decision-making with low purchase frequencies in general. At the same time, virtual shelf space is unlimited and consumers face a variety of products, which exceeds their rational capabilities. This condition requires online shop operators to implement search tools in their web sites that allow consumers to structure and reduce complexity, both on a catalogue and a product level. Consumers in general do not always possess product expertise, especially in the case of low frequency purchases such as digital cameras. Virtual product advisors intend to fill this gap.
The primary objective of the thesis is to investigate the interaction effect between different levels of consumer knowledge and a chosen product search approach. A special focus is put on a virtual product advisor and a facet search as a structuring tool. Based on theoretical work in marketing, psychology, information system management a set of hypotheses was developed pertaining to the interaction effect and how it affects the perceived quality of the online feature of a product search interface from a consumer perspective.
A randomized experiment with a control group design in a live Online Shop was conducted to test the hypotheses. In sum, the findings suggest a contingency between the consumer knowledge and a product search interface in regard of the impact on antecedents of e- satisfaction. The results provide two different angles from a marketing perspective in terms of usefulness and from an information system management point of view in terms of usability.
(Keywords: Complexity, Electronic Commerce, E-Satisfaction, Expertise Knowledge, Means-End Chain, Product Search Interfaces, Usability, Virtual Product Advisors, WebQual, Web Site Quality)
ZUSAMMENFASSUNG
Konsumenten kaufen heutzutage viele, verschiedenartige Produke aus unterschliedlichen Gründen online ein. Im Allgemeinen erfordern bestimmte Produkte eine starke Kundeneinbindung bei niedriger Kaufhäufigkeit. Dabei besitzen Online Shops unbegrenzten Platz für die Warenpräsentation und Konsumenten stehen einem Produktangebot gegenüber, welches sie rational überfordert. Diese Ausgangssituation verlangt eine Einbindung von Produktsuchwerkzeugen in den Online Shops seitens der Betreiber. Dabei stehen verschiedene Möglichkeiten zur Verfügung, die Komplexität sowohl auf Produkt- als auch auf Katalogebene zu strukturieren und zu reduzieren. Verbraucher haben verallgemeinert nicht immer Produktexpertise, vor allem für unregelmäßige Einkäufe wie beispielsweise einer digitalen Kamera. Virtuelle Produktberater sollen diese Lücke schließen.
Das Hauptziel dieser These besteht aus der Ermittlung der Wechselwirkungen zwischen zwei Graden der Produktexpertise und zweier Produktsuchschnittstellen. Mittelpunkt der Untersuchung sind die Suchansätze eines Virtuellen Produktberaters sowie einer Facettensuche als Struktierungswerkzeug. Arbeiten aus den Fachbereichen Marketing, Psychologie und Informationssystem Management bilden das Fundament der entwickelten Hypothesen, welche die Wechselwirkungen betreffen und die Auswirkungen auf die wahrgenommene Qualität des Webseitenmerkmals Produktsuchschnittstelle aus seiner Verbrauchersicht ermitteln.
Ein randomisiertes Experiment mit einem Kontrollgruppen Design wurde während des laufenden Betriebs eines Online Shops durchgeführt, um die Hypothesen zu testen. Kurzum suggerieren die Ergebnisse eine Kontingenz der Merkmale Verbraucherwissen und Produktsuche. Diese Wechselwirkung wiederum beeinflusst die Antezendenzien der Online-Zufriedenheit. Die Resultate bestehen aus zwei unterschiedlichen Betrachtungswinkel, einerseits aus einer Marketingperspektive bezüglich der empfunden Nützlichkeit, und andererseits aus einem Informationssystem Management Blickwinkel bezüglich der Benutzerfreundlichkeit.
(Stichw ö rter: Benutzerfreundlichkeit, E-Commerce, Expertenwissen, Komplexit ä t, Means-End Chain Theorie, Online Zufriedenheit, Produktsuche, Virtuelle Produktberater, WebQual, Webseiten Qualit ä t)
RÉSUMÉ
Les consommateurs d’aujourd'hui achètent pour des raisons diverses toutes sortes de produits dans des boutiques en ligne. D’une façon générale, certains produits exigent une forte intégration du client, même si sa fréquence d’achat reste faible. Les boutiques en ligne ont un espace illimité quant à la présentation de produits, ce qui a pour conséquence que les consommateurs sont parfois dépassés d’un point de vue rationnel par la gamme quasi-indéfinie de produits. Cette situation exige donc des opérateurs la mise en place et l'intégration efficace d’outils de recherche de produits au sein des boutiques en ligne. Pour ce faire, il existe diverses options permettant de structurer et réduire la complexité du catalogue et des produits. Bien souvent, les consommateurs n'ont pas une connaissance fondée d’un produit, notamment lorsqu’il s’agit d’un investissement irrégulier tel que l'achat d'un appareil photo numérique, par exemple. Ainsi, la mise à disposition de conseillers virtuels pourra combler ce déficit.
L'objectif principal de cette thèse consiste à déterminer les différents types d’interactions aux niveaux respectifs de l’expertise des produits et d’autre part de l'interface de recherche de produits. L’objet de l'étude présente l’analyse de la fonction de recherche propre aux conseillers virtuels ainsi que de la recherche à facettes en tant qu'outil structurant. Des analyses dans les domaines du marketing, de la psychologie et de la gestion des systèmes d'information constituent la base des hypothèses développées et ont pour objectif d’établir l’impact de la fonction « interface de recherche d’un produit » sur la perception du consommateur.
Une étude a ainsi été effectuée sur un groupe de contrôle type lors de l'utilisation d'une boutique en ligne en vue de tester les hypothèses présentées. Les résultats montrent notamment d’une part le rapport intrinsèque entre la connaissance du consommateur d’un produit particulier et d’autre part le procédé de recherche de ce même produit, cette interaction affectant à son tour le degré de satisfaction des clients en ligne. Cette étude offre ainsi la possibilité d’observer les résultats sous deux angles différents: d'une part, du point de vue marketing focalisé sur la perception de l'utilité puis sous l’optique de la gestion des systèmes d'information en termes de facilité d'utilisation.
(Mots-cl é s: Commerce é lectronique, Complexit é , Conseillers virtuels, Expertise des produits, Facilit é d'utilisation, Qualit é de site web, Recherche de produits, Satisfaction é lectronique, Th é orie de Means-End Chain, WebQual)
List of Tables
Table 1: Comparison of Product Search Approaches
Table 2: Selected Antecedents of Customer E-Satisfaction
Table 3: Recommended Actions for Web Site Areas by WebQual
Table 4: Overview of Data Labels
Table 5: Overview of Socio-Demographic Descriptive Statistics
Table 6: Cross-Tabulation of Independent Variables
Table 7: Overview of Descriptive Statistics of Continuous Variables
Table 8: Test Statistics for the Manipulation Check
Table 9: Descriptive Statistics of Manipulation Check
Table 10: Descriptive Statistics for DV by Group
Table 11: Correlation Output for Informational Fit-to-Task Items
Table 12: Correlation Output for Tailored Communication Items
Table 13: Homogeneity of Variance of the Dependent Variables
Table 14: Descriptive Statistic for Means of USF1 Scale
Table 15: SPSS Output of two-way between subjects ANOVA (USF1)
Table 16: Test Statistics for Mann-Whitney U Test (FS - USF1)
Table 17: Descriptive Statistics for Mann Whitney U Test (FS - USF1)
Table 18: Test Statistics for Mann-Whitney U Test (VPA - USF1)
Table 19: Descriptive Statistics for Mann Whitney U Test (VPA - USF1)
Table 20: Descriptive Statistics for Means of USF2 Scale
Table 21: SPSS Output of two-way between subjects ANOVA (USF2)
Table 22: Descriptive Statistics for Means of EFCT Scale
Table 23: SPSS Output of two-way between subjects ANOVA (EFCT)
Table 24: Descriptive Statistics for Means of EFCY Scale
Table 25: SPSS Output of two-way between subjects ANOVA (EFCY)
Table 26: Test Statistics for Mann-Whitney U Test (FS - EFCY)
Table 27: Descriptive Statistics for Mann Whitney U Test (FS - EFCY)
Table 28: Test Statistics for Mann-Whitney U Test (VPA - EFCY)
Table 29: Descriptive Statistics for Mann Whitney U Test (VPA -EFCY)
Table 30: Descriptive Statistics for Means of SAT Scale
Table 31: SPSS Output of two-way between subjects ANOVA (SAT)
Table 32: Summary of Test Results
List of Illustrations
Illustration 1: Principle of an Online Shop
Illustration 2: „Anytime/Anyplace“ - Matrix
Illustration 3: Nikon D600 System Chart
Illustration 4: Venn Diagram of a Facet Search
Illustration 5: Virtual Product Advice Process
Illustration 6: Saturn Online Shop Search Interface
Illustration 7: The Memory Process by Atkinson, Shiffrin
Illustration 8: Means-End-Chain Model
Illustration 9: Attribute-Need Inference
Illustration 10: Generic Online Shopping Strategies
Illustration 11: Usability - Web Site Quality Framework
Illustration 12: E-Satisfaction Antecedent Contingency
Illustration 13: Histograms of the Variables SFQ and ASC
Illustration 14: Histograms of the Manipulation Check Variables
Illustration 15: Profile Plot of USF1
Illustration 16: Profile Plot of USF2
Illustration 17: Profile Plot of EFCT
Illustration 18: Profile Plot of EFCY
Illustration 19: Profile Plot of SAT
List of Abbreviations
Abbildung in dieser Leseprobe nicht enthalten
1. Introduction
Different industries have undergone structural changes, especially in the fast moving consumer goods (FMCG) sector due to advances in information and communication technologies (ICT) in the past decade. Prominent examples such as Zalando, Amazon and iTunes induced disruptive changes of the retail landscape. Anytime and anywhere shopping possibilities with a competitive pricing scheme led to the decline of ‘brick and mortar’ shoe retailers and book stores, and accessibility to digitalized music caused diminishing returns in the music industry. On the one hand, brick and mortar firms are facing increased competition from dot- com firms. It seems to be just a matter of time, when current offline product categories such as fresh groceries will be purchased through online channels. On the other hand, mature online retailer find themselves already in a highly competitive environment, as the next store is literally just ‘one click away’.
Major investments in the width and depth of product catalogues are realized in order to attract a great number of potential customers. A portfolio can be even dedicated to a narrow single product category - so called long tail offers -, which would not be viable in a physical retail environment. In the net economy , single product category offers become viable in the so-called long tail. Other online shop operators pursue the strategy to offer many product categories with a high variety of products. The arising complexity of product catalogues requires the use of product search tools, which can help to reduce complexity. Not only growing product catalogues impose a challenge on consumers for finding the right product, but the products themselves might be complex and require an extensive information search process beforehand. Virtual product advisors are developed to aid consumers in the decision-making process and in the selection of a consideration set.
The purpose of decision-making aids is to increase the customer’s perception of the quality of an online shop. Multiple factors exist, that constitute the quality perception, such as usability of the online shop, affective components such as entertainment of the online shopping experience or back-end processes such as product delivery impact the customer experience immediately. The logic is that satisfied customers are likely to revisit, purchase and repurchase from an online shop. They might perform more up- and cross-selling and tend to have higher, average shopping cart values. As an ever increasing number of virtual and/or interactive decision making aids1 can be observed in the marketplace, a purely product attribute and keyword search seems outdated and the evolution towards an integrated search2 promise a potential for differentiation in the hyper- competitive online B2C business.
Are ‘traditional’ search interfaces contingent to consumer knowledge for complex products? Do virtual product advisors facilitate search tasks? Are customers able to encode their stated needs into technical product attributes effectively? To what extent does the expertise knowledge of a customer impact the preference for a product search interface?
The research ambition of this master thesis is to answer those questions among others. The paper is centered on the product (pre-)selection process in online shops and to assess certain antecedents of customer satisfaction in regard to product search tools. A dyadic perspective on the perceived quality of the product search features is offered through marketing and information system management lenses. The research paper shall provide en epistemological contribution to academic research, practical insights for online shop operators and formulate further research questions.
I. Theoretical Framework
2. The Net Economy
During the last decade the term Net Economy evolved with such prominent example like the rise of Amazon and its far-reaching, structural impact on the book retail business. With the emergence of the online player, the ‘traditional’ book retail sales collapsed and their business model became troublesome. The rapid development of technologies such as electronic ink, e-book readers and tablet computers as well as the increasing availability of e-book content accelerated the decay of the industry in the later half.3 The phenomena can be observed for different industries in the decade due to the abundant availability of digital information and available products online, which might nowadays simply be called the Net Economy:
The Net Economy is characterized by the management of information about customers, competitors, products, prices, delivery policies and by the efficient exchange of goods and services through e-marketplaces, e-shops and e-procurement.4
2.1. THE CONCEPT OF AN ONLINE SHOP
Amazon was able to gain a competitive advantage against its physical competitors through a superior offer of product variety than any brick and mortar book store could possibly offer. Moreover, the disruption it caused in the book industry was achieved through an innovative, online purchasing process paired with a customer- oriented delivery policy (free shipment). Nevertheless, Amazon’s success heavily depended on the favorable product characteristics of books in the first place, which is to say that a book possesses perfect search attributes5. A book represents physical information, that can be perfectly digitalized and accessed effectively through a product search via the internet. The information might be coded by author, date of publication, abstracts, summaries etc., which allows a consumer to retrieve it. Thus, the information itself is the product, which is consumed. It can be summed up by the phrase ‘what you see is what you get’ (WYSIWYG). The favorable product attributes in this example allow the customers to find a desired product by searching for the information and to determine the product quality a priori.
As information is exchanged via contact points on the internet in the process of preparing and/or executing the exchange of property rights, the concept of an online shop shall be defined as following:
The term online shop or e-shop in general can be defined as “ the electronic sale of goods or services by a company via digital networks ” 6
It represents a virtual salesroom of a company, where goods and services are displayed, offered and sold. Online shops are characterized by three factors which distinguish it from physical retailers. All sales interactions between customers and sales agents are virtual and based on a human-machine-interaction, which is the constituent feature. Second, the offered goods might not only be physical/tangible, but also digital in nature. Third, the characteristics of the merchandise in turn impact the outbound logistic process of an online shop. Depending on the nature of the product, the distribution might be realized completely electronically7. The following Illustration 1 depicts the basic idea of online shop.
illustration not visible in this excerpt
Illustration 1: Principle of an Online Shop8
Regarding the offer of product variety, an online shop overcomes the constraint of limited shelf space. The variety of products, which can be virtually offered, is limitless and opens opportunities for providers in the long tail as mentioned earlier. This increase in product variety, however, leads to a complexity problem from a customer’s perspective in regard of the amount of available products. Moreover, the offered goods themselves are subject to complexity in terms of their usage and technical understanding.
The information collection of Amazon’s book catalogue and its efficient retrieval and delivery is a key success factor for its underlying real economy value chain.
Also Kotler (2003) noted that “companies can operate powerful information and promotion channels by dispensing complex product catalogues to their online customers explaining their products and services.”9 According to Kollmann (2011) online shops in general offer added-value to the consumer through overview (structuring value), choice (selection value), product advice (retrieval value) and in the procedure (transaction value) to its customers10. First, the structuring value consists of the online offer of a product catalogue, which allows a consumer to get an overview of products. Second, the selection value of an online shop creates a possibility for customers to identify products more efficiently that are suited for their needs, rather than browsing multiple websites11. The combination of the structuring and selection values can be considered as an aggregation of information, which lower the cost of information search. Third, the product advice is considered to have a retrieval value of information, which is based on the aggregated information in the database and executed by algorithmic inference processes. Last the transaction value of an online shop offer derives from efficiency and effectiveness gains through virtual transactions.12 In conclusion the new economy leads to structural market changes due to:
- a substantial increase in buying power
- a greater variety of available goods and services (long-tail)
- a great amount of information at low cost (decreased search cost)
- a greater ease in interacting, placing, receiving orders (anytime/anywhere)
- an ability to compare products and services (market transparency)13
Due to the fact that product catalogues can be complex, the search process becomes pivotal for success by turning qualified leads into paying customers, also known as conversion rate. Regarding other product categories, however, the product search process will likely be less linear. As product information might not be perfectly, digitally representative (WYSISWYG) as in the Amazon example, products can be complex themselves as mentioned before. The question arises how the product properties and its quality of product information determine a product search process and what implications it bears for satisfying the needs for information.
2.2. ELECTRONIC SALE OF PRODUCTS
Thanks to ICT, the search cost for information has dropped significantly, because the availability of information determines the economic cost of time spent to find an adequate product, which satisfies consumer needs. Directing the attention to the starting point of this consideration, the product itself has to be analyzed in terms of its suitability for electronic sale. Kollmann (2011)14 offered three evaluation criteria for the assessment of a product’s potential for electronic sale which is also termed product potential for digitalization:
(1) digital representativeness
(2) feasibility of digital evaluation
(3) need for digital product advice
First and foremost, a product shall possess characteristics that are transformable into digital information or at least a potential for the digitalization of product information. An example for the former would be the digital sale and delivery of music as mp3 files via an online shop such as Beatport15, whereas examples for the latter include products that are presented and sold digitally, but dependent on a physical delivery such as a camera purchase for instance. The technical specifications of a camera (e.g. size of a photo chip) in an online shop permit a high digital representativeness, because the functional benefits can be assessed on objective grounds via digital product information. Second, the feasibility of digital evaluation from a consumer perspective is high, if a consumer can evaluate a product solely on information provided via digital channels. Regarding technical products with a standardized quality, the evaluation is highly comprehensive but not perfect, because certain characteristics such as ergonomics are difficult to describe via text and images. Third, the need for digital product advice describes the degree of information needs concerning a product purchase from a provider’s perspective. In the case of a camera purchase, the need for information is relatively high due to the technical complexity, whereas the service of product advice can be represented via digital information. That is to say that the high digital representativeness of a camera permits a high level of virtual product advice. Based on the provided arguments, it is concluded that a camera possess a high suitability for an electronic sale and a high potential for delivering comprehensive information in an online product purchase.
2.3. SEARCH QUALITY OF A CAMERA
Nelson (1970) coined the concepts of search quality and experience quality of a good. In his terms, the search process of a product is determined by the cost of search related to price and quality from a utilitarian point of view. Nowadays, the search cost for price dropped significantly thanks to available price comparison websites on the internet, such as www.guenstiger.de in the German market, which are also referred to as meta search16 engines. In contrast, the experience cost for quality characteristics are directly linked to the price of a good, because the goods have to be purchased in order to evaluate its quality.
Considering the quality of a camera for instance, Nelson classified it as a durable search good in his study, which is to say the quality can be determined prior to purchase by means of information search.17 It is posited that consumers will search for information prior to a purchase, as the cost for experience of a digital camera is relatively high, thus it can be regarded as a high involvement purchase with a low purchase frequency. A low purchase frequency in turn suggests that consumers have less opportunities to evaluate quality in terms of experience. Based on the search quality classification, a camera’s quality can be evaluated a priori according to available information on technical product attributes. Even though, quality dimensions of a camera such as product reliability, battery life etc. can be considered experience qualities, consumers might find information in user ratings, consumer reviews etc. Nevertheless, this research project is limited to product dimensions that are regarded as search qualities. The search qualities include the following technical dimensions:
- resolution (amount of megapixels)
- camera type
- size of viewfinder
- camera ‘functions’ and ‘programs’
In conclusion the technical specifications of products requires knowledge that allows consumers to interpret them correctly. On the one hand the interpretation concerns the quality evaluation (other than a price-quality construct) of a technical product and on the other hand it also concerns the value in use or use context of the product. This relates to the user’s needs in terms of functional consequences, which is discussed in chapter 3.2.
In the following chapter, general definitions and terms will be introduced to provide a common language and understanding for the reader in the course of the research paper.
3. Definitions
In this chapter definitions for key terms and notions will be provided that are used in the following chapters of the thesis. As many different definitions exist, the provision will not be an exhaustive attempt to review all definitions and unify them, however, following a pragmatic approach, it shall offer practicable definitions, which will serve the purpose and the context of this thesis. Kollmann’s work (2003) on e-business proved to be a very rich source of definitions for basic terms regarding the Net Economy.
3.1. GENERAL TERMS
According to the Oxford American Dictionaries the term virtual means “carried out, accessed, or stored by means of a computer, esp. over a network”.
Virtual Communication refers to information network systems which are a comprised of data flows and information channels and handled through software and programs in an impersonal manner.18
Following this definition of virtual communication, the exchange of information is realized digitally, since the term digital relates to “signals or data expressed as series of the digits 0 and 1” and “involving or relating to the use of computer technology”19. Hence the notion virtual precedes the term digital, both shall be used synonymously in the course of this paper.
Virtual communication offers possibilities of communication independent of time and geographic location, which is depicted in the anytime/anyplace - matrix below.
illustration not visible in this excerpt
Illustration 2: „Anytime/Anyplace“ - Matrix20
Virtual communication is an enabler of electronic commerce. Since communication is the basis of transactions, it can be realized 24 hours a day, 7 days a week through the use of a computer anywhere in the world. For example, a consumer in Germany (GMT +1 hour) can place an order at 7 a.m. at an ebay seller residing in Los Angeles in the United States. The transaction is either realized in an automated way immediately (synchronous response) independent of ‘regular’ opening hours or processed during the opening hours (GMT -8 hours), thus triggering an asynchronous, interactive response. From the buyer’s perspective, the virtual communication can be realized at anytime and from anyplace.
Virtual communication and its presentation of information usually involves different types of media, which is referred to as multimedia. It is a concept of virtual contact design which combines different sources of information such as text based information, auditive and visual stimuli (still frames and motion pictures) in an online environment.
Multimedia is the integration of multiple sources of information and its simultaneous, conscious and real usage of at least two media in an application which offers added value in the information process21
In general terms most of the online shops can be considered as multimedia platforms, because text based and visual information usually constitute a product catalogue. On the electronic retailer Saturn’s Online Shop22 for instance, the products are displayed and described in the center of the browser window. Multimedia however should clearly be distinguished from multichannel information, as multichannel information describe the usage of various platforms to deliver information. In the mentioned example, multichannel information via external platforms such as YouTube and Facebook can be found in the sidebar of the web page, which yet offer multimedia contents on their behalf.
According to the Oxford American Dictionaries the adjective interactive means “( ) allowing a two-way flow of information between [a computer or other electronic device] and a user, responding to the user's input”. In this lexicographic definition, the focus lays on the bidirectional character of communication with the Newtonian principle of actio and reactio. Virtual interactivity shall thus be defined in the following way: Interactivity in an online environment describes the active, bidirectional communication and mutual information exchange between a sender and a receiver via a virtual contact point.23
The degree of interactivity is determined by the design of the virtual contact point, whereas interactivity can be further differentiated. True interactivity exists, when the premise of a bidirectional information exchange is fulfilled, that is to say that both agents in a communication process can assume the role of a sender and a receiver. The design of a virtual contact point determines the degree of flexibility, or in other words the extent of scripted response mechanisms. 24
For example, the pharmaceutical skin care company Avène uses a product finder tool25, which offers an interaction script through a questioned based contact point. It allows the user to ‘pull’ information from its database that are most relevant to his/her search query. The opportunity for influencing and altering the communication process (in terms of interaction), however, is limited due to the product finder’s program code. Thus, the product finder only offers limited interactivity or scripted interactivity.
An example for true interactivity is the Adobe technical support chat26, which offers a one-to-one communication interface with an Adobe technician regarding a closed universe of topics like order status, serial numbers, product activations and the like. The list of topics can be understood as a Frequently Asked Questions section and thus be limited in itself, nevertheless, the interactivity between a customer and a technician is realized in real time through natural language over a machine network. The company Whisbi27 for instance offers true interactive solutions for digital environments in a B2B setting.
3.2. COMPLEXITY
Since no absolute, generally accepted definition of complexity exists, which above all is applicable to various industries and various products, complexity can be understood as relative in nature28. According to the Oxford American Dictionaries the term complex means “consisting of many different and connected parts” or “not easy to analyze or understand”.
3.2.1. Catalogue Complexity
The complexity of a catalogue is determined by a high variety of products, which possess multiple levels of attributes. The variety of products in terms of product categories and subcategories can be understood as the different parts of a catalogue or as the catalogue width. The connected parts of a catalogue is to say that similar products with different attribute levels are subsumed under a product category or subcategory respectively, which will hence be termed catalogue depth.29 Since space is limited on a computer screen, a catalogue index of an online shop is basically structured as a list either on multiple pages or continuously. From a usability perspective, this is not very convenient as the displayed information in catalogue with great width and or depth is not easy to analyze30. As a rule of thumb, the catalogue complexity is increasing with the increasing number of offered items, variety and attribute levels. The introduction of a cut-off point for classification would appear too dogmatic at this instance, however, it is assumed that customers will use some kind of structuring approach or filter option (product search) to reduce the complexity of catalogue information to make comparisons on a rational level. Hence, a pragmatic definition of catalogue complexity shall be provided as following:
An online product catalogue can be considered as complex, if structuring and filtering tools are necessary to help customers navigate through categories, subcategories and attribute levels in order to ease analysis of information and reduce cognitive effort.
The Saturn Online Shop for instance offers 550 products in the category cameras (catalogue width), which are structured in 5 subcategories (catalogue depth). At this point the distinction between tools that are ‘nice to have’ versus a ‘must-have’ in the sense of providing an acceptable level or structure of information display shall be emphasized. In other words the appliance of filtering and structuring tools per se are not sufficient to classify a catalogue as complex. In the above mentioned example, a product search tool is necessary in terms of usability or user friendliness.
3.2.2. Product Complexity
With increasing product variety and attribute levels, a product might be considered complex from a operations management perspective. In this logic, external complexity evolving from the market is translated into complex, internal production operations of an organization31. Generally speaking, a higher product variety leads to a more complex production program. A definition solely based on this ground, however, does not appear to be satisfactory for the purpose of the thesis, and the understanding shall thus be extended by and shifted towards the concept of consumer knowledge or expertise.
The greater the amount of attribute levels, the more complex a product. Hence, the complexity of a product determines the need for information for a purchase decision. This need for information is dependent on the product expertise that a consumer possesses. It is posited that a customer, who has gained consumer knowledge through learning will require less information than a product novice. That is to say, an expert can interpret technical product attributes correctly in terms of ‘objectified needs’, whereas a novice is subject to advice through external information sources such as a sales agents or product recommendations. Objectified in this instance means, that needs can be objectively expressed in technical specifications. In conclusion, product complexity is relative and can be considered as a dyadic relationship between consumer knowledge and product variety. Focus of this thesis will be on technical products with objectified user needs that allows an interpretation in terms of attributes, hence product complexity shall be defined as following:
Products can be considered as complex, if they possess a high variety of attribute levels, which require a degree of consumer knowledge to correctly interpret them.
According to this definition, digital cameras can be considered as complex products, because the technical specifications have a high variety of attribute levels and they require consumer expertise in terms of evaluation of technical performance. The following illustration displays a system chart of the Nikon D600 Single Lense Reflex (SLR) camera, which underlines the understanding of product complexity.
illustration not visible in this excerpt
Illustration 3: Nikon D600 System Chart
(http://imaging.nikon.com/lineup/dslr/d600/compatibility01.htm)
Next, different structuring and filtering tools, which are widely in use, are defined and discussed. The following paragraphs will give an overview, how these tools cope with both product and catalogue complexity.
3.3. PRODUCT SEARCH APPROACHES
Basically four different search interface possibilities exist. In the course of the paper, the search interfaces are termed product search approaches (PSA). In practice, the use of only one type is hardly found, because the combination of the various approaches offer a more natural costumer experience regarding the search process as well as better or more effective product search results. This is referred to as a hybrid search serves as a triangulation to a search task by combining different advantages of the various methods. Nevertheless, in order to understand hybrid search, the pure archetypes of product search tools shall now be defined. Excentos’ Whitepaper on Guided Selling (2011) offered a good overview of the various search methods existent in the online marketplace.
Local Keyword Search or Full Text Search is an interface, that offers a possibility to directly type in strings32. The search engine matches the search query with hits in the full text contents as well as in the online shop taxonomy such as product tags, categories and datasheets. Usually the result list is indexed through a ranking logarithm and the best matching results are presented on top of the list, whereby the ranking depends on different variables such as keyword counts, average space between keywords etc. A ranking logarithm can be comprised of a multitude of variables and the relative importance of the variables can be determined idiosyncratically, thus it represents a domain of proprietary technology33. A very popular plugin or widget is the Google Custom Search for Websites34 which offers webpage operators the possibility to integrate Google’s search engine to ‘crawl’ local content.
Hierarchical Category Search “hierarchically organizes individual (..) categories and characteristics, starting with general main categories and moving down into sub-categories that branch off from the main categories. The list of results is filtered according to the sub-category that is selected.”35 A Hierarchical Category Search can be regarded as a decision tree with varying degrees of detail among its branches or respectively the amount of subcategories. It is also referred to as a hard search interface, because an unambiguous allocation to one category is necessary. In practice, this search interface is generally suitable for online shops with a limited amount of products. With growing complexity of the content e.g. number of available product categories and so forth, this search method’s disadvantages regarding its user friendliness become apparent. The process of categorization itself can become ambiguous, and turn into a so called ‘category war’. A hypothetical example is the widely used category ‘gifts’ in some online shops, which is a very wide and relative product term, as anything can be regarded as a gift from a user’s perspective. Nevertheless, a hierarchical category search can offer a first clue for rough filtering of a wide and deep product catalogue, as categories are mutually exclusive.
A Facet Search (FS) is an interface that allows a user to make multiple choices to refine a content search query through options (facets) set by a online shop operator. The facets relate to the content or product attributes36. For example, a FS is offered on Amazon.com to browse categories and filter product results by facets such as price range, manufacturer/brand, novelties by launching period inter alia. This search method generally performs a selection based on a ‘k.o. criteria’, which is to say that selected search facets will be matched against tags in a database on a per item basis. Facets can be understood as the front-end content attributes, whereas tags can be understood as the back-end content labeling. ‘Untagged’ content or content, which doesn’t possess all selected tags will not be displayed in the search results. Hence, the FS can be labeled as an attribute-oriented search approach. Following illustration clarifies the hard selection criteria of a FS tool.
illustration not visible in this excerpt
Illustration 4: Venn Diagram of a Facet Search
A Virtual Product Advisor37 (VPA) can be defined as a closed-question based PSA, which is centered on the user’s functional needs and translates given information input into technical product specifications or attributes. The understanding of needs within the framework of VPAs shall be classified by Kotler’s proposition that “customers are not always fully conscious of their needs, some cannot articulate needs or use words that require interpretation”38. That is to say, that consumers tend to express functional benefits or context-in-use value rather than needs. In this line of argumentation, the term needs represent a use-oriented understanding of product functionalities and their respective attributes. Thus, in the course of the research paper, the term needs refers to a request for a pragmatic utility derived from abstract attributes and thus it differs from the classical Maslow ‘an understanding of needs.
The standalone characteristic of a VPA is the assessment of a consumer’s stated, use-oriented needs through an interactive process, which allows the translation into abstract/concrete attributes. Illustration 5 below depicts a 4 step virtual product advice process as proposed by excentos GmbH:
illustration not visible in this excerpt
Illustration 5: Virtual Product Advice Process 39
Analogous to a sales talk in a physical environment, a VPA will assess a customer’s needs based on scripted interactivity. The sales process can be broken down into 4 main steps, namely (1) assessment of needs, (2) analysis of needs, (3) product recommendation and (4) provision of further relevant information. The process is interactive, that is to say the transaction partners exchange information and essentially require information input, which might be referred to as customer co-creation. Finally a product or a selection of products is recommended to a potential customer. In some cases, additional information is provided to bridge the customer’s needs and the available information in search results in order to further facilitate a decision-making process40.
In a basic understanding a VPA shifts a product search from an attribute- orientation towards a needs-orientation with a selection criteria based on either exact match (k.o. principle) or best match (soft selection). An example of this shift are closed questions, which are posed for a needs assessment, such as: “For what purpose do you want to use your pictures?”, whereas the scripted interaction offers 3 response possibilities, which are: “Print-outs in a standard format, large-scale print-outs, images for digital retouching”41
An elaborated approach of a VPA is realized by excentos’ Guided Selling technology, which realizes a soft selection in combination with a reasoning engine that explains a certain product recommendation. The reasoning engine is an artificial intelligence that manages a trade-off between expressed needs and a best match with available products in the catalogue. The satisficing principle avoids an empty result universe. Therefore, the soft selection criteria in the product recommendation step mimics a more dynamic, flexible interaction, opposed to a static product recommendation based on a k.o. principle. A practical implication is that a customer might be willing to pay (marginally) more for a best match product, than having displayed an empty result list and not buy a product at all.
In practice none of the different search methods can be found solely but rather more a combination of those. The combination of search methods adds more flexibility and a better user experience to the search queries of online customers.
Table 1 below provides a comparison between the most important points of the different search methods that have been introduced in this chapter.
illustration not visible in this excerpt
Table 1: Comparison of Product Search Approaches 42
A widely applied practice is the combination of a Local Keyword Search with one of the search methods, if not an essential feature of an online shop. For example, a keyword search allows a user to narrow down further a product selection that has been made with a FS, if the result universe is still very extensive. Many online shop operators also offer a dynamic result display which is to say that each selection criterion (category, facet, keyword string) narrows down the product catalogue in real-time. This contributes to the user’s experience and offers instantaneous feedback in a virtual interaction. Illustration 6 below displays the different product search algorithms in the Saturn Online Shop for cameras.
illustration not visible in this excerpt
Illustration 6: Saturn Online Shop Search Interface
(http://www.saturn.de/mcs/shop/kaufberater-digitalkamera.html?et_cid=26&et_lid= 53)
Stanoevska-Slabeva (2000) distinguished between 4 different kinds of catalogues43, which represent outputs of the product search - database function. On the input side, the product database can be considered a constant with the product search as an independent variable that manipulates the result list from a catalogue search (output). The 4 kinds are termed (1) attribute-based (2) constructive (3) natural language (4) consulting catalogues. Focus of the thesis are on the one hand attribute-based catalogues, where product attributes serve as an identifier (FS) and second consulting catalogues, that use an artificial intelligence for a needs analysis to provide advice on a product selection (VPA).
In conclusion a product search can be best described as an interface for customer interaction with an online shop’s product catalogue, that allows to reduce complexity in a relational database. Next, the consumer behavior shall be reviewed to draw a more holistic framework regarding product search approaches.
4. Consumer Behavior
In this chapter salient aspects and concepts of consumer behavior covering perception, memory, expertise knowledge and motivation for webpage visits will be reviewed. The different concepts will be illustrated by technical product examples in order to provide a practical link to the theoretical frameworks discussed.
4.1. INFORMATION PROCESSING STRATEGIES
Perception is a three stage neurophysiological process in which external stimuli and sensations are selected, organized and interpreted. The meaning of the stimuli is interpreted to be consistent with individual biases, needs and experiences.44 The inference of meaning is a highly subjective manner that is to say that subjects exposed to external stimuli project meaning on events either based on memory via top-down concept-driven approach or via bottom up data-driven approach45.
The perception of external sensations is managed through sensory receptors, which provide information to the cognitive apparatus for further processing. The sensory receptors are the following:
- See (vision)
- Hear (auditory)
- Smell (olfactory)
- Taste (gustatory)
- Touch (tactition)
As a practical example of a camera, the good is perceived and evaluated by the product design (vision), the weight and ergonomic design of the camera (tactition) or by the quality of sound replay (auditory), if the camera has a video function. The perceptive sensations therefore provide information to a consumer, which needs to be interpreted. Moreover, it also has to bypass a perceptual filter for further processing. A perceptual filter can be referred to as a subjective control mechanism for providing attentional resources on processing certain stimuli. In terms of a planned product purchase, for instance a consumer might be more aware and susceptible to information regarding a product of interest. This factor is also referred to as perceptual vigilance.
From a computational-thinking perspective as proposed by Atkinson, Shiffrin (1968) the selected external information is then encoded and stored in the memory and finally retrieved when necessary. A simplified information processing memory model is depicted in illustration 7 below.
illustration not visible in this excerpt
Illustration 7: The Memory Process by Atkinson, Shiffrin46
The encoding process step is influenced by many factors itself and research findings suggest that the selection and its attached meaning of information is dependent on prior knowledge and expectations47, which can be subsumed under the term experience. “Experience [ ] is the result of acquiring and processing stimulation over time, [which] is one factor that determines how much exposure to a particular stimulus a person accepts.”48 Regarding prior knowledge, it can be understood as familiarity. The more prior knowledge exist, the more familiar a consumer is within a product domain. Familiarity reduces information processing cost in terms of cognitive effort for encoding as well as the level of acceptance for external stimuli. By level of acceptance it is meant, how consumer’s embed new information into existing knowledge structures.
4.2. EXPERTISE KNOWLEDGE
Extensive research in cognitive psychology has been conducted in order to develop a conceptual model for defining expertise. The areas of problem solving, leaning and decision-making, to name a few, are touched in this wide domain of psychological studies. Without diving too far into the realm of cognitive psychology, also due to a lack of expertise on the author’s behalf, a level of expertise can be inferred from the cognitive development and the knowledge structure. Briefly, experts have “greater amounts of knowledge and better-organized structure of knowledge than do less knowledgeable .. novices”49. In the following paragraphs the link between expertise knowledge regarding a subject and the need for information will be regarded.
In general novices possess a “superficial and literal understanding of problems”, whereas experts are apt to have “an articulated, conceptual, and principled understanding” of problems. The level of skill (expertise) is determined by the experience and practice of a user within a certain realm50. Since the cognitive effort is high for goods such as consumer electronics, online shoppers will apply a problem-solving strategy that is characteristic for high involvement. That is to say that the need for information is high for decision-making processes51 for complex products, which in turn is also determined by existing consumer knowledge and expertise. The degree of expertise knowledge in terms of reasoning is dependent on the efficiency of cognitive capabilities and the skills applied to problem solving that differentiate experts from novices.52 A focus on a basic problem solving strategy such as the ‘means-end analysis’ is particularly useful in this context. The model shall be introduced now to facilitate the understanding of PSAs and the relation between functional benefits (needs) and product attributes.
The means-end chain approach proposed by Gutman (1982) is a conceptual model that addresses the linkage between values important to consumers and specific product attributes. The model consists of six levels ranging from concrete attribute knowledge to abstract self-knowledge53:
illustration not visible in this excerpt
Illustration 8: Means-End-Chain Model 54
The levels can be further grouped into grouping level distinctions (1)(2), consequence level distinctions (3)(4) and value level distinctions (5)(6)55. The three distinctive levels shall be illustrated by the means of a digital photo camera. On the distinction level, products can be grouped into categories according to their concrete and abstract values e.g. weight, ergonomic design (concrete) and quality of pictures in terms of megapixels (abstract) of a digital camera. The consequence level distinction links the product attributes with the value level of consumer self- knowledge. For instance, a camera is an instrument for external visual memory storage - that is to say snapshots - or realistic reproductions of landscapes as a professional user (functional consequence). It allows to share visual sensations (psychosocial consequence). The sharing of photographic memories (instrumental value) can be regarded as a behavior that is instrumental to achieve a preferred end-state of existence (terminal value) such as appearing as a creative individual.
Special attention shall be paid to the link between the distinctive level of attributes and the functional level of consequences. The knowledge structure at the attribute level is of prime importance to distinguish an expert from a novice, as an inference process from functional consequences to abstract attributes. The intermediate benefits derived from the functional consequences in the means-end chain define the term needs or functional benefits as termed before.
As an example, an expert will correctly infer the necessary resolution that serves the purpose for a big format print of a photograph. Moreover, he will also know, which color space to apply to the picture, in order to achieve a realistic reproduction. As previous research indicated “that experts draw more complex conceptual distinctions than novices”56, therefore, product expertise expresses itself in more sophisticated levels of abstraction and complexity57, which will be called the knowledge structure. Experts possess more detailed knowledge about concrete details of a subject. To sum up, experts are able to both form more abstract dimensions of a subject within a more general framework and to describe a subject along a wider variety of dimensions than a novice. The mentioned knowledge structure relates to the width and depth of knowledge about a subject in general terms.58 As a consequence, the expertise knowledge of customers influences their need for information. Previous research also suggests that there is an existing link between the degree of expertise knowledge and the characteristic of the provided information. Therefore, the search approach might also impact customer satisfaction dependent on the degree expertise knowledge. In this logic product information is delivered either through an attribute-oriented or through a needs-oriented product search interface.
What I will call the attribute-need inference at this point is congruent with the steps
(2)-(3) in the means-end chain model. This process is subject to learning processes based on user experience or as Gutman (1982) stated that ”all consumer actions have consequences ( ) and consumers learn to associate particular consequences with particular actions.”59
illustration not visible in this excerpt
Illustration 9: Attribute-Need Inference60
[...]
1 See also www.whisbi.com for a new generation of interactive product advice systems
2 An integrated search means the use of various search tools for the same search task
3 It is noteworthy that in 2011 the online book market grew by 5,0% and constituted 14,8% (Börsenverein des Deutschen Buchhandels, 2012) of the total revenue in Germany. This channel shift in revenue composition stresses the importance of online distribution for consumer goods.
4 Compare Kollmann (2011), p. 36 - 46
5 The product characteristics defined by Nelson et al. will be reviewed more elaborately in subchapter 2.3
6 Kollmann (2011), p. 207
7 Compare Kollmann (2011), p. 207ff
8 Illustration by Kollmann (2011), p.207
9 Kotler (2003), p. 38
10 Compare Kollmann (2011), p. 38
11 Product search as an information service can be regarded as a stand-alone business model. See URL: www.larovo.com , for a closed-question based virtual product advisor for a variety of product categories which reduces complexity within product choice.
12 Kollmann (2011), p. 38
13 Compare Kotler (2003), p. 41 and Kollmann (2011)p, 208f.
14 Kollmann (2011), p.267
15 see URL: http://www.beatport.com
16 A meta search engine is a search algorithm, which ‘pulls’ catalogue data in XML format from external websites and displays the catalogue information in a comparable output format. Their values derives from comparability through information aggregation.
17 Nelson (1970), note: Nelson noted in a revised classification that cameras should rather be classified as experience goods, however more objective information was available for cameras as search goods.
18 Compare Kollmann (2011), p. 26
19 Oxford American Dictionaries
20 Based on O'Hara-Devereaux, Johansen (1994), p.199
21 Compare to Kollmann (2011), p.28
22 URL: http://www.saturn.de/
23 Own definition
24 Based on Kollmann (2011), p.29-31
25 URL: http://www.avene.de/produktfinder#
26 URL: https://www.adobe.com/support/chat/ivrchat.html
27 See www.whisbi.com
28 Pasche (2008)
29 Compare to Magento (n.d.)
30 Compare Häubl, Trifts (2000)
31 Compare to Pasche (2008)
32 According to American Oxford Dictionary, strings are “a linear sequence of characters, words, or other data”
33 Google’s exact scoring system in its ranking logarithm is regarded a corporate secret. Many Search Engine Optimization (SEO) professionals conduct experiments to evaluate the importance of certain variables for a more favorable listing in the ‘organic’ search result list. For more information see URL: http://www.google.com/patents/US8234147
34 URL: http://www.google.com/cse/
35 Tangermann (2011), p.10
36 Based on Tangermann (2011), p.10
37 Other researchers such as Häubl, Trifts (2000) refer to it as a recommendation agent or interactive decision-making aid.
38 Kotler (2003),
39 Based on Kühn (2011), p. 3
40 Based on Tangermann (2011), p. 16-26
41 Example taken from Saturn Online VPA for digital cameras
42 Own table
43 Compare also to Kollmann (2011), p. 211ff.
44 Solomon (2011), p. 82f
45 Hoch, Deighton (1989)
46 Illustration by Solomon (2011), p. 131 based on Atkinson’s computational-thinking perspective
47 Hoch, Deighton (1989)
48 Solomon (2011), p. 102
49 Walker, Celsi, Olson (1987)
50 Hoffmann (1996)
51 Kimmel (2013)
52 Hoffmann (1996)
53 Walker, Celsi, Olson (1987)
54 Ilustration based on Walker, Celsi, Olson (1987), p.2
55 Gutman (1982)
56 Hoffmann (1996)
57 Solomon (2011) p.136
58 Walker, Celsi, Olson (1987)
59 Gutman (1982) p. 61
60 own illustration
-
-
-
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X.