Identifying the Ideal Types of Online Shoppers: A Qualitative Analysis of Online Shopping

The tremendous increase in online shopping has created a growing demand to understand online shopping behavior. This study contributes to this understanding by identifying ideal types among online shoppers. An ideal type is an analytical construct used to ascertain similarities and deviations to concrete cases in an individual phenomenon. Theoretically, the study draws from different perspectives to create a multifaceted view of online shoppers. The purpose is not to categorize online shoppers under a specific category but rather to help understand different typically occurring online shopping behaviors. Through thematic analysis of the data from 31 participants, this study presents five ideal types of online shoppers: Conservative Shoppers, Rational Shoppers, Hedonistic Shoppers, Spontaneous Shoppers, and Vanguard Shoppers. The formed ideal types serve as the main theoretical contribution of this study. From a practical standpoint, implications for online shop providers on how to accommodate the needs of each ideal type are provided.


INTRODUCTION
Understanding consumer behavior plays a crucial role in all kinds of business fields.Since shopping and purchasing have shifted from physical "brick-and-mortar" stores to online environments in recent years, it is important to further understand the motives, habits, and expectations of online shoppers.Despite some similarities between online shopping and traditional offline shopping in physical environments, the motives for online shopping can be fundamentally different (Ganesh, 2010).
The purpose of this study is to increase the understanding of online shopping behavior by analyzing consumer narratives describing their online shopping experiences and purchase paths from initial need recognition to post-purchase activities.The study brings together relevant consumer behavior literature from marketing and information systems (IS) studies to examine the antecedents of online shopping and their effects on consumers' online shopping preferences.Based on the theoretical antecedents of online shopping, ideal types (Weber, 1904a(Weber, , 1904b) ) of online shoppers are identified.
An ideal type is an analytical construct originally presented by sociologist Max Weber.It can be defined as an entity created by a researcher, with which s/he aims to first understand and then describe a multifaceted social action or phenomenon.Ideal types are internally fully coherent constructions representing reality but not found in reality as they are presented.As such, they are abstractions of reality (Giddens & Sutton, 2009;Weber, 1904aWeber, , 1904b)).Ideal types either fit or deviate from a given situation of the reality and thus cannot be understood as being empirically right or wrong.According to Weber, ideal types have a twofold function: serving as a measurement tool for reality and to contrast the created ideal types with the empirically observed phenomenon.This enables identifying the regularities and causalities within the observed phenomenon or the observed social system.According to Kaesler (2003), these regularities and causalities can be further utilized in understanding and explaining the social system with a higher level of abstraction.
The purpose of forming ideal types is to reduce the complexity of the social reality by making it easier to understand in an analytical way and to differentiate and highlight the most essential aspects of a social phenomenon (Weber, 1988).Contemplating and applying Weberian ideal types has enabled the use of fictional people in the design process and had a strong impact on designing technologies and products.From a research perspective, the concept of ideal type has been previously used, for example, in relation to consumers of sport and wellness products (Kettunen et al., 2017) and services and tourism (cf., Ahtiainen, Piirainen, & Vehmas, 2015).Although the ideal type approach can be considered as a suitable method for investigating and describing online shopping behavior as well, it has been unused in the context of online shopping.

Previous Research and the Contribution of This Study
Although online shopper typologies and categorization methods have been discussed particularly in marketing research (e.g., Bhatnagar & Ghose, 2004a;Bhatnagar & Ghose, 2004b;Brown et al., 2003;Kau et al., 2003;Rohm & Swaminathan, 2004;Barnes et al., 2007;Ganesh et al., 2010), in IS research, online shopping behavior has more generally been studied by focusing on the acceptance and use of technology (cf., Tandon et al., 2016;An et al., 2016;Yaprakli et al., 2013).A common approach for consumer segmentation in marketing is to form consumer clusters, which are usually based on the quantitative characteristics of consumers, such as their demographic or psychographic facts.
Instead of forming consumer clusters -which has been done, for example, by Brown et al. (2003), Kau et al. (2003), Rohm and Swaminathan (2004), Barnes et al. (2007), as well as Ganesh et al. (2010) -our qualitative approach and the use of the Weberian ideal types concept allows us to give a different insight to consumers' online shopping behavior, as described by consumers themselves, and thus provide a complementary typology to the ones generated by quantitative data, categorization and clustering methods.In contrast to the consumer clusters formed by clustering algorithms (e.g., Brown et al., 2003;Kau et al., 2003;Rohm & Swaminathan, 2004;Barnes et al., 2007;Ganesh et al., 2010) or consumer categories, Weberian ideal types are a logical and thus not necessary an empirical representation of consumer typologies, and they do not aim for a perfect and exclusive representation of the consumer groups.Ideal type method does not try to categorize people into one category based on their behavior.Instead, ideal type approach aims to combine sets of causal behaviors under one ideal type, leaving the possibility for people to identify themselves among more than one ideal type.In other words, a person can share qualities from several ideal types.
In this study, ideal types provide insights on consumer logic on the online shopping context; how consumers make sense of online services and involve different online providers into their lives by utilizing online offerings in their daily activities.The ideal types do not present individual participants but, as in our case, show what types of behavior related to online shopping go together with one another.Here, the purpose is not to classify each online shopper under a specific category but rather to help understand different kinds of typically occurring online shopping behaviors.Characterizing consumers in the form of ideal types advances the understanding of consumer behavior as well as its antecedents, and allows online service providers to better understand how online shops could further develop themselves in order to meet the needs of different ideal types.Marketing researchers (e.g., Heinonen et al., 2010;Heinonen & Strandvik, 2018;McColl-Kennedy et al., 2019), in particular, have recently called for deeper and truly consumer-oriented understanding, as marketing studies have traditionally investigated consumer behavior from a company's perspective -such as by measuring consumers' reactions to specific company offerings.It has been argued that identifying and comprehending customer logic "what matters to the customer in both the present and the future" (Heinonen & Strandvik, 2018, p. 10) still remains a key challenge from a managerial perspective.Furthermore, as the business landscape and consumer behavior are constantly undergoing drastic changes due to the technological development, the individual's point-ofview on the use of technological solutions should, in particular, be emphasized (Heinonen & Strandvik, 2018).Currently, there is a significant need for novel concepts and tools which enable deep insights into customers and knowledge beyond customer demographics, statistics, and other traditional measures.
By utilizing the previous research in the fields of marketing and IS, this study introduces a novel, multidisciplinary approach to online shopping behavior.As mentioned, both streams have their own traditions in studying online shopping behavior; online shopper typologies have not been discussed that profoundly in the field of IS, and in the marketing field, the research on online shopping behavior has not focused on technology and how it translates into specific user behaviors.Thus, by discussing online shopping through marketing and IS lenses and by combining theoretical concepts from both fields we can provide increased understanding of consumer behavior and are able to contribute to the research fields of both IS and marketing.Furthermore, there are no previous studies that have combined Weberian ideal types, online shopper typologies, and online shopping behavior in such a manner.
Through thematic analysis of qualitative data collected from 31 participants, five ideal types of online shoppers, including; (1) conservative shoppers, (2) rational shoppers, (3) hedonistic shoppers, (4) spontaneous shoppers, and (5) vanguard shoppers, are identified.As a result of our research, we find most of the ideal types to consider online shopping more convenient compared to physical stores, being in line with previous studies (Sareen & Jain, 2014;Yaprakli et al., 2013).As suggested by An et al. (2016), hedonic motivations are found to be a significant factor for a couple of ideal types.Finally, supported by the results of Tandon et al. (2016) and Martins et al. (2013), we also find perceived risks and habits to have an important role on the online shopping behavior of all identified ideal types.
The antecedents of online shopping selected for the theoretical framework of the study are introduced and discussed in more detail in Section 2. Section 3 presents the methodology for developing the ideal types of online shoppers, and these developed ideal types are described in Section 4. Section 5 presents the main conclusions of the study, and finally, Section 6 shortly discusses the limitations and some suggestions for future research.

ANTECEDENTS OF ONLINE SHOPPING
Based on previous IS and marketing literature on online shopping (e.g., Bilgihan, Kandampully, & Zhang, 2016;Bonera, 2011;Lee & Koubek, 2010;Chang & Chen, 2008;Dash & Saji, 2008;Pavlou, 2003), a variety of online shopping antecedents can be identified.These antecedents can be roughly divided into three main groups, including; (1) personal factors (e.g., Dash, & Saji;2008;Hsu & Chiu, 2004), (2) online store characteristics (e.g., Lee & Koubek, 2010;Chang & Chen, 2008), and (3) social factors (e.g., Cheung, Liu, & Lee, 2015;Sareen & Jain, 2014).As discussing all the antecedents of online shopping is not reasonable in one study, for this study we have chosen the antecedents which were considered the most beneficial and descriptive for identifying the ideal types and for better understanding the online shopping behavior as a multidimensional phenomenon.Similar antecedents have previously been discussed in, for example, Bilgihan, Kandampully and Zhang's (2016), and Kawaf and Tagg's (2017) studies on customer experience in an online shopping context.
First, previous research states that online shopping behavior is influenced by personal factors; the way individuals perceive the online shopping context, the benefit and risks of shopping, and themselves as online shoppers.Perceived utilitarian and hedonic value of shopping (e.g., Holbrook & Hirchsman, 1982) and perceived risk (e.g., Pavlou, 2003) have been recognized as important antecedents of online consumer behavior as they determine consumer's purchase choices.In addition, self-efficacy; individual's beliefs of one's capabilities to perform a certain task (Bandura, 1986) in an online environment, has been identified as a factor influencing one's willingness to shop online.
Second, online shopping behavior is affected by online store characteristics.Previous research (e.g., Zhou, 2011) demonstrates that website quality affects user adoption and usage, and higher quality of a website leads to more positive attitudes towards online retailers.Therefore, website usability has traditionally been seen as the most important factor when examining consumer preferences in online contexts (Lee & Koubek, 2010).In addition to usability, which often highlights the technical elements of an online store, the amusement of an online store is as an important antecedent of online shopping behavior.Previous research suggests that, for instance, gamification (Deterding et al., 2013;Kari et al., 2016) can generate more enjoyable user experiences.According to Bilgihan, Kandampully, and Zhang (2016) gamification has become increasingly relevant in online shopping due to its critical role in an online shopping experience.
Third, in addition to the personal factors and online store characteristics, consumers' online shopping behavior is influenced by social factors; the social environment in which consumption takes place.Previous research suggests that social interaction is today an important antecedent of online shopping behavior as consumers rely on their social networks, opinion leaders and online recommendations when making purchase decisions (Cheung, Li, & Lee, 2015).
The above-mentioned themes and antecedents of online shopping are in this study used as a theoretical framework for developing and describing the ideal types of online shoppers.By characterizing online shoppers through these viewpoints, this study aims at providing important insights into online shopping behavior and its multidimensionality.In the following, each of the antecedents as well as the arguments for incorporating them into the used theoretical framework are discussed in more detail.

Perceived Utilitarian and Hedonic Value
Marketing research has long recognized the duality of utilitarian and hedonic consumer value derived from both the shopping process and the intended outcome, arguing that consumer choices are driven by utilitarian and hedonic considerations (e.g.Holbrook & Hirchsman, 1982;Babin et al., 1994;Jones, Reynolds, & Arnold, 2006).However, the two dimensions are not mutually exclusive as several authors have acknowledged that shopping can also produce both utilitarian as well as hedonic value (e.g.Babin et al., 1994).
Utilitarian consumer behavior has been described as rational and task and outcome-oriented (Batra and Ahtola, 1991).The utilitarian consumers have been regarded as problem solvers who have a conscious pursuit of an intended consequence; hence, consumption is cognitively driven, instrumental and helps in accomplishing a practical task (Strahilevitz & Myers, 1998).On the contrary, hedonic consumer behavior is related to fun, pleasure, and excitement.It has been argued that shoppers oriented to hedonic value seek a pleasurable experience (Bradley & LaFleur, 2016).Hedonic value is thus more subjective and personal, reflecting shopping's potential emotional and experiential worth (Holbrook & Hirschman, 1982).
Emotions are an integral part of hedonic online shopping experience and they moderate the quality of the relationship between consumers and service providers (Sanchez-Franco & Rondan-Cataluna, 2009).In fact, Bonera (2011) stresses that either high or low hedonistic emotional commitment is vital when the consumer perceives and evaluates the usefulness of an online shop.

Perceived Risk
Perceived risk has been a diligently researched topic in online consumer behavior studies.Perceived risks can be defined as one's subjective belief of a loss followed by certain actions online (Pavlou, 2003;Pires et al., 2004;Chang & Chen, 2008).Perceived risk derives from the feeling of uncertainty that is present in the online environment, due to its distant and impersonal nature (Pavlou, 2003;Kim et al., 2008).Researchers have identified various types of risks associated with online shopping.According to Kim et al. (2008), three particular risks can be said to be the predominant ones: financial risk, product risk, and information risk.Financial risk relates to the possibility that a consumer loses their money in the transaction.Product risk derives from the possibility that the product is not as expected, being, for example, defective or the wrong size.Information risk is related to the security and privacy of transactions; for example, a consumer might fear their credit card information will be stolen.
According to Kim et al. (2008) perceived risk is a significant factor that might inhibit consumers from making online purchases.Pavlou (2003) also found that perceived risk has a negative effect on consumers' intention to transact in online shops.According to Pires et al. (2004) and Comegys et al. (2006), perceived risk influences consumers' behavior when searching for information.Consumers set purchase goals for themselves after they have recognized a need for purchasing something, and seeking to meet these goals influences the way they search for information.Pires et al. (2004) state that if a consumer's actual purchase experience is likely to be very different from the goal that they have set for themselves, the level of the perceived risk will be high.Therefore, during the information search stage consumers will try to search for information that is close to their goals, and do not seem risky (such as being too far from their purchase goals).
Perceived risk is also a factor that influences consumers' behavior when evaluating the alternatives.In the evaluation stage, the consumers weigh their options and ultimately make the decision to continue or not to continue to the purchase stage.In addition, perceived risk has an effect on post-purchase behavior.One of the important aspects of post-purchase behavior is the intention to repurchase (Comegys et al., 2006).If a consumer has a positive purchasing experience, perceived risk decreases, and they are more likely to make repurchases.Chang and Chen (2008), suggested that key aspects of reducing perceived risk were how well-known the brand is, and the quality of the website.If consumers recognize a brand as being well-established and have a high-quality website, the perceived risk will be lower.The same principle also works in reverse.

Self-efficacy
When looking at online shopping from the self-efficacy theory point of view, it can be understood that people have a perception of their capabilities of using the internet and shopping online.A person's selfefficacy refers to an individual's own beliefs in their capabilities to perform a certain task and it is developed by external experiences and self-perception (Bandura, 1986).Self-efficacy has been divided to general internet self-efficacy, which refers to perceptions of using various internet service domains in general, and to web-specific self-efficacy (Hsu & Chiu, 2004), which means an individual's perceptions of their capabilities to use a specific web domain.Whereas general internet self-efficacy is trait-oriented and is developed over time, web-specific self-efficacy is state-oriented and develops throughout using a specific web domain.Whereas the first one is more stable, the second one can vary a lot between using different web domains.
According to Dash and Saji (2008) consumer self-efficacy affects perceived usefulness, trust, and perceived risk.This has a positive influence on the intention of the consumer to purchase online.Whereas consumers with high self-efficacy feel comfortable, consumers with lower self-efficacy feel less comfortable and more uncertain about shopping online.Therefore, they need and appreciate simple methods that do not require too much knowledge.This indicates that in order to create a positive online shopping experience for people with low internet self-efficacy, the website needs to be easy to use.However, it is important to remember that people may also experience different levels of self-efficacy towards different online shops.This might be an important factor that leads to purchase decision.Website specific self-efficacy might be affected by customers' previous experiences of using that particular online shop, the familiarity of the online shop, the perceived usability of the online shop, and the feedback received from other people.Self-efficacy judgments are positively correlated to the outcome expectations of online shopping (Oliver & Shapiro, 1993).This means that the lower the person's self-efficacy is regarding online shopping the less likely he/she is to give effort on trying to achieve their goal, which could be, for example, finalizing the purchase.This is because consumers are more likely to continue to perform a task that they feel capable of doing it (Eastin & LaRose, 2000).

Usability
Usability has traditionally been the most discussed factor when examining consumer preferences in online contexts (Lee & Koubek, 2010).There is no single determinant of what makes an appealing or usable website as there are various norms of what is considered appropriate (Lee & Koubek, 2010).According to Flavián et al. (2006), usability in an online environment most commonly consists of the following elements: structural clarity of the website, simplicity, the time required to search the wanted options or content in a web site, and feeling of control.
As Flavián et al. (2006) note, usability can play a similar role as the appearance of a traditional physical store: a well-planned display and an organized look of a shop attract potential customers to go inside and to take a closer look.In an online environment, this means having all the necessary information easily accessible.Inferior usability not only frustrates users but may also give incorrect information about products or services that an online vendor is offering (Konradt et al., 2003).If a customer searching for information feels uncertain about a web page and the provided information, they are unlikely to return later.Users make choices quickly and often subconsciously (Lee & Koubek, 2010), which leaves no room for error as shoppers quickly shift their loyalty if they feel the online shop is not found useful or appealing.
By ensuring that customers find what they are looking for intuitively, without unnecessary challenges in navigation, an online shop can stand out from others.Puccinelli et al. (2009) note that a customer is not always shopping for a certain product or service as shopping can also be entertainment, for example.However, regardless of the customer's intentions, improving web site usability can have a positive impact on user experience.

Gamification
As today's consumers value the enjoyment of consuming, including the immersive and experiential aspects of shopping (Bilgihan, Kandampully, & Zhang, 2016) a potentially effective way to motivate an online shopper towards a certain behavior is making the user experience more playful and enjoyable and thus increase the hedonic value of the experience.One central way to achieve this is gamification (Deterding et al., 2013;Kari et al., 2016).As mentioned, hedonic consumer value can be derived from both the shopping process and the intended outcome, that is, the product or service.What distincts gamification's value from the general hedonic value of shopping is that instead of deriving hedonic value from the product or the service, gamification focuses on deriving hedonic value from the shopping process.
Gamification, from its two classical definitions, refers to "the use of game design elements in non-game contexts (Deterding et al., 2011, p. 1) or to "a process of enhancing a service with affordances for gameful experiences in order to support user's overall value creation" (Huotari & Hamari, 2012, p. 19).Kari et al. (2016) propose that these two definitions follow from two different perspectives on gamification and further divide gamification into "process of gamification" and the "experience of gamification", where the process refers to "using a set of activities with the aim to implement game elements to non-game context" and the experience refers to "a use experience in non-game context that the user perceives as gameful" (Kari et al., 2016, p. 395).Typically, the purpose of the process of gamification is to generate more gameful and enjoyable use experiences (i.e., the experience of gamification) and subsequently motivate the user towards desired behaviors (Deterding et al., 2013;Kari et al., 2016).
During the last 10 years, gamification has become a major model for business growth attempts in various fields and is these days widely used.Utilizing the process of gamification is largely based on the idea that games engage people and that the engagement lasts for some time.In online retail environment this often means engaging people to stay loyal to a particular online shop, thus increasing the customer retention.Customer loyalty can be improved by offering more than just the sold product or service (Insley & Nunan, 2014) and many online businesses aim to enrich the experience of online shopping by means of gamification, that is, the application of game mechanics to the shopping process.Typical ways for this are, for example, rewards, bonus programs, customer competitions and similar, which are derived from such concepts of game mechanics as status, reward, competition, and achievement (Zichermann & Cunningham, 2011).
Different businesses are utilizing gamification for different reasons and in different ways.For example, online clothing retailers have run into the problem of not being able to allow customers feel and interact with the products before purchase, which can hinder the appeal of purchasing clothing online.Insley and Nunan (2014) have suggested that utilizing gamification could be a way to combat such missing components of customer experience; user experience design can benefit by drawing from gamification, for example, by implementing a graphical user interface similar to ones in digital games.However, some game mechanics could be seen as punishing and implementing them into the online shopping process could have a negative outcome on the experience.Thus, developers need to understand their customers' needs and wants, and what is feasible to create in terms of different types of customers and gamification.

Social Interaction
The social cognitive theory by Bandura (1986) emphasized the role of social experiences and observational learning in personality development by suggesting that individual's actions, reactions, and social behavior are influenced by actions observed in other people.In online shopping context, social interactions enable consumers to perceive and evaluate themselves as members of a community (Bilgihan, Kandampully, & Zhang, 2016).Previous research demonstrates that social presence of other consumers and service representatives in an online store affect consumer's perceived usefulness, trust, and perceived risk, which have a positive influence on the intention of the consumer to purchase online (Dash & Saji, 2008).Studies also show that if an online shop uses images of humans, instead of generic cues, it stimulates higher levels of positive emotions and enjoyment (Wang et al., 2014).
Characteristics of online shopping, in general, emphasize the social aspect as well.For example, the amount of interaction among consumers regarding online shops is active nowadays with the introduction of social media, discussion forums, and review sites.The positive or negative online shopping experience can lead to word-of-mouth (WOM) behavior, which refers to information sharing among peers (Hennig-Thurau et al., 2004).According to Yeganegi and Elias (2016) WOM is positively correlated with social influence and social influence is again positively related to behavioral intention to shop online.
According to Lievonen (2017), online reviews and public discussions concerning the previous shopping experiences are important for consumers and have a direct influence on the online shops they concern.Chu & Kim (2011) argue that by commenting, liking or passing along to their social connections, consumers voluntarily display their brand preference along with their persona (e.g.name and picture).This can further engender socially constructed communication about the online shop, which, in turn, can have a direct effect on the behavior of the consumers including site revisit (Gounaris et al., 2010) and repurchase intentions (Matute et al., 2016).

METHODOLOGY
This study uses a multimethod design in data collection.By using more than one method within a research problem, a more complete picture of the phenomena can be obtained (Morse, 2003).The data for this study includes individual online shopper interviews (N = 3), small-group discussions (N = 10, 2-3 participants per group), and written personal reflections (N = 18).The information from all these three qualitative datasets was utilized as one integrated qualitative dataset in the analysis and in the creation of ideal types reported below.
Of the total of 31 participants for the study, 21 were women and 10 were men and their ages varied from 23 to 86 years.However, a majority of them were young adults aged under 35 years.In terms of the socioeconomic status of the participants, 71 % students, whereas 19 % were employed or entrepreneurs and 10 % were pensioners.More precise sample statistics of the participants can be found in the Appendix.The interviewees and the participants for the small-group discussions were recruited through university mailing lists targeted to students and staff.The interviews and small-group discussions each lasted approximately 60 minutes, and they were both recorded and transcribed.The written personal reflections were submitted as essays with an average length of about 500-1,000 words, which were collected during a marketing course held at the university.The essays were guided by open-ended questions in order to gather customer-led data.In all the cases, the participants were asked to describe themselves as online shoppers in order to find out how and why they shop online, how do they make purchase decisions, what are the perceived benefits of online shopping, and what is a good or bad online shopping experience.Focus was also on the preferences regarding the visual and technical elements of online shops as well as on their usability factors, marketing elements, and trust elements.
The analysis method used in this study was thematic analysis.It is a method for "identifying, analyzing, and reporting patterns (themes) within data" (Braun & Clarke, 2006, p. 79).Thus, it was considered suitable for studying the concept of ideal types.Thematic analysis is the most widely used analysis method in qualitative research (Guest et al., 2012), and allows describing and organizing the data set in rich detail (Braun & Clarke, 2006).It is often divided into the following six phases: (1) organizing the data, (2) generating categories or themes, (3) coding the data, (4) testing emergent understandings of the data, (5) searching for alternative explanations of the data, and (6) writing the analysis (Marshall & Rossman, 1999).
This study follows the guidelines for thematic analysis by Braun and Clarke (2006).However, as suggested by the authors, the guidelines were adopted with some flexibility in order to fit the data and the research objective.The first phase of the analysis was categorizing information received from the interviews and reflections into a table.This data included information about the participants' personal habits, motives, characteristics, likes, and dislikes.The next phase of the analysis was searching for recurring themes and recursively reviewing them in relation to the data.The baseline for searching the recurring themes was derived from the theoretical framework, and it concentrated on the more indirect psychological, behavioral, social, hedonic, and technological themes rather than on comparing the participants directly based on their online shopping activity or purchase preferences.The next phase was determining what kind of similarities and differences there were between the participants in terms of these themes.These similarities and differences then resulted into the development and description of distinct ideal types.
According to Gerhardt (1994), in ideal type analysis, individual cases are the focus of three different stages of research.First, the cases are the units of analysis used in data processing.In this stage, the cases are arranged into sequential patterns so that each case can be compared with the other cases in the data set.Second, the cases are selected based on their relative capacity for ideal type presentation.This means that when the clusters of similar cases emerge, a few paradigmatic cases can be highlighted in order to present a typical pattern.Third, the case explanations are the objective of an ideal type analysis explaining systematically the dynamics developed of the empirical cases.
Here is an example of the development process of one ideal type in order to promote the understanding of the entire process of ideal type analysis.The process begun by analyzing each participant based on the antecedents of online shopping incorporated into the theoretical framework of the study.The next phase was searching for similarities based only on one antecedent at a time, for example, the level of selfefficacy.Participants who seemed to have a very low self-efficacy were then taken under closer scrutiny.The next phase was searching for similarities within these participants.Other characteristics that were very common among these participants were the high level of risk avoidance and the high influence of website usability.The different combinations of the antecedents were presented only if they were found meaningful and common in the data.After finding the participants who shared all these characteristics, the focus then shifted onto their needs, preferences, and behaviors.This resulted into the forming of the ideal types.

FINDINGS
This study was able to identify five ideal types of online shoppers based on the characteristics of online shopping highlighted by the participants.The objective was to reveal the relationships and connections between the substantial characteristics.Thus, none of the participants were directly classified into an ideal type category, whereas the characteristics of one participant may be found simultaneously in different ideal types.Below, the five developed ideal types are described in detail, added with the citations of participants' comments matching each ideal type.Each ideal type representation includes quotes from participants whose behavior in general illustrates typical elements essential to the ideal type.

You also have to be careful about extra hidden costs. (Female, age 67)
Conservative Shoppers consider physical stores as the main source for shopping.Online shops are an alternative only when needed, and physical stores are preferred because the product is wanted to be seen and felt before making a purchase decision.Only certain purchases -which "cannot go wrong" -may be made online, and in these cases, a considerably lower price is needed.The information search and evaluation are done in physical stores, but sometimes a lower online price leads to an online purchase.Typically travel and event tickets, sometimes clothing and other low-value everyday life purchases may be made online.Conservative Shoppers' do not like to manage their life via the internet and thus also online shopping activity is low.

I want to see the product in real life. Physical stores are my thing. (Male, age 79)
For Conservative Shoppers online shopping has to be easy and simple.A purchase will not be made from an unclear website.A free, easy and fast delivery service is required, and simplicity and clarity are appreciated.A computer is used for making purchases, as a big screen is perceived more secure.Credit cards are considered unsafe and online banking is perceived as the most convenient and safest payment method.Analyzing and avoiding potential risks are in an important role of the Conservative Shoppers' decision making.Familiar and domestic online shops are preferred, whereas new or foreign shops are considered risky and troublesome and are thus avoided.Conservative Shoppers are ready to pay more for the exact same product in order to get it from a familiar shop and avoid risks related to an unfamiliar shop.

I prefer the same familiar and reliable online shop even though I know I could get the product cheaper from somewhere else. (Male, age 24)
Conservative Shoppers always base their purchases on a rational and careful decision and actual need, and they are conscious of their money.Online shopping is not considered pastime and visits to online shops have a serious aim of making a purchase.Emotional aspects are almost obsolete.Online shopping is considered personal and shopping experiences or feedback is not shared, and gamification related elements are not considered important.However, recommendations of others are considered valuable.

I'm not an emotional buyer. I don't do impulsive purchases, there is always a need. Finding the best and cheapest option after long research makes me feel like a winner. (Male, age 25)
Active but careful online shopping behavior characterizes Rational Shoppers.Various products and services are bought online, but always after careful consideration and thorough research.Behind a purchase exists an actual need and a lot of time might be used for searching information of products, payment or delivery and comparing alternatives by detail.Thus, the first visit to an online shop rarely leads to purchase.Even if a satisfying product, price and online shop has been found, it might not lead to purchase immediately, or at all.

It is easier to see the selections and compare prices and products online. (Male, age 49)
Generally, shopping does not provide pleasure for Rational Shoppers but online shopping is perceived as fast, easy and efficient.In-depth online information search and comparison between physical and online alternatives is considered necessary and needed to building trust towards the product and online shop.Also, foreign online shops are used if perceived reliable.Credit cards, and in domestic shops also online banking, are preferred as the most secure payment options.Rational Shoppers possess fairly good online skills.They get annoyed by advertisements rather than consider them beneficial.Personalized marketing and obtrusive advertising are seen as a trick to confuse the customer and thus reduce the reliability of an online shop.Advertisements work when a purchase decision already has been made.Chat services are also be perceived as an unwanted advertisement, but on the other hand, chat might be used to get the necessary information quickly.

When choosing an online shop, I appreciate simplicity and functionality but most of all the security. I always read all the certificates and other information regarding security issues. (Female, age 24)
Only if the online service or product has been exceptionally good or bad, Rational Shoppers tend to give feedback, and thus an easy channel for giving feedback is appreciated.Returning purchases is not liked but done if the product does not match the descriptions in the online shop.Rational Shoppers appreciate clear and simple online shops, which are technically and logically sound.For example, an unexpected cost in the checkout phase or other single negative experience likely leads to abandoning the online shop permanently.Thus, Rational Shoppers often end up in using only those online shops they have good experiences with.

Hedonistic Shoppers
I used to buy a lot of clothing from online shops because it felt so easy and was almost a habit.Nowadays, I don't feel like I need to buy anything but can get enjoyment from only browsing and desiring.(Female,age 26) Online shopping is easily and always available.For Hedonistic Shoppers, this offers an opportunity to shop whenever they feel like it.Online shops provide an easy channel to browse and compare products or prices without having to deal with the trouble of going to a physical store, and online shops serve to a constant source of inspiration for shopping.However, emotional impulsive shopping tends to take place more often in a physical store.Hedonistic Shoppers' online purchase is not always need-based as they are rather often grounded on personal desire, price offers and personalized advertisements.Hedonistic Shoppers order clothes, cosmetics and alike, and often similar or same products in different sizes are ordered simultaneously without an intention to keep all ordered items but to return at least some of these.Easiness and risk avoidance in the form of free delivery, easy return policy, and flexible payment methods as well as familiar domestic shop is appreciated.

When I feel sad or bored, I amuse myself by shopping without no intentions to buy. (Female, age 24)
Hedonic Shoppers see online shopping as entertainment which brings pleasure and a happy mood, sometimes reducing anxiety.Seeing pictures of the offered products and services in online shops bring joy and thus online shopping might not lead to a purchase decision immediately.As online shopping is providing an emotional experience, it is important that it is always available when needed to regulate one's emotions.Thus, information search and alternative evaluation are often done mobile but for the purchase, a computer is used.Friends' recommendations are of high value and are trusted, and also sharing shopping information with friends and in social media is an important part of the shopping experience.
Shopping needs to feel like an experience.This feeling should be easy to create with current technology.However, it is underutilized.(Female,age 42) Hedonistic Shoppers appreciate online shops that are visually attractive, easy to use and offer easy product comparison possibilities as well as gamification.Online shops have to be in line with their personal style, and good customer service is essential.An online shop should take care of the whole shopping experience starting from an initial visit to the online shop to post-purchase service in order to provide Hedonistic Shoppers a feeling that they are valued.On the other hand, a sign of bad customer service is a reason to not use the online shop anymore.An online shop should also offer various communication possibilities.Online chats are considered easy and fast and appreciated as a mean to communicate with online shops.Hedonistic Shoppers are interested in loyalty and membership programs, and thus are also loyal customers to online shops which are able to treat them well.

Spontaneous Shoppers
I am an impulsive person and make decisions fast.If I get an idea, I tend to execute it immediately.(Female,age 86) Spontaneous Shoppers' online buying behavior is emotional.They are very impulsive and active and consider online shopping easy, simple -and most importantly -a fast shopping option whenever they feel like it.The internet enables Spontaneous Shoppers to receive sales advertisements and keeps them informed of important sales.They look for cheaper prices and a broader selection and may purchase from foreign online shops' sales to get cheaper prices than in the domestic shops.Bought products include art, food, clothes, tickets and also high-end fashion.
Advertisement is a present-day phenomenon.The boundaries of privacy should be redefined again in people's own minds.(Female,age 24) An online shop is able to get the attention of Spontaneous Shoppers by advertisements.The initiative for purchasing can be created by the online shop, for example by a Facebook, email or banner advertisement.Spontaneous Shoppers don't get annoyed by such personalized advertisements but see these as opportunities, and seasonal sales are golden times for them.Spontaneous Shoppers may be encouraged by online shops to spend more by providing giveaways or by removing or reducing shipping costs along with bigger purchases.Smoothness is appreciated and the products need to be delivered fast and without trouble.Thus, products are rarely returned as it is considered troublesome.Spontaneous Shoppers rather sell or give unwanted goods to friends or family.Online shopping is not on the top priority list of discussions with friends, although suggestions and opinions on online shops are seriously considered.Spontaneous Shoppers prefer chat or call service as they appreciate a fast and personalized service, and they do not tend to give feedback unless it is of their personal benefit.Spontaneous Shoppers use various devices for online shopping and appreciate especially payment methods which give price discounts or extra bonuses for future purchases.Rewards from previous purchases are perceived positively, and generally, gamification related elements are seen as having growing importance.
When I need something, I don't have the patience to wait and get it from a store the next day.(Female,age 24) Spontaneous Shoppers appreciate online shops which are easy to use and enable customers to find information quickly without spending much effort in searching, providing a possibility to make a quick decision.Extensive visualization of the product and its end use is appreciated, for example, in the form of photos which can be zoomed-in or rotated, or show the product in actual use.Services providing a possibility to imagine a product in its own use and highlighting the product's personality, are highly valued.

I am a person who is usually among the first to test out new things. I visit various online shops just out of curiosity. (Male, age 49)
Vanguard Shoppers make most of their purchases online and are very experienced online shoppers, considering online shopping comfortable and modern.Various goods and services are bought by them from online shops, not even expensive purchases are hesitated.Online shopping is seen to provide a platform for a better offering, information search, and comparison possibilities, and it is perceived as the easier, cheaper and faster alternative.Even if a purchase is eventually made in a physical store, online shops are used for information search and evaluation.Other people see Vanguard Shoppers as forerunners always aware of new online shopping trends and able to adjust to new technical features.Vanguard Shoppers make their online purchase decisions rather on rational than emotional grounds.The decisions are based on needs, research, and evaluation instead of impulses.Thus, Vanguard Shoppers are not easily affected by advertisements, however, situations when emotions persuade to find rationality for a purchase exist.
As a consumer, I have more power when shopping online.(Female,age 24) Vanguard Shoppers are in the opinion that online shops need to make a profit, like physical stores, and thus they are not expecting big discounts from online shops.Good experiences, functional online shops and broad selections, quality products as well as novelty products make Vanguard Shoppers return to an online shop.Gamification is seen as a part of the development of online shops but perceived as neutral.Foreign shops are not separated from domestic shops and thus used as frequently.Instead, Vanguard Shoppers choose the online shop based on quality, clarity, aesthetics and nature-friendly green values.Also, price-quality relationship and fast delivery are important and might lead to a purchase in a foreign online shop instead of a domestic shop.Lack of English or native language is not a barrier if the online shop seems reliable and of high quality.Multiple devices are used and multiple payment choices are appreciated, but a credit card is preferred.
Price and quality go hand in hand.I don't expect to get big discounts from quality products.The seller needs to get some benefit as well.(Male,age 45) Vanguard Shoppers do provide online shop recommendations if needed.Instead, others' online recommendations of online shop service and functionality are considered important but reviewed skeptically if these recommendations are in conflict with their own views, instincts, and opinions.Trendsetters such as fashion bloggers may be followed in order to stay up with the latest trends and to stand out from the mass.

CONCLUSIONS
The purpose of this study was to promote the level of understanding of online shopping behavior by forming ideal types of online shoppers, which can be considered valuable for both business purposes and future research.The study was based on the data of the online shopping behaviors and preferences of 31 study participants.This data was collected by using individual interviews, small-group discussions, as well as written personal reflections, and it was analyzed by using thematic analysis focusing on the antecedents of online shopping, including personal factors, online store characteristics, and social factors.The findings of the study highlighted five different ideal types: the Conservative Shoppers, the Rational Shoppers, the Hedonistic Shoppers, the Spontaneous Shoppers, and the Vanguard Shoppers.Their differences and similarities are summarized in Table 1.As can be seen from Table 1, the high level of risk avoidance plays a key role in the behavior and decision-making of the Conservative Shoppers.Because their Internet self-efficacy (and sometimes even their computer self-efficacy) is low, the best way for the service providers to reduce their level of perceived risk is to develop simple and clear websites.For the Rational Shoppers, the clarity and usefulness of online shops play an important part, yet their main focus is on receiving a high amount of information that they feel they need, and this information must be found easily and reliably.Because advertising is not a beneficial tool for getting their attention, the best way for the service providers to attract them is by developing online shops that have both high technical quality and excellent content.The Hedonistic Shoppers consider online shopping as an entertainment experience.Therefore, visual elements and gamification are important for them and can be used for customer attraction and retainment.Because they view online shopping as a social phenomenon, social interaction is considered as a meaningful part of online shopping, and high quality and personalized service is an important factor that service providers should pay attention to.These customers can be loyal and may bring along others, but if treated badly, they do not hesitate to share their bad experiences.The Spontaneous Shoppers attach emotions to online shopping.Advertisements, even personalized ones, are seen as good service, which enables them to receive important information and do impulsive purchases whenever wanted.Ensuring fast and smooth buying and delivery process is something that service providers should focus on if they want to entice these consumers to purchase again.The Vanguard Shoppers have a very high self-efficacy for online shopping and are used to purchase from high-quality sites that offer high-quality products, independent of the origin of the online shops.Their social contribution on other consumers' purchases can be prominent; since they are often seen as pioneers, their opinions are listened to and respected.This is why it is important for the service providers to ensure they have pleasant shopping experiences by focusing on experience's overall quality.
Whereas there exist differences between the ideal types, there are also a few similarities worth addressing.In terms of online store characteristics, all ideal types appreciated clear and easy to use online shops.Whereas for some ideal types, website clarity promotes the level of visual aesthetics or enables faster purchases, for other ideal types, having a clear website is the key element determining whether the purchase will or will not be made in the first place.This supports the findings of Ha and Stoel (2009), which state that website design is the biggest factor affecting the perceived quality of online shops and that positive perceptions of quality predict the beliefs of consumers about ease of use as well as the level of trust and enjoyment.In addition, gamification elements could be perceived as positive by some ideal types as long as they do not compromise the level of website functionality or clarity.More precisely, the two ideal types among which gamification is perceived important (and also positively) are the Hedonistic Shoppers and Spontaneous Shoppers, which are also the two ideal types who are more driven by the hedonic than the utilitarian value of online shopping.This supports the notion that in an online shopping context, gamification focuses on deriving hedonic value from the shopping process.All ideal types appreciated online shopping as it was considered to offer better selections and cheaper prices in comparison to physical stores.Most ideal types also considered online shopping to be more convenient, giving them a chance to shop whenever and wherever they want.This goes together with the findings of Sareen and Jain (2014) as well as Yaprakli et al. (2013) who highlighted the importance of performance expectancy and effort expectancy as predictors of online shopping behavior.
Similarities among the ideal types were also found within the personal factors.Despite perceived hedonic value was an influential factor only for a couple of ideal types, its impact was still significant for these particular ideal types.This is in line with the findings of An et al. (2016).Perceived risks and habits, in contrast, seemed to have an important role in the online shopping behavior of all ideal types, although in different ways.This finding goes together with findings of Tandon et al. (2016) as well as Martins et al. (2013).Whereas some ideal types returned to the same online shops in order to reduce perceived risks, some ideal types liked to use familiar shops because of reasons related to quality, functionality, and purchase reward.The level of self-efficacy goes hand in hand with the concept of risk avoidance, confirmed also in the study by Hsu and Chiu (2004).The general rule seems to be that the higher the level of self-efficacy regarding online shopping, the lower the level of risk avoidance, meaning that pricier purchases can be made and less attention to be paid to the location of the online shop.Since it is hard for online shops to affect the level of consumers' self-efficacy, our recommendation is that their attention should be focused on promoting the level of online shop usability and providing fast and personalized customer service.
When comparing the five ideal types of online shoppers identified in this study to the clusters of online shoppers identified in prior studies (e.g., Brown et al., 2003;Kau et al., 2003;Rohm & Swaminathan, 2004;Barnes et al., 2007;Ganesh et al., 2010), some notable similarities can be found.For example, the Conservative Shoppers share similarities especially with the on-off shoppers and the traditional shoppers by Kau et al. (2003), the store-oriented shoppers by Rohm and Swaminathan (2004), as well as the riskaverse doubters by Barnes et al. (2007).The Rational Shoppers share similarities especially with the economic shoppers by Brown et al. (2003), the comparative shoppers by Kau et al. (2003), the convenience shoppers by Rohm and Swaminathan (2004), the reserved information-seekers by Barnes et al. (2007), as well as the bargain seekers by Ganesh et al. (2010).In turn, the Hedonistic Shoppers share similarities especially with the recreational shoppers by Brown et al. (2003) as well as the e-window shoppers by Ganesh et al. (2010), whereas the Spontaneous Shoppers share similarities especially with the information surfers by Kau et al. (2003) as well as the variety-seekers by Rohm and Swaminathan (2004).
Finally, the Vanguard Shoppers are most similar to the dual shoppers by Kau et al. (2003) as well as the open-minded online shoppers by Barnes et al. (2007).However, in addition to the above, the prior studies have also identified several clusters of online shoppers that cannot be clearly connected to any of the five ideal types of online shoppers identified in this study.Some example of these are the personalizing shoppers, the involved shoppers, and the community-oriented shoppers by Brown et al. (2003) as well as the e-laggards by Kau et al. (2003).All in all, when comparing the clusters of the prior studies to the ideal types of this study, many of them can also be seen to significantly overlap with each other in terms of one or more dimensions.For example, the balanced buyers by Rohm and Swaminathan (2004) basically resemble all the three other clusters identified in their study, whereas the overlap between the recreational shoppers, the convenience-oriented recreational shoppers, and the apathetic convenience-oriented shoppers by Brown et al. (2003) is quite evident when merely reading the cluster labels.In part, this overlap results from the traditional statistical clustering methods used in these studies, which aim at grouping the online shoppers together by averaging their observed behavior or behavioral antecedents.This often ends up highlighting more the similarities rather than the differences between the distinct clusters.In contrast, as mentioned in the introduction, the ideal type approach used in this study offers much more freedom for highlighting also the distinctions between the different types of online shoppers.As a result, the overlap between the types often becomes less of an issue.
Because of the aforementioned reasons, we see that the wider application of the ideal type approach, instead of the traditional statistical clustering methods, could benefit the whole IS field and that the current study could serve as an exemplar of how this application can be conducted in one specific IS context.From a theoretical perspective, in addition to addressing the aforementioned overlap problem, the ideal type approach also offers the benefit of highlighting that not all IS users can always be clearly allocated to one specific cluster but that they can have associations with several distinct ideal types.Moreover, these associations are not necessarily static but dynamic, meaning that they can change from one context or point of time to another.Respectively, from a more practical perspective, as mentioned also by Doty and Glick (1994), a particular benefit of ideal types is that they allow both researchers and practitioners to move beyond the limitations of the current empirical world.For example, in the context of the current study, the managers of the online stores do not necessarily have to limit their thinking of the potential marketing actions only to the empirically identified clusters of online shoppers, but they can approach the issue through the ideal types.Although these ideal types may not actually exist in the empirical world, this kind of thinking may inspire novel ideas, which can then be adapted to the realworld settings by taking into account that most online shoppers are actually hybrids of two or more ideal types rather than personifications of any single one of them.It may also open-up their thinking towards the potential contradictions and conflicts between the different marketing actions targeted at different ideal types.For example, if one marketing action is likely to cause an increase in sales among the Rational Shoppers, but a decrease in sales among the Hedonic Shoppers, whereas another marketing action is likely to have the opposite effects, how could these two be merged together so they would amplify each other's positive effects and cancel out each other's negative effects?.Overall, we can conclude that the study provides both theoretical insights for the scholars as well as practical implications for IS and marketing stakeholders in the online store setting.

LIMITATIONS AND FUTURE RESEARCH
A notable limitation of this study is the relatively homogenous group of participants.Even though the age of the participants varied from 23 to 86 years, 74 % of them were under 35 years old.However, as described by Comegys et al. (2006), young people born between 1977 and 1997, are called the "Net Generation".Since these people have been using information technology throughout their entire lives, they form the key target group for online shops.Therefore, the homogeneity of the participants in the study can also be seen as focusing on the behavior of this key consumer group.Having a larger number of female participants can also been seen as a limitation.However, due to the high total number of participants, the data set of the study also contained enough male participants in order to ensure its heterogeneity.It is also important to notice that cultural elements can play an important part when analyzing online shopping behavior.Since this study included participants from only one country, Finland, it can be suggested that similar studies should be done in other countries or even comprising a mixed set of participants that are from different countries and cultures.

Table 1 .
Main characteristics of the five ideal types