Customer Perceived Value: A Comprehensive Meta-analysis

Abstract

Customer perceived value (CPV) is a cornerstone of marketing literature. However, myriad studies have generated contradictory empirical findings. In addition, though some existing literature reviews help clarify the conceptual foundations of CPV, the literature lacks a meta-analysis of empirical evidence about the CPV model and its effects. To consolidate existing research, the current meta-analysis assesses the findings of 687 articles, involving 780 independent samples and 357,247 customers. The most integrative CPV model, which includes benefits, sacrifices, and overall value, performs best. Empirical generalizations also reveal the relative weights of various benefits and sacrifices integrated into this CPV model and causal chains between CPV and different outcomes (satisfaction, word-of-mouth, and repurchase intentions). Finally, this analysis uncovers moderating effects of multiple relational contexts: nonprofit/for-profit, public/private, contractual/non-contractual, online/offline, business-to-business/business-to-consumer, and goods/services. For scholars, this article synthesizes existing findings on CPV; for managers, the results provide suggestions for ways to increase CPV.

Graphical Abstract

Keywords

customer perceived value meta-analysis benefits sacrifices satisfaction word-of-mouth repurchase intentions

Customer perceived value (CPV) occupies a prominent place in academic research and managerial practice. Defined as a trade-off between the benefits of the offering and the sacrifices perceived by the customer (Dodds, Monroe, and Grewal 1991; Zeithaml 1988), CPV is an essential metric for successful businesses, because it functions to enhance customer satisfaction and loyalty and thus drive long-term profitability (Kumar and Reinartz 2016). As a reflection of this central managerial role, CPV also has become a cornerstone of marketing and service research. Scholars have provided critical insights into its definition (Zeithaml et al. 2020), measurement (Leroi-Werelds et al. 2014), and influence on customer decision-making (Gallarza et al. 2017).

However, these myriad studies also have generated contradictory empirical findings, suggesting the need for a meta-analysis of CPV and its effects. In particular, the marketing discipline lacks a united, conceptual view of CPV that reflects all the available empirical evidence about the construct (Grönroos and Voima 2013). Extant literature reviews provide meaningful insights into its definitions, epistemological foundations, characteristics, and conceptualizations (Table 1), but major challenges and gaps in understanding remain.

Table 1.

Reviews Dedicated to Customer Perceived Value.

Authors	Study Focus	Study Type	Empirical?	Key Insights	Differences with the Current Study
Sánchez-Fernández, Iniesta-Bonillo, and Holbrook (2009)	Organize and synthesize the major research streams on CPV	Narrative literature review	No	- Summarizes the unidimensional and the multidimensional approaches of CPV	- Does not compare different conceptualizations (i.e., unidimensional and multidimensional) of CPV - Does not provide empirical evidence
Boksberger and Melsen (2011)	Provide a comprehensive overview of CPV and identify measures associated with CPV in service context	Narrative literature review	No	- Identifies various multidimensional conceptualizations of CPV	- Does not compare different conceptualizations of CPV - Does not provide empirical evidence
Gallarza, Gil-Saura, and Holbrook (2011)	Identify research deficiencies in value-related research	Essay	No	- Clarifies conceptual and methodological links between quality, satisfaction, and value	- Does not develop and test a model between CPV and other identified constructs - Does not provide empirical evidence
Lindgreen et al. (2012)	Provide an overview of CPV in B2B markets	Narrative literature review	No	- Identifies seven areas for future research on CPV in B2B settings	- Focuses exclusively on B2B markets - Does not provide empirical evidence
Vieira (2013)	Test CPV relationships	Meta-analysis	Yes	- Tests the effects of five antecedents (quality, trust, image, risk, and expectancy) of CPV and five outcomes (word-of-mouth, satisfaction, loyalty, intention, and commitment) - Tests two dimensions of CPV (hedonic and utilitarian)	- Does not compare different conceptualizations of CPV - Does not test the relative weights of various dimensions of CPV
Kumar and Reinartz (2016)	Provide an extensive historical and theoretical overview of CPV	Essay	No	- Outlines the importance of CPV to firm performance - Links CPV to customer lifetime value	- Does not examine the effect of CPV on key outcomes (satisfaction, word-of-mouth, and loyalty) - Customer lifetime value is beyond the scope of the current study - Does not provide empirical evidence
Eggert et al. (2018)	Understand how CPV conceptualization has evolved over time in business markets	Narrative literature review	No	- Highlights the key definitions of CPV on the B2B markets	- Does not compare different conceptualizations of CPV - Focuses exclusively on B2B markets - Does not provide empirical evidence
Leroi-Werelds (2019)	Provide a theoretical update on CPV	Narrative literature review	No	- Revisits Holbrook’s value typology (14 positive and 9 negative value types) - Highlights recent (academic and managerial) changes that affect the analysis of CPV	- Does not examine the boundary conditions of CPV - Does not provide empirical evidence
Zeithaml et al. (2020)	Clarify the theoretical foundation of CPV	Integrative (qualitative) literature review	No	- Identifies different CPV conceptualizations in different paradigms (positivist, interpretive, and social constructionist) - Identifies commonalities and differences among CPV conceptualizations among different paradigms	- Does not provide a model of CPV - Does not examine CPV boundary conditions - Focuses exclusively on B2C markets - Does not provide empirical evidence
Current study	Provide and test a comprehensive model of CPV	Meta-analysis	Yes	- Compares performance of various CPV models - Weighs and compares the effects of various benefits and sacrifices related to CPV - Highlights the context-dependency of CPV’s effects	N/A

First, prior reviews are mostly conceptual (cf. Vieira 2013), so they cannot provide clear empirical evidence about CPV conceptualizations and their effects. Although Vieira (2013) offers a meta-analysis of CPV, in support of efforts to accumulate knowledge about the topic, that study does not compare different conceptualizations or clarify which CPV conceptualization is the most accurate or predictive. With the current meta-analysis, we therefore seek to establish a quantitative summary of all the collected empirical evidence on CPV across multiple studies and thereby test the theoretical basis of the construct and its underlying predictions (Grewal, Puccinelli, and Monroe 2018).

Second, we find four prominent conceptualizations of CPV in prior literature: (1) a unidimensional perspective focused on overall value, (2) a multidimensional approach centered on benefits, (3) a multidimensional version spanning both benefits and sacrifices, and (4) a combination of uni- and multidimensional elements. Because the operationalizations thus differ across studies, they generate different results that can trigger validity concerns. For example, some studies rely on a two-dimensional model of CPV that emphasizes utilitarian and hedonic value (Babin, Darden, and Griffin 1994), but others propose four- or six-dimensional models (Mathwick, Malhotra, and Rigdon 2001; Sweeney and Soutar 2001). Furthermore, some studies exclusively consider CPV-related benefits (Gallarza et al. 2017), but others include sacrifices (Petrick 2002). Therefore, today’s scholars still face challenges defining CPV and determining which dimensions to include in their models, and managers struggle to find clear guidelines for improving CPV. With our meta-analysis, we attempt to provide such guidance by comparing different models using structural equation modeling (SEM). The results of such comparisons can produce more valid conclusions than individual studies can (Grewal, Puccinelli, and Monroe 2018).

Third, contradictory empirical findings prevent a clear determination of which benefits and sacrifices to include in CPV discussions, much less what their relative weights should be. For example, Williams and Soutar (2009) report a positive effect of emotional value and a nonsignificant effect of social value on behavioral intentions, whereas Kim et al. (2019) highlight a positive effect of hedonic value but a negative effect of social value on loyalty intentions. Such inconsistency further highlights the need for meta-analyses: Because meta-analyses synthesize empirical evidence in a research domain in a way that overcomes the imperfections of individual studies (e.g., measurement errors and sampling bias) and deals with the potential influences of contextual differences, they can reveal the relative importance of benefits and sacrifices in CPV models (Hunter and Schmidt 2004).

Fourth, CPV is often depicted as highly contextual (Holbrook 1999), such that both the value and its outcomes (e.g., satisfaction, word-of-mouth [WOM], and repurchase intention) likely are influenced by contextual variables (e.g., relational contexts, such as business-to-business [B2B] vs. business-to-consumer [B2C]). However, we lack empirical evidence of such context-dependent effects, such that managers do not have guidance regarding which contextual factors to consider when deciding which benefits to provide and which sacrifices to request of customers. By identifying variance in empirical findings across studies and using moderator analysis to explain it, our meta-analysis promises to determine the generalizability of extant findings and clarify their inconsistencies (Grewal, Puccinelli, and Monroe 2018).

Therefore, to move the CPV field forward, we conduct a meta-analysis that synthesizes the empirical results of 687 articles, which report on 780 samples involving 357,247 customers. With an integrative conceptual framework, we (1) identify which CPV conceptualization has the most predictive power, (2) estimate the relative weights of the most widely examined benefits and sacrifices, (3) assess the influence of CPV on frequently examined customer outcomes (satisfaction, WOM, and repurchase intentions), and (4) highlight key contextual moderators that influence CPV and its relationship with different outcomes. We thus extend prior conceptual literature reviews (Kumar and Reinartz 2016; Eggert et al. 2018; Leroi-Werelds 2019), answer calls to reconceptualize heterogeneous empirical findings (Zeithaml et al. 2020), and support the development of a more unified view on CPV mechanisms.

Literature Review

What Is CPV?

As noted, CPV refers to the “consumer’s overall assessment of the utility of a product based on perceptions of what is received and what is given” (Zeithaml 1988, p. 14). It represents one of the most fundamental and widely studied concepts in marketing, especially in service contexts (Leroi-Werelds 2019). Several reviews have attempted to clarify its conceptualization theoretically (see Tables 1 and 2) and have established some common points (Table 3): CPV is subjective and individual, rather than objectively determined by firms; it is multidimensional rather than unidimensional; and it is context-dependent, such that CPV is a function of the situation in which the customer makes a judgment of the value being provided. Along with this consensus about the fundamental characteristics of CPV, a range of conceptualizations has been developed over time, leading to some contestation (Zeithaml et al. 2020), as we discuss next. Then, using CPV theory, we propose a conceptual framework of the direct and moderated influence of CPV.

Table 2.

Main Conceptualizations of Customer Perceived Value.

CPV Conceptualization	Example of References	Constructs Related to CPV			Main Advantage (A) and Limitation (L) of the CPV Conceptualization
CPV Conceptualization	Example of References	Benefits	Sacrifices	Overall Value
Model 1: Unidimensional	Zeithaml (1988)	—	—	Value-for-money	(A): Simple, concise, and easy to process (L): Value-for-money conceptualization is too narrow to account for all value provided by consumption experiences
Model 1: Unidimensional	Dodds, Monroe, and Grewal (1991)	—	—	Value-for-money
Model 2: Multidimensional with focus on benefits	Sheth, Newman, and Gross (1991)	Functional value, emotional value, social value, conditional value, and epistemic value	—	—	(A): Comprehensive view of benefits related to perceived value (L): No consideration of sacrifices; no aggregated measure that reflects a global value judgment
	Babin, Darden, and Griffin (1994)	Utilitarian value and hedonic value	—	—
	Holbrook (1999)	Efficiency, play, excellence, aesthetics, status, ethics, esteem, and spirituality	—	—
Model 3: Multidimensional including benefits and sacrifices	Sweeney and Soutar (2001)	Quality, emotional value and social value	Price value	—	(A): Comprehensive view of both benefits and sacrifices related to perceived value (L): No aggregated measure that reflects a global value judgment
	Petrick (2002)	Reputation, quality and emotional response	Behavioral price, monetary price	—
	Leroi-Werelds (2019)	Convenience, excellence, status, self-esteem, enjoyment, aesthetics, escapism, personalization, control, novelty, relational benefits, social benefits, ecological benefits, and societal benefits	Price, time, effort, privacy risk, security risk, performance risk, financial risk, physical risk, ecological costs, and societal costs	—
Model 4: Uni- and multidimensional	Kleijnen, de Ruyter and Wetzels (2007)	Convenience, control and service compatibility	Risk, cognitive efforts	Value-for-money	(A): Most comprehensive view of perceived value based on a combination of uni- and multidimensional (L): Complexity of the conceptualization

Table 3.

Characteristics of CPV.

Premises of CPV	References
1. CPV is the consumer’s overall assessment of the utility of a product, based on perceptions of what is received and what is given (trade-off)	Zeithaml 1988
2. CPV is subjective and individual, rather than objectively determined by firms	Holbrook 1999; Grönroos and Voima 2013
3. CPV is multidimensional rather than unidimensional	Sheth, Newman, and Gross 1991; Sweeney and Soutar 2001
4. CPV is context-dependent and a function of the situation in which the judgment is made by the customer	Holbrook 1999; Leroi-Werelds et al. 2014; Leroi-Werelds 2019

Note. Leroi-Werelds (2019) and Eggert et al. (2018) discuss these and other characteristics in more detail.

How Is CPV Conceptualized in the Literature?

For this discussion, we number the various conceptualizations of CPV that have been developed in marketing literature. Model 1 (Table 2) regards CPV as a unidimensional concept, specifically, as a general evaluation derived from a quality/price ratio, often labeled as “value-for-money” (Dodds, Monroe, and Grewal 1991; Gale 1994; Grewal, Monroe, and Krishnan 1998). This conceptualization has the advantage of simplicity, but it might not account for the theoretical complexity of CPV (Lin, Sher, and Shih 2005).

In response, multidimensional conceptualizations have become more prevalent in marketing literature. An initial multidimensional approach (Model 2) focused on identifying different dimensions or types of perceived value, akin to the benefits in Zeithaml’s (1988) definition. Studies that adopt this perspective move beyond the predominant utilitarian vision of CPV to identify affective and experiential dimensions (Babbin, Darden, and Griffin 1994; Sheth, Newman, and Gross 1991; Sweeney and Soutar 2001). A prominent and comprehensive approach, developed by Holbrook (1999), specifies three dimensions (intrinsic vs. extrinsic, active vs. passive, and self-oriented vs. others-oriented) and identifies eight value types: efficiency, excellence, status, esteem, ethics, play, aesthetics, and spirituality.

Some scholars suggest extending this multidimensional perspective to perceived sacrifices (Model 3), in line with Zeithaml’s (1988) original value definition. Many studies incorporate sacrifices into CPV conceptualization but limit their definition to monetary prices (Kumar and Reinartz 2016). Yet customers frequently bear nonmonetary sacrifices too, such as time, cognitive effort, and various types of risk (Kleijnen, De Ruyter, and Wetzels 2007; Leroi-Werelds 2019). Although efforts to account for such sacrifices better capture the complexity of customers’ value judgments, their consideration of a wide variety of benefits and sacrifices prevents these models from establishing an aggregated measure that reflects a global value judgment (Lin, Sher, and Shih 2005).

Accordingly, another option is to combine uni- and multidimensional conceptualizations (Model 4). Customers perceive both benefits and sacrifices, and they also make trade-off calculations that inform their overall value perceptions (Lin, Sher, and Shih 2005). As informed by several studies (Gallarza et al. 2017; Kleijnen, De Ruyter, and Wetzels 2007; Lin, Sher, and Shih 2005; Ulaga and Eggert 2006), this model incorporates two levels: different benefits and sacrifices that affect overall value and the effects of CPV assessed at an overall value level.

Some empirical studies compare selected CPV models (see Web Appendix A), primarily by addressing psychometric issues (e.g., reflective or formative model and first- or higher-order model). However, they do not offer evidence about which models achieve better predictive power, with the notable exception of Leroi-Werelds et al. (2014). To fill this gap, we turn to a meta-analysis, as an effective method to assess the predictive power of the different models identified.

Conceptual Framework

Building on CPV theory, we propose the model displayed in Figure 1. It builds on Model 4 and includes different benefits and sacrifices that can affect overall value, which in turn determines the different outcomes. First, CPV theory builds on Zeithaml’s (1988) review and is rooted in the positivist tradition (Zeithaml et al. 2020). It predicts that customers evaluate the benefits and sacrifices associated with a product or service to determine its overall value (Zeithaml et al. 2020). The net evaluation, that is, whether the benefits received outweigh the sacrifices made, defines the value that customers perceive (Kumar and Reinartz 2016). Higher overall value leads to increased customer satisfaction, WOM, and repurchase intentions (Vieira 2013). Satisfaction also might drive WOM and repurchase intentions (Keiningham et al. 2018). Although our model includes such effects, we do not propose formal hypotheses about these main effects, which already have received support in previous empirical studies.

Figure 1.

Conceptual model of customer perceived value.

Second, CPV theory emphasizes the contextual nature of the CPV construct (Holbrook 1999; Leroi-Werelds 2019). Customer expectations of benefits and sacrifices are subjective and vary in their relevance. Brady and Robertson (1999, p. 472) similarly conclude that “the weight attributed to either the benefit or the sacrifice side of the model appears to vary across individuals and service settings.” Individual customer differences are difficult to test as moderators in a meta-analysis, but with the comprehensive database we gather for our meta-analysis, we can assess the effects of CPV across various relational contexts. We define a relational context specifically, as distinctive from other different contextual factors that previous studies have investigated as influencing CPV. For example, the CPV and customer experiences likely differ for nonprofit organizations (vs. for-profit organizations), in public (vs. private) contexts, in contractual (vs. non-contractual) contexts, in online (vs. offline contexts), in B2B (vs. B2C contexts), and for goods (vs. services). Furthermore, we control for the influence of other, largely methodological differences that may influence the relationships in our model (Blut, Wang, and Schoefer 2016).

Conceptualization and Effects of CPV

In Table 4, we list and define the constructs included in our conceptual model. We rely on the typology of 14 benefits and 10 sacrifices developed by Leroi-Werelds (2019), which constitutes a robust framework of customers’ value judgments and provides practitioners a fine-tuned diagnostic tool. Leroi-Werelds’s (2019) typology also reflects a conceptual approach that is widely accepted among scholars (Holbrook 1999; Zeithaml et al. 2020) and that currently represents the most recent academic advances pertaining to the conceptualization of CPV. Accordingly, we include the following benefits in our proposed framework: convenience, excellence, status, self-esteem/esteem, enjoyment/play, social benefits, aesthetics, escapism/spirituality, personalization, control, novelty, relational benefits, ecological benefits/ethics, and societal benefits/ethics. We also account for the following sacrifices: price, time, effort, privacy risk, security risk, performance risk, financial risk, physical risk, ecological costs, and societal costs.

Table 4.

Constructs, Definitions, and Aliases Related to the Customer Perceived Value Model.

Construct	Definition	Alias(es)
Benefits
Convenience	Offer that adds to customer’s comfort and/or makes his life easier (Leroi-Werelds 2019)	Efficiency, utilitarian value, and ease-of-use
Excellence	Product(s) and/or service(s) of high quality (Mathwick, Malhotra, and Rigdon 2001; Leroi-Werelds 2019)	Quality, productivity, reliability, and confidence in product quality
Status	How an offer can be actively manipulated by the customer to make a favorable and positive impression on others (Holbrook 2005)	Reputation and impression
Self-esteem/esteem	Global perception of the self (Leroi-Werelds 2019)	—
Enjoyment/play	Intrinsic enjoyment derived from engaging in activities that are absorbing (Leroi-Werelds 2019)	Affective value, hedonic value, fun, pleasure, joy, and negative emotions (r)
Social benefits	Improving relationship with other customers (Leroi-Werelds 2019)	Social value
Aesthetics	Passive reaction of the customer to the salient sensory elements of an offer (Mathwick, Malhotra, and Rigdon 2001)	Product design, facilities, and physical environment
Escapism/spirituality	Allowing the customer to relax and temporarily escape from reality or daily routine (Mathwick, Malhotra, and Rigdon 2001)	Escapism, sacredness, and spirituality
Personalization	Degree to which an offer is tailored to meet the unique needs of each customer (Leroi-Werelds 2019)	Customization and co-creation
Control	Competence, superiority, and mastery of the customer over the environment (Leroi-Werelds 2019)	Self-efficacy
Novelty	Creation of curiosity, satisfaction of a desire for knowledge, and/or alleviating boredom (Leroi-Werelds 2019)	Epistemic value (i.e., desire for knowledge)
Relational benefits	Better and trustful relationship with the provider (Leroi-Werelds 2019)	Trust, commitment, and engagement
Ecological benefits/ethics	Customer’s attitude toward environmental concerns (e.g., pollution, energy conservation, and resource waste; Leroi-Werelds 2019)	—
Societal benefits/ethics	Societal well-being and/or pursuit of morality (Leroi-Werelds 2019)	Fairness, solidarity, and community support
Sacrifices
Price	Price of an offer encoded by the customer (Zeithaml 1988)	Price, monetary costs, monetary price, perceived price, and price value (r)
Time	Customers’ perceptions of the required time when interacting (i.e., buying, using, etc.) with the offer (Leroi-Werelds 2019)	Waiting cost and time cost
Effort	Customers’ perceptions of the required efforts (cognitive and physiological) when interacting (i.e., buying) with the offer (Leroi-Werelds 2019)	Perceived effort, cognitive cost, and switching cost
Privacy risk	Uncertainty of disclosing personal information (Leroi-Werelds 2019)	—
Security risk	Uncertainty of protection and safety during the transaction and usage (Leroi-Werelds 2019)	—
Performance risk	Uncertainty of offer performance and the losses if the offer fails to meet customers’ expectations (Lei, de Ruyter, and Wetzels 2008)	—
Financial risk	Potential economic-related losses if the offer does not perform as expected (Lei, de Ruyter, and Wetzels 2008)	Concerns about costs
Physical risk	Offer that is unsafe and/or causes physical harm to the customer (Leroi-Werelds 2019)	—
Ecological costs	Negative impact on environmental well-being (Leroi-Werelds 2019)	—
Societal costs	Negative impact on societal well-being, such as child labor (Leroi-Werelds 2019)	—
Overall value	Customer’s overall assessment of the utility of a product based on perceptions of what is received and what is given (Zeithaml 1988)	Overall value, perceived value, and global perceived value
Outcomes
Satisfaction	Customer’s evaluation of a specific transaction (Bolton and Drew 1991)	—
Word-of-mouth	Spread of information about products, services, companies, sales, or customer managers from one customer to another (Leroi-Werelds 2019)	Intentions to recommend, online word-of-mouth, advocacy and intentions
Repurchase intentions	Likelihood of using a service provider again in the future (Jones, Mothersbaugh, and Beatty 2003)	Purchase intentions, patronage intentions, and loyalty intentions

In line with CPV theory, we complement this model by integrating overall value. Consistent with influential studies on CPV (Zeithaml 1988; see also Model 4 in Table 2), overall value represents a more abstract construct, distinct from but complementary to benefits and sacrifices. Lin, Sher, and Shih (2005) explain that models without overall value are incomplete, because they cannot truly map the customer “give–get” trade-off experience that defines CPV. According to the principle of maximizing utility, perceived benefits positively influence overall value, whereas perceived sacrifices negatively influence it (Kumar and Reinartz 2016).

Our model also includes three outcomes, reflecting theory that suggests that CPV directly and positively influences outcomes. To ensure sufficient effect sizes, we select the most widely examined outcomes in value literature: satisfaction, WOM, and repurchase intentions (Lin, Sher, and Shih 2005). This selection also reflects the central role of these outcomes in service research. In addition to their direct relationships with overall value, we predict relationships among these outcomes, in that satisfaction positively and directly influences WOM and repurchase intentions (Hutchinson, Lai, and Wang 2009).

Moderating Effects of Relational Context

The effects of benefits and sacrifices on overall value, as well as the effects of overall value on outcomes, likely vary across contexts (Brady and Robertson 1999; Zeithaml et al. 2020). Our selection of moderator variables is guided by theoretical CPV considerations, as well as differences across studies that we identified by coding our data set, which spans research published in the past 30 years. We observe that CPV has been examined using data gathered from both nonprofit and for-profit organizations, in public or private contexts, in both contractual and non-contractual contexts, online and offline, in B2B or B2C settings, and for goods and services. All these contextual variables have potential implications for both elements of our framework, that is, how benefits and sacrifices affect overall value and then how overall value affects outcomes. We derive hypotheses about the specific moderating effects that we anticipate are theoretically and managerially relevant (Table 5), in line with the predictions of CPV theory; however, we test moderating effects for all the benefits and sacrifices in our model.

Table 5.

Overview of Hypotheses.

	Nonprofit (vs. For-Profit)	Public (vs. Private)	Contractual (vs. Non-Contractual)	Online (vs. Offline)	B2B (vs. B2C)	Goods (vs. Services)
Benefits → overall value
Convenience/efficiency → overall value	H1b (NP < FP)		H5b (CO < NC)	H7a (ON > OF)	H9 (B2B > B2C)
Excellence → overall value	H1b (NP < FP)				H9 (B2B > B2C)
Status → overall value	H1a (NP > FP)		H5a (CO > NC)			H11a (G < S)
Social benefits → overall value		H3a (PU > PR)		H7b (ON < OF)
Enjoyment/play → overall value		H3b (PU < PR)
Aesthetics → overall value
Personalization → overall value			H5a (CO > NC)	H7b (ON < OF)
Novelty → overall value		H3b (PU < PR)
Relational benefits → overall value			H5a (CO > NC)		H9 (B2B > B2C)	H11a (G < S)
Sacrifices → overall value
Price → overall value
Effort → overall value		H3c (PU < PR)	H5c (CO < NC)			H11b (G < S)
Risk → overall value				H7c (ON > OF)
Overall value → outcomes
Overall value → satisfaction, WOM, and repurchase intentions	H2 (NP > FP)	H4 (PU < PR)	H6 (CO < NC)	H8 (ON > OF)	H10 (B2B > B2C)	H12 (G > S)

> Benefits and overall value (sacrifices) have a stronger positive (negative) effect in this context; < weaker effect.

Nonprofit vs. For-profit Contexts

Nonprofit organizations operate for collective, public benefits, often by providing charitable offerings; for-profit organizations aim to generate profit with conventional offerings (Hull and Lio 2006; Marwell and McInerney 2005). Consistent with CPV theory, the importance or weight of certain benefits likely varies between these contexts. In particular, nonprofit organizations prioritize interactions with users, based on shared values and prioritization of societal, ethical, or psychological well-being benefits (Marwell and McInerney 2005). Customers often take pride in supporting nonprofit organizations and do so to enhance their social standing (Arnett, German, and Hunt 2003). However, they also anticipate that for-profit organizations provide better service, due to their greater expertise (Hull and Lio 2006). Thus, social rewards that reflect status-related benefits may be more influential in driving overall value in nonprofit organizations, whereas competence-related benefits, such as convenience/efficiency and excellence, may be more relevant in for-profit organizations.

H_1a. For customers of nonprofit organizations, status has a stronger positive effect on overall value than it does for customers of for-profit organizations.

H_1b. For customers of for-profit organizations, convenience/efficiency and excellence have stronger positive effects on overall value than they do for customers of nonprofit organizations.

With respect to customer outcomes, CPV theory suggests that the influence of overall value on different outcomes varies across contexts too. Customers likely appreciate the overall value they receive in nonprofit due to the charitable nature of their business mission (Marwell and McInerney 2005). Because they are evaluated more favorably, we expect overall value to have a stronger positive impact on customer outcomes (i.e., satisfaction, WOM, and repurchase intentions) in nonprofit contexts.

H₂. For customers of nonprofit organizations, overall value has stronger positive effects on satisfaction, WOM, and repurchase intentions than it does for customers of for-profit organizations.

Public vs. Private Contexts

Public and private contexts can be considered opposing poles of a continuum, based on the degree to which buying or consuming is witnessed by other customers (Kulviwat, Bruner, and Al-Shuridah 2009). In comparison with private contexts, public contexts—such as sporting events, restaurants, and movie theaters—are characterized by the presence of other customers who can influence interactions with the product or service and thereafter affect customer judgments (Grove and Fiske 1997). Consistent with CPV theory, we posit that the importance of benefits and sacrifices varies between public and private contexts. In a public context, customer choices are influenced by the desire to convey a certain impression or adhere to social norms (Blut, Wang, and Schoefer 2016). Accordingly, in public contexts, social benefits provided by the firm become more relevant for customers, whereas they hold less significance in private contexts, where customers primarily expect enjoyment and novelty from the activity. Thus, they personalize their spaces, select products and services that cater to their specific tastes, and align with their unique preferences. The effort required to consume the offering might be perceived less negatively in a public, compared with a private context, because the presence of other people can generate social motivation, social reinforcement, or positive distraction (Blut, Wang, and Schoefer 2016). Thus, the negative effect of this sacrifice on overall value likely is stronger in private than in public contexts.

H_3a. For customers in public contexts, social benefits have stronger positive effects on overall value than they do for customers in private contexts.

H_3b. For customers in private contexts, enjoyment/play and novelty have stronger positive effect on overall value than they do for customers in public contexts.

H_3c. For customers in private contexts, effort has a stronger negative effect on overall value than it does for customers in public contexts.

Regarding customer outcomes, CPV theory suggests that public versus private contexts can influence customers’ consideration of overall value in their decision-making and judgment. In a public context, customers may feel observed by others and thus find it challenging to appreciate their consumption experience, because they lose some ability to focus on and appreciate the overall value, which also implies weaker effects on outcomes (Grove and Fiske 1997; Stieler, Weismann, and Germelmann 2014). In a private context, such influence does not exist, and customers can focus on and fully appreciate the value they receive. Therefore, the absence of social presence in private contexts should strengthen the positive effects of overall value on various customer outcomes.

H₄. For customers in private contexts, overall value has stronger positive effects on satisfaction, WOM, and repurchase intentions than it does for customers in public contexts.

Contractual vs. Non-contractual Contexts

Another moderator is the contractual nature of a commercial relationship (Lemon, White, and Winer 2002). This contextual characteristic refers to whether the relationship between customers and the company is governed by a contract (Pick and Eisend 2014). In non-contractual settings, customers can easily switch between businesses, so they likely seek to identify the most convenient offer, a search process that entails significant effort for customers. Consequently, these benefits and sacrifices have more relevance in non-contractual settings. In contrast, the contractual context limits opportunistic customer behavior: Customers cannot change their purchase behavior without informing the company (Lemon, White, and Winer 2002). To offset this lack of flexibility, customers expect some benefits, including relationship and social advantages, such as personalization, relational benefits, and status. In line with CPV theory, these factors should exert a stronger positive effect on overall value in contractual settings.

H_5a. For customers in contractual contexts, personalization, relational benefits, and status have stronger positive effects on overall value than they do for customers in non-contractual contexts.

H_5b. For customers in non-contractual contexts, convenience/efficiency has a stronger positive effect on overall value than it does for customers in contractual contexts.

H_5c. For customers in non-contractual contexts, effort has a stronger negative effect on overall value than it does for customers in contractual contexts.

With respect to customer outcomes, we note that contractual contexts are characterized by switching costs, so the impact of overall value on customer outcomes may be weaker. The contracts create barriers to switching and encourage relationship continuation (Pick and Eisend 2014). However, in non-contractual contexts, overall value is a more crucial evaluation factor, with a stronger potential to affect satisfaction, WOM, and repurchase intentions (Lam et al. 2004). Therefore, in line with CPV theory, we predict:

H₆. For customers in non-contractual contexts, overall value has stronger positive effects on satisfaction, WOM, and repurchase intentions than it does for customers in contractual contexts.

Online vs. Offline Contexts

The effects of benefits and sacrifices on overall value may differ depending on whether the offering is purchased online or offline (Harris et al. 2006). The Internet can enhance convenience and facilitate customer purchases, but offline settings provide opportunities for face-to-face interactions with other customers or company staff, through which customers can derive social benefits and access personalized information (Laroche et al. 2005). Furthermore, purchasing offline can reduce risk, because customers can more easily assess the offering in person.

H_7a. For customers in online contexts, convenience/efficiency has a stronger positive effect on overall value than it does for customers in offline contexts.

H_7b. For customers in offline contexts, personalization and social benefits have stronger positive effects on overall value than they do for customers in online contexts.

H_7c. For customers in online contexts, risk has a stronger negative effect on overall value than it does for customers in offline contexts.

Regarding customer outcomes, we anticipate that the effect of overall value on customer outcomes also differs between online and offline contexts. Specifically, customers in online contexts generally sense greater control, due to the availability of goal-oriented attributes like greater product selection, accessibility, convenience, and price comparisons that help them achieve desired results (Harris et al. 2006). Accordingly, online customers also feel more responsible for the success or failure of the transaction, and so they place greater importance on the overall value they receive when assessing their satisfaction and considering their WOM and repurchase intentions.

H₈. For customers in online contexts, overall value has stronger positive effects on satisfaction, WOM, and repurchase intentions than it does for customers in offline contexts.

B2B vs. B2C Contexts

Relative to B2C customers, B2B customers likely possess more expertise and should be guided by more rational criteria (Ulaga and Eggert 2006). According to CPV theory, certain benefits thus may be of greater relevance in B2B contexts, in that the offering itself and the supplier’s capacity to innovate are important to customers, as is customer support, as a means to increase the ease of product/service usage (Pick and Eisend 2014). Thus, excellence and convenience benefits should hold more significance for these business customers. They also likely focus on quality and seek to develop long-term relationships with suppliers (Ulaga and Eggert 2006). Accordingly, relational benefits should drive overall value too.

H₉. For customers in B2B contexts, convenience/efficiency, excellence, and relational benefits have stronger positive effects on overall value than they do for customers in B2C contexts.

These rational and relational perspectives suggest that the overall value provided in B2B contexts may carry more weight in determining customer outcomes than it does in B2C contexts (Pick and Eisend 2014; Ulaga and Eggert 2006). Because B2B customers assess their relationship with the supplier carefully, they place greater emphasis on the overall value they receive when determining their satisfaction and intentions to engage in WOM or make repurchases (Ulaga and Eggert 2006).

H₁₀. For customers in B2B contexts, overall value has stronger positive effects on satisfaction, WOM, and repurchase intentions than it does for customers in B2C contexts.

Goods vs. Services Contexts

Finally, the goods context refers to a model of exchange based on physical products, whereas services entail a model of exchange based on intangible processes (Vargo and Lusch 2004). Product customers receive goods; services customers function as co-producers of value (Vargo and Lusch 2004). Service production and delivery also occur simultaneously, and as Parasuraman, Zeithaml, and Berry (1985) argue, due to this inseparability, customer participation can affect value creation. Customers also can interact with front-line staff and other clients to derive greater relational benefits and status from services rather than from goods. Because of their active role though, services demand greater effort from customers than goods do (Vargo and Lusch 2004). Therefore, we propose:

H_11a. For customers of services, relational benefits and status have stronger positive effects on overall value than they do for customers of goods.

H_11b. For customers of services, effort has a stronger negative effect on overall value than it does for customers of goods.

Turning to customer outcomes, CPV theory suggests that the differences between goods and services influence the importance of overall value for driving customer outcomes. Zeithaml and Bitner (2003, p. 359) argue that “because service customers must participate in service delivery, they frequently blame themselves (at least partially) when things go wrong.” Therefore, both customers’ participation and the variability of service performance can affect the significance of overall value and its impact on outcomes. In turn, the relationship between overall value and customer outcomes should be stronger in goods than in services contexts.

H₁₂. For customers of goods, overall value has stronger positive effects on satisfaction, WOM, and repurchase intentions than it does for customers of services.

Control Variables

We control for several methodological differences in our meta-analysis, according to journal quality, year of the study, sample composition (student vs. non-student), and the nature of the size effect (correlation vs. converted regression coefficient; Hunter and Schmidt 2004).

Method

Literature Search and Criteria for Inclusion

To compile the database for this meta-analysis, we first applied pertinent keywords, such as “customer perceived value,” “perceived value,” “overall value,” “consumer value,” “customer value,” “functional value,” “emotional value,” “social value,” “price value,” “monetary value,” “experiential value,” “consumption value,” and “shopping value,” to search different electronic databases, including ABI/INFORM, ProQuest, and EBSCO (Business Source Premier). We also searched Google Scholar for unpublished studies, such as working papers and conference papers. Next, we examined the reference lists of the collected studies to identify any missing contributions and citations to key studies. We also reached out to authors to request their correlation matrices and unpublished studies.

When selecting studies for our meta-analysis, we applied two inclusion criteria. First, we sought studies that measured CPV as defined in Table 4 and at least one other construct of the meta-analytic framework. Second, the studies needed to be quantitative and report information that could be transformed into effect sizes to describe the relationship between CPV and other constructs. We excluded qualitative studies and those that did not report on the relevant relationships or effect sizes. Through this search, we identified 687 articles that examined CPV, using 780 independent samples; they reported results involving 357,247 customers. The list of studies is provided in Web Appendix B.

Coding of Studies

Four coders extracted the required information, including the coded effect sizes, sample sizes, reliability, and moderating effects, as well as study characteristics (e.g., context). Following Lipsey and Wilson (2001), we coded effect sizes for the entire data set when feasible. For studies that presented separate subsample or experimental results, we treated each subsample as an independent study. Three coders then classified the effect sizes using the construct definitions in Table 4. Each coder assessed approximately two-thirds of the effect sizes, ensuring that each effect size was independently evaluated at least twice. The agreement rate among the coders was excellent, with a Cohen’s kappa value of 0.91 (>0.81). Any inconsistencies among the three coders were resolved through discussion with a fourth coder. The final sample consists of 5969 correlations.

The coders also gathered information about the study context and moderators, using a predefined coding scheme (Table 6). The relational moderators were dummy-coded (e.g., 1 = contractual context and 0 = non-contractual). For method moderators, we also used dummy-coded variables, except for the study year and journal quality. The journal quality rating was obtained from the Association of Business Schools (2015). Regarding the study context, 6% examined nonprofit contexts and 94% focused on for-profit contexts; 54% examined the public context, whereas 46% focused on the private context; 22% addressed contractual contexts and 78% examined non-contractual contexts; 24% considered online, but 76% focused offline; 7% investigated on B2B, and 93% focused on B2C contexts;¹ and 19% addressed goods, whereas 81% dealt with services. Moreover, 60% of the studies reported correlations, and 40% reported information that we could convert into correlations. In terms of sample composition, 14% consisted of student samples, whereas 86% included non-student samples. Regarding journal quality, 11% of the studies were from 4-rated publications, 21% from 3-rated publications, 26% from 2-rated publications, 15% from 1-rated publications, and 27% were unrated. The studies were published between 1994 and 2023.

Table 6.

Coding of Moderators.

Moderator	Operationalization
Relational moderators
Nonprofit vs. for-profit	Dummy variable: Nonprofit (1) or for-profit (0)
Public vs. private	Dummy variable: Goods for public (1) or private (0) consumption
Contractual vs. non-contractual	Dummy variable: Relationship is contractual (1) or non-contractual (0)
Online vs. offline	Dummy variable: Business is online (1) or offline (0)
B2B vs. B2C	Dummy variable: Business is B2B (1) or B2C (0)
Goods vs. services	Dummy variable: Business is goods (1) or services (0)
Method moderators
Correlation vs. regression	Dummy variable: Effect size is a correlation (1) or converted regression coefficient (0)
Student vs. non-student	Dummy variable: Student (1) or non-student sample (0)
Journal quality	Rating of the publication outlet: Low (1) to high (4) quality (Association of Business Schools, 2015)
Year	Year of the study

Effect Size Integration

The effect size we examine is the correlation coefficient (r), which is a scale-free measure that allows for the integration of various effect sizes and is commonly reported. We also converted other statistical information, such as t-tests, into correlation coefficients. To convert regression weights into correlations, we used Peterson and Brown’s (2005) formula. If any independent sample included multiple correlations for the same relationship, we calculated the average correlation to avoid giving disproportionate weight to that sample.

To integrate the effect sizes across the collected samples, we followed the approach of Hunter and Schmidt (2004) for a random-effects meta-analysis. This approach accounts for measurement reliability and sampling errors. We corrected the correlations for measurement error by dividing each correlation by the square root of the product of the reliabilities of the two constructs involved (Hunter and Schmidt 2004). In addition, we weighted the corrected correlations by the sample size to correct for sampling error. Finally, we calculated the standard errors and 95% confidence intervals for the sample size-weighted and reliability-adjusted correlations (rwc).

We relied on several statistics to assess the practical significance and distribution of the observed correlations. The R² values for each correlation indicate the proportion of variance explained. The Q-test of homogeneity, a chi-square test, provides an assessment of whether the effect sizes are homogeneous or heterogeneous. A significant Q-test suggests the need for moderator analyses. We calculated credibility intervals to determine the variance in effect sizes; wider intervals indicate greater variability. To evaluate the robustness of the results and the potential for publication bias, we calculated the fail-safe N (FSN), or how many studies with null results would be needed to render a significant correlation nonsignificant (p > .05). Its tolerance can be assessed using the formula 5 × k + 10, where k represents the number of correlations. Furthermore, we examined funnel plots of effect sizes plotted against sample sizes for potential publication bias, particularly if they exhibited asymmetry. The trim-and-fill method helped us detect and adjust for publication bias (Grewal, Puccinelli, and Monroe 2018).

Path Model Estimation

We used SEM to examine the various construct conceptualizations in prior literature (Figure 2). Specifically, we compared a unidimensional conceptualization (Model 1) with a multidimensional approach focused on benefits (Model 2), a multidimensional approach incorporating both benefits and sacrifices (Model 3), and the proposed integrative model with benefits, sacrifices, and overall value (Model 4). For this analysis, we prepared a correlation matrix that encompassed the 12 constructs of the meta-analytic framework (Web Appendix C). We included only those constructs that allowed for the compilation of a complete correlation matrix. Accordingly, we merged the different facets of risk to incorporate them into the analysis. Furthermore, we calculated the harmonic mean of all sample sizes (N = 4,611) and used it, along with the correlation matrix, as input for our analyses in LISREL 9.2. The error variances of the constructs were set to 0, because we already had adjusted the correlations for measurement reliability during the integration of effect sizes.

Figure 2.

Comparison of different conceptualizations of customer perceived value.

Moderator Analysis

We employed meta-regression (Grewal, Puccinelli, and Monroe 2018) to test the moderators in our meta-analysis. That is, we regressed the effect sizes on various moderator variables. The moderators included the coded relational moderators: nonprofit versus for-profit, public versus private, contractual versus non-contractual, online versus offline, B2B versus B2C, and goods versus services. We also considered the methodological moderators: correlation versus regression, student versus non-student, journal quality, and study year. Thus, we can test the unique impact of each moderator while controlling for the influences of the others. We ran 15 regression models, each corresponding to a relationship with sufficient effect size, to test the moderators for the relationships of benefits/sacrifices with overall value, as well as between overall value and the three outcomes (i.e., satisfaction, WOM, and repurchase intentions). To assess multicollinearity, we calculated variance inflation factors for each regression model.

Results

Relative Weights of Determinants and Effects of Value

The results of the effect size integration in Table 7 detail the average correlations of benefits, sacrifices, and overall value, determined with Model 4 in Figure 2. Recall that we included all 14 benefits and 10 sacrifices proposed by Leroi-Werelds (2019). Although some selected benefits and sacrifices have been addressed frequently in prior literature, others have been studied less intensively. As detailed in Web Appendix D, most significant relationships are robust to publication bias, with FSN values much greater than the tolerance levels. Furthermore, the funnel plots do not indicate any publication bias (Web Appendix E). However, the significant chi-square tests of homogeneity suggest the presence of moderators.

Table 7.

Descriptive Results on CPV Model.

IV	DV: Overall Value				DV: Satisfaction				DV: Word-Of-Mouth				DV: Repurchase Intentions
IV	k	rwc	CI_95-	CI₉₅₊	k	rwc	CI_95-	CI₉₅₊	k	rwc	CI_95-	CI₉₅₊	k	rwc	CI_95-	CI₉₅₊
Benefits
Convenience/efficiency	73	0.51*	0.45	0.56	116	0.51*	0.47	0.55	30	0.46*	0.36	0.56	172	0.47*	0.43	0.52
Excellence	221	0.55*	0.52	0.59	172	0.55*	0.52	0.59	33	0.36*	0.22	0.49	198	0.49*	0.45	0.53
Status	27	0.20*	0.07	0.34	19	0.44*	0.34	0.55	6	0.47*	0.37	0.58	30	0.41*	0.32	0.50
Self-esteem/esteem	2	0.78*	0.37	1.18	3	0.56*	0.47	0.66	‒	‒	‒	‒	5	0.63*	0.49	0.77
Social benefits	34	0.54*	0.44	0.63	42	0.34*	0.27	0.42	16	0.58*	0.49	0.67	62	0.47*	0.41	0.53
Enjoyment/play	76	0.47*	0.40	0.54	116	0.51*	0.47	0.56	32	0.59*	0.52	0.65	188	0.52*	0.48	0.56
Aesthetics	28	0.52*	0.44	0.60	29	0.52*	0.46	0.59	9	0.48*	0.32	0.64	58	0.45*	0.39	0.51
Escapism/spirituality	6	0.64*	0.51	0.77	6	0.50*	0.27	0.73	2	0.12	−0.25	0.50	11	0.43*	0.22	0.64
Personalization	12	0.69*	0.57	0.80	10	0.55*	0.46	0.64	3	0.73*	0.64	0.83	11	0.43*	0.32	0.53
Control	8	0.43*	0.28	0.58	4	0.55*	0.38	0.72	‒	‒	‒	‒	11	0.31*	0.12	0.51
Novelty	14	0.33*	0.22	0.44	17	0.50*	0.40	0.61	7	0.57*	0.42	0.72	27	0.43*	0.35	0.51
Relational benefits	81	0.53*	0.48	0.59	63	0.49*	0.40	0.58	12	0.47*	0.32	0.62	93	0.44*	0.38	0.50
Ecological benefits/ethics	5	0.44*	0.22	0.66	2	0.66*	0.63	0.68	‒	‒	‒	‒	11	0.60*	0.52	0.69
Societal benefits/ethics	3	0.33*	0.15	0.51	2	0.34*	0.09	0.59	‒	‒	‒	‒	2	0.30*	0.28	0.33
Sacrifices
Price	78	−0.34*	−0.43	−0.25	50	−0.11	−0.21	0.00	13	−0.37*	−0.58	−0.16	96	−0.29*	−0.36	−0.23
Time	6	−0.25	−0.57	0.07	9	−0.10*	−0.35	0.15	3	−0.13*	−0.17	−0.09	8	−0.04	−0.37	0.30
Effort	32	−0.21*	−0.33	−0.09	22	−0.18*	−0.33	−0.03	4	−0.32*	−0.54	−0.10	49	−0.12*	−0.23	−0.01
Privacy risk	10	−0.27*	−0.40	−0.14	‒	‒	‒	‒	‒	‒	‒	‒	8	−0.04	−0.20	0.13
Security risk	8	−0.42*	−0.66	−0.19	4	−0.12	−0.53	0.28	1	−0.39	‒	‒	9	−0.32*	−0.60	−0.04
Performance risk	11	−0.40*	−0.51	−0.29	1	−0.83	‒	‒	‒	‒	‒	‒	9	−0.32*	−0.48	−0.15
Financial risk	8	−0.29*	−0.50	−0.07	3	−0.53*	−0.74	−0.33	3	−0.24	−0.83	0.35	7	−0.15	−0.42	0.11
Physical risk	‒	‒	‒	‒	‒	‒	‒	‒	‒	‒	‒	‒	1	−0.01	‒	‒
Ecological costs	‒	‒	‒	‒	‒	‒	‒	‒	‒	‒	‒	‒	1	−0.41	‒	‒
Societal costs	‒	‒	‒	‒	‒	‒	‒	‒	1	−0.23	‒	‒	3	−0.05	−0.28	0.19

Overall value	‒	‒	‒	‒	195	0.61*	0.57	0.65	33	0.49*	0.37	0.62	282	0.53*	0.50	0.57
Satisfaction	‒	‒	‒	‒	‒	‒	‒	‒	49	0.57*	0.47	0.68	233	0.64*	0.61	0.67

k = number of effect sizes; rwc = sample-weighted reliability-adjusted average correlation; CI = 95% confidence interval. *p < .05. Detailed results in Web Appendix D. Confidence intervals were calculated when at least two effect sizes were available.

As expected, most benefits are positively associated with overall value, as well as satisfaction, WOM, and repurchase intentions (Table 7). The only exception is the effect of escapism/spirituality on WOM, which is not significant. The strength of these effect sizes varies, indicating that some benefits are more influential than others. With regard to overall value, the effect sizes of various benefits range between 0.20 and 0.78; self-esteem/esteem emerges as the most powerful determinant of overall value (rwc = 0.78), whereas status is the weakest (rwc = 0.20). The effect sizes between benefits and satisfaction as an outcome range between 0.34 and 0.66, such that ecological benefits/ethics have the strongest effect sizes (rwc = 0.66), whereas social benefits and societal benefits/ethics exhibit the weakest (rwc = 0.34). In the link between benefits and WOM, effect sizes range from 0.12 to 0.73, revealing that personalization has the strongest effect on WOM (rwc = 0.73), and escapism/spirituality has the weakest effect (rwc = 0.12), which is also nonsignificant. Finally, the effect sizes for repurchase intentions range between 0.30 and 0.63. In this case, self-esteem/esteem has the strongest effect size for repurchase intentions (rwc = 0.63), followed by ecological benefits/ethics (rwc = 0.60), and societal benefits/ethics have the weakest effects (rwc = 0.30). In turn, these varying effect sizes of different benefits produce a range of explained variances (ranging from 1 percent to 61 percent; Web Appendix D).

Turning to sacrifices, we uncover mixed findings. All sacrifices relate negatively to overall value, satisfaction, WOM, and repurchase intentions, but some of the effects are nonsignificant. Hunter and Schmidt (2004) suggest focusing on the strength and direction of effect sizes, rather than their significance level; we report all the results for comprehensiveness though. We calculated confidence intervals if at least two effect sizes were available. The effect sizes of sacrifices generally are weaker than those of benefits, though again, they vary in strength, from −0.42 to −0.21, such that some sacrifices are more influential than others. In detail, security risk appears to be the most powerful determinant of overall value (rwc = −0.42), but effort emerges as the weakest (rwc = −0.21), and time is nonsignificant. Turning to satisfaction outcomes, the effects sizes for sacrifices range between −0.83 and −0.10, dominated by performance risk (rwc = −0.83), whereas time has the weakest effect size (rwc = −0.10). In this case, the effects of price and security risk are nonsignificant. When we consider the effects of sacrifices on WOM, we find a range between −0.39 and −0.13, from security risk with the strongest effect size (rwc = −0.39) to time with the weakest (rwc = −0.13). The effect of financial risk on WOM is nonsignificant. The range of effect sizes of sacrifices on repurchase intentions is between −0.41 and −0.01. Ecological costs display the strongest effects (rwc = −0.41), physical risk has the weakest effect (rwc = −0.01), and time, privacy risk, financial risk, and societal costs are all nonsignificant. The explained variances also vary for the different sacrifices (ranging from 0 percent to 69 percent; Web Appendix D).

In addition, overall value exhibits strong effect sizes for all three outcomes: satisfaction (rwc = 0.61), WOM (rwc = 0.49), and repurchase intentions (rwc = 0.53). Along with the strong direct effects of the different benefits and sacrifices on overall value, this finding indicates partial mediating effects of overall value. We employ SEM to test these effects.

Different CPV Conceptualizations

We conducted formal tests to compare the four value measurement approaches, using SEM (Figure 2), on the basis of the number of significant relationships, explained variances (given that some models are saturated), and the insights provided. The correlation matrix in Web Appendix C served as the input, and we provide the results of the tests of Models 1–4 in Table 8. They reveal that the unidimensional model (Model 1) performed well, but the multidimensional model focused on benefits (Model 2) attained higher explained variances. Various sacrifices also appear significant, indicating the need to differentiate among them. Accordingly, the multidimensional model incorporating both benefits and sacrifices (Model 3) explains even more variance. The best performing Model 4 encompasses benefits, sacrifices, and overall value; we discuss this model in detail.

Table 8.

Results of Structural Equation Modeling.

Relationship	Model 1	Model 2	Model 3	Model 4
Relationship	Unidimensional	Multidimensional with Focus on Benefits	Multidimensional with Benefits and Sacrifices	Unidimensional and Multidimensional
Excellence → overall value	‒	‒	‒	0.20*
Social benefits → overall value	‒	‒	‒	0.20*
Enjoyment/play → overall value	‒	‒	‒	0.08*
Aesthetics → overall value	‒	‒	‒	0.14*
Relational benefits → overall value	‒	‒	‒	0.12*
Price → overall value	‒	‒	‒	−0.09*
Effort → overall value	‒	‒	‒	−0.07*
Risk → overall value	‒	‒	‒	−0.08*
				R² = 46.3%
Excellence → satisfaction	‒	0.24*	0.20*	0.14*
Social benefits → satisfaction	‒	‒	‒	‒
Enjoyment/play → satisfaction	‒	0.21*	0.19*	0.16*
Aesthetics → satisfaction	‒	0.19*	0.23*	0.16*
Relational benefits → satisfaction	‒	0.17*	0.13*	0.07*
Price → satisfaction	‒	‒	‒	‒
Effort → satisfaction	‒	‒	−0.02*	‒
Risk → satisfaction	‒	‒	−0.12*	−0.08*
Overall value → satisfaction	.61*	‒	‒	0.32*
	R² = 37.2%	R² = 42.2%	R² = 43.3%	R² = 48.9%
Excellence→ word-of-mouth	‒	‒	‒	‒
Social benefits→ word-of-mouth	‒	‒	‒	‒
Enjoyment/play → word-of-mouth	‒	0.34*	0.33*	0.33*
Aesthetics → word-of-mouth	‒	0.12*	0.04*	0.04*
Relational benefits → word-of-mouth	‒	.11*	‒	‒
Price → word-of-mouth	‒	‒	−0.21*	−0.21*
Effort → word-of-mouth	‒	‒	−0.12*	−0.12*
Risk → word-of-mouth	‒	‒	−0.04*	−0.04*
Overall value → word-of-mouth	0.23*	‒	‒	‒
Satisfaction → word-of-mouth	0.43*	0.28*	0.32*	0.32*
	R² = 35.7%	R² = 46.8%	R² = 53.5%	R² = 53.6%
Excellence → repurchase intentions	‒	0.05*	‒	‒
Social benefits → repurchase intentions	‒	0.21*	0.20*	0.20*
Enjoyment/play → repurchase intentions	‒	0.14*	0.13*	0.13*
Aesthetics → repurchase intentions	‒	‒	‒	‒
Relational benefits → repurchase intentions	‒	‒	‒	‒
Price → repurchase intentions	‒	‒	−0.17*	−0.17*
Effort → repurchase intentions	‒	‒	‒	‒
Risk → repurchase intentions	‒	‒	‒	‒
Overall value → repurchase intentions	0.22*	‒	‒	‒
Satisfaction → repurchase intentions	0.50*	0.47*	0.49*	0.49*
	R² = 44.1%	R² = 50.6%	R² = 54.6%	R² = 54.9%

*p < .05.

Model 4 exhibits the highest explained variances: 46.3% for overall value, 48.9% for satisfaction, 53.6% for WOM, and 54.9% for repurchase intentions. All benefits and sacrifices display positive relationships with overall value: excellence (µ = 0.20), social benefits (µ = 0.20), enjoyment/play (µ = 0.08), aesthetics (µ = 0.14), relational benefits (µ = 0.12), price (µ = −0.09), effort (µ = −0.07), and risk (µ = −0.08). Furthermore, we find effects of overall value, benefits, and sacrifices on satisfaction. The effect of overall value (β = 0.32) is significant, as are the effects of excellence (µ = 0.14), enjoyment/play (µ = 0.16), aesthetics (µ = 0.16), relational benefits (µ = 0.07), and risk (µ = −0.08). Moreover, various benefits and sacrifices relate to WOM: enjoyment/play (µ = 0.33), aesthetics (µ = 0.04), price (µ = −0.21), effort (µ = −0.12), and risk (µ = −0.04). Satisfaction also indicates significance (β = 0.32). Several benefits and sacrifices are associated with repurchase intentions, including social benefits (µ = 0.20), enjoyment/play (µ = 0.13), and price (µ = −0.17). Satisfaction (β = 0.49) is significant too.

In summary, CPV is best conceptualized as a combination of benefits, sacrifices, and overall value. All studied benefits and sacrifices are significantly related to overall value. In addition, many of them, along with overall value, significantly influence satisfaction, WOM, and repurchase intentions.

Moderator Analysis Results

The tests of homogeneity indicate significant variance in the effect sizes, confirming the need for a moderator analysis. We conducted moderator tests for each relationship with at least 12 effect sizes. In so doing, we identified several moderating effects for each relational moderator, as we detail in Table 9, along with the regression weights.

Table 9.

Results of Moderator Tests Using Meta-Regression.

Relationship	k	Relational Moderators						Controls				R²	VIF
Relationship	k	Nonprofit (vs. For-Profit)	Public (vs. Private)	Contractual (vs. Non-Contractual)	Online (vs. Offline)	B2B (vs. B2C)	Goods (vs. Services)	Correlation (vs. Regression)	Student (vs. Non-Student)	Journal Quality	Year	R²	VIF
Benefits → overall value
Convenience/efficiency → overall value	73	−0.25*	0.06*	−0.38*	0.20*	0.12*	0.03	0.27*	0.09*	0.17*	0.24*	37%	1.57
Excellence → overall value	221	−0.04*	0.06*	0.12*	−0.10*	0.10*	−0.08*	−0.07*	0.03*	−0.14*	0.24*	11%	1.48
Status → overall value	27	0.20*	−0.37*	0.29*	−0.15*	‒	−0.44*	−0.03	−0.06*	0.10*	−0.25*	60%	1.90
Social benefits → overall value	34	‒	0.31*	−0.13*	0.04	‒	0.00	−0.11*	−0.59*	0.07*	0.31*	54%	2.66
Enjoyment/play → overall value	76	0.03*	−0.22*	0.03	−0.17*	‒	−0.44*	0.12*	−0.03	0.31*	0.22*	35%	1.93
Aesthetics → overall value	28	‒	−0.20	‒	−0.11	‒	−0.29*	0.06	−0.14*	0.17*	0.38*	33%	6.76
Personalization → overall value	12	‒	−0.14	−0.64*	0.16	‒	0.71*	0.41*	−0.40*	−0.70*	−0.63*	82%	3.24
Novelty → overall value	14	0.10	−0.50*	0.38*	−0.15	‒	−0.42*	0.81*	−0.32*	−0.18	−0.26	64%	4.93
Relational benefits → overall value	81	−0.15*	0.08*	−0.04	0.06*	0.09*	−0.06*	0.08*	0.04	0.02	−0.04	4%	1.53
Sacrifices → overall value
Price → overall value^a	78	−0.03*	0.09*	0.27*	−0.32*	−0.10*	0.03	−0.05*	0.02	0.30	0.25	21%	1.83
Effort → overall value^a	32	0.58*	−0.78*	−0.28*	0.07*	−0.16*	−0.09*	−0.11*	−0.28*	0.02	−0.31*	60%	3.73
Risk → overall value^a	55	0.06	0.08*	0.04	−0.12*	‒	−0.07*	0.13*	0.01	0.09*	−0.27*	13%	2.22
Overall value → outcomes
Overall value → satisfaction	195	0.02	−0.06*	0.04*	−0.20*	0.14*	−0.07*	0.33*	0.08*	0.06*	0.15*	20%	1.51
Overall value → word-of-mouth	33	0.17*	−0.10*	−0.46*	0.17*	0.12*	0.09*	0.31*	−0.15*	−0.12*	−0.03	33%	1.90
Overall value → repurchase intentions	282	0.11*	−0.21*	−0.05*	−0.08*	−0.02	−0.05*	0.44*	0.01	−0.12*	0.02	31%	1.57
		H1–H2	H3–H4	H5–H6	H7–H8	H9–H10	H11–H12

*p < .05. The table show standardized coefficients. A dash indicates that a moderator could not be tested because of data availability. A positive regression coefficient suggests benefits and overall value (sacrifices) have a stronger positive (negative) effect in this context, while a negative coefficient suggest a weaker effect.

^aInterpreting the moderator results requires consideration of the direction of main effects; to ease interpretation, we reverse-coded the effect sizes of sacrifices.

First, in line with H₁, the positive effect of status (β = 0.20) on overall value is stronger in nonprofit contexts, whereas the effects of convenience/efficiency (β = −0.25) and excellence (β = −0.04) are stronger in for-profit contexts. Consistent with H₂, the effect of overall value on repurchase intentions (β = 0.11) and WOM (β = 0.17) are stronger in nonprofit contexts; the moderating effect on satisfaction is not significant. Although not hypothesized, we observe a stronger positive effect of enjoyment/play (β = 0.03) on overall value in nonprofit contexts; the positive effect of relational benefits (β = −0.15) also is stronger in for-profit contexts. Whereas the negative effect of effort (β = 0.58) on overall value is stronger in nonprofit contexts, the negative effect of price (β = −0.03) gains strength in for-profit contexts.

Second, the positive effect of social benefits (β = 0.31) on overall value is stronger in public contexts, and the positive effects of enjoyment/play (β = −0.22) and novelty (β = −0.50) are stronger in private contexts. The negative effect of effort (β = −0.78) is stronger in private contexts too. These findings support all the predictions in H₃. Consistent with our expectations in H₄, the positive effects of overall value on satisfaction (β = −0.06), WOM (β = −0.10), and repurchase intentions (β = −0.21) are stronger in private contexts. We also find that the positive effects of convenience/efficiency (β = 0.06), excellence (β = 0.06), and relational benefits (β = .08) on overall value are stronger in public contexts, whereas the positive effect of status (β = −0.37) is stronger in private contexts. Moreover, the negative effects of price (β = 0.09) and risk (β = 0.08) on overall value are stronger in public contexts.

Third, we find mixed results related to contractual versus non-contractual contexts. As we expected, the positive effect of status (β = 0.29) on overall value is stronger in contractual contexts, and the positive effect of convenience/efficiency (β = −0.38) is stronger in non-contractual contexts. The negative effect of effort (β = −0.28) on overall value also is stronger in non-contractual contexts. But the positive effect of personalization (β = −0.64) on overall value is stronger in non-contractual contexts, and we find no significant effect for relational benefits. Thus, H₅ receives partial support. The positive effect of overall value on repurchase intentions (β = −0.05) and WOM (β = −0.46) are stronger in non-contractual settings, but its effects on satisfaction (β = 0.04) are stronger in contractual settings, which also indicates partial support for H₆. The positive effects of excellence (β = 0.12) and novelty (β = .38) on overall value are stronger in contractual contexts, whereas social benefits (β = −0.13) have a stronger positive effect in non-contractual contexts. Finally, the negative effect of price (β = .27) is stronger in contractual contexts.

Fourth, the positive effect of convenience/efficiency (β = 0.20) on overall value is stronger in online contexts. However, we do not find significant effects for either personalization or social benefits. Contrary to our expectations, the negative effect of risk (β = −0.12) on overall value is stronger in offline contexts. Thus, we obtain only partial support for H₇. The positive effect of overall value on WOM (β = 0.17) is stronger in online contexts, whereas its positive effects on satisfaction (β = −0.20) and repurchase intentions (β = −0.08) are stronger in offline contexts. These results provide partial support for H₈. Moreover, the positive effect of relational benefits (β = 0.06) on overall value is stronger in online contexts, but the positive effects of excellence (β = −0.10), status (β = −0.15), and enjoyment/play (β = −0.17) are stronger in offline contexts. Online contexts produce a stronger negative effect of effort (β = 0.07) on overall value; offline contexts indicate a stronger negative effect of price (β = −0.32).

Fifth, in line with H₉, B2B contexts reveal stronger positive effects of convenience/efficiency (β = 0.12), excellence (β = 0.10), and relational benefits (β = 0.09) on overall value. We also find that B2B contexts feature stronger positive effects of overall value on satisfaction (β = 0.14) and WOM (β = 0.12), though not on repurchase intentions, in partial support of H₁₀. Although not hypothesized, we note that B2C contexts indicate stronger negative effects of price (β = −0.10) and effort (β = −0.16) on overall value.

Sixth and finally, services contexts show stronger positive effects of status (β = −0.44) and relational benefits (β = −0.06) on overall value, as well as a stronger negative effect of effort (β = −0.09), in support of H₁₁. Services indicate stronger positive effects of overall value on satisfaction (β = −0.07) and repurchase intentions (β = −0.05), but in goods contexts, we find a stronger positive effect of overall value on WOM (β = 0.09), indicating partial support for H₁₂. In addition, goods contexts reveal a stronger positive effect of personalization (β = 0.71) on overall value, whereas services lead to stronger positive effects of excellence (β = −0.08), enjoyment/play (β = −0.44), aesthetics (β = −0.29), and novelty (β = −0.42). Services also show a stronger negative effect of risk (β = −0.07) on overall value.

The variance inflation factors of all these regression models are acceptable, ranging from 1.48 to 6.76. The results are also robust when controlling for the influence of methodological moderators. Journal quality enhances some effects (e.g., convenience/efficiency) but weakens others (e.g., excellence and personalization). Prior meta-analysis approaches reflect the notion that the rigid control mechanisms imposed by quality journals prevent effect size inflation; we find these mechanisms support empirical studies that are likely to detect some effects. In addition, many benefits gain importance over time (e.g., social benefits), but others lose importance (e.g., status and personalization). In contrast with predictions that customer expectations always increase over time, we find that some benefits lose importance.

General Discussion

Theoretical Contributions

Empirical research on CPV has generated contradictory empirical findings. Responding to calls to “synthesize extant research findings by engaging in meta-analytical research endeavors” (Zeithaml et al. 2020, p. 18), we conduct a meta-analysis designed to summarize this empirical evidence and test the theoretical bases of the CPV construct. In turn, this article contributes to extant literature in three main ways.

First, we identify which conceptualization of CPV offers the best explanation of customers’ positive responses (Table 8). Our findings indicate that the model that integrates benefits, sacrifices, and overall value surpasses the others. It also matches Zeithaml’s (1988) foundational definition and represents the most comprehensive conceptualization of CPV. Because it accounts for advantages and disadvantages that shape value judgments, and also captures this process at a more abstract level (i.e., overall value), this model enhances theoretical understanding of the intricacies of CPV and therefore should be given preference in continued research on CPV.

Second, by leveraging Leroi-Werelds’s (2019) typology, we provide a more complete view of what creates (i.e., benefits) or hinders (i.e., sacrifices) overall value perceptions. We also provide meta-analytical evidence of the effects of overall value on satisfaction, WOM, and repurchase intentions. Precisely, self-esteem is the most important driver of overall value, and security risk is the variable that impedes it most (Table 7). These primary roles of status-related variables highlight the need to move beyond simple conceptualizations of value as a quality/price ratio (Dodds, Monroe, and Grewal 1991; Grewal, Monroe, and Krishnan 1998). We also note that status, as another social variable, is the benefit with the least importance, which may seem surprising at first glance. But this insight suggests that benefits that strengthen customers’ evaluations of their own worth (e.g., self-esteem), as opposed to external evaluations (e.g., status), may have the most influence on overall value perceptions. The finding that ecological benefits/ethics exert particularly strong direct effects on satisfaction and repurchase intentions also deserves more attention.

Third, our meta-analysis of CPV offers a more precise view, by identifying the relational contexts in which overall value is likely either to be affected or else to influence satisfaction, WOM, and repurchase intentions. These moderator tests contribute to a better understanding of the generalizability of different benefits/sacrifices across contexts (Grewal, Puccinelli, and Monroe 2018). As summarized in Table 10 and Figure 3, as well as detailed in the following paragraphs, we find support for various moderating effects that we predicted, as well as some unexpected effects that warrant more qualitative research to explain them.

Table 10.

Summary of Hypothesis Testing and Explored Moderating Effects.

Hypotheses	Results	Conclusions
Nonprofit vs. for-profit
Status, convenience/efficiency, and excellence → overall value	H1 (✓)	Status has a stronger positive effect on overall value in nonprofit contexts; convenience/efficiency and excellence have stronger positive effects in for-profit contexts
Overall value → customer outcomes	H2 (✓; WOM and RI)	Overall value has stronger positive effects on WOM and repurchase intentions in nonprofit contexts. The effect on satisfaction is nonsignificant and appears to be conditional
Explored moderators	—	Enjoyment/play has a stronger positive effect on overall value in nonprofit contexts; relational benefits have a stronger positive effect in for-profit contexts. Effort has a stronger negative effect in nonprofit contexts; price has a stronger negative effect in for-profit contexts
Public vs. private
Social benefits, enjoyment/play, novelty, and effort → overall value	H3 (✓)	Social benefits have a stronger positive effect on overall value in public contexts; enjoyment/play and novelty have stronger positive effects in private contexts. Effort has a stronger negative relationship in private contexts
Overall value → customer outcomes	H4 (✓)	Overall value has stronger positive effects on satisfaction, WOM, and repurchase intentions in private contexts
Explored moderators	—	Convenience/efficiency, excellence, and relational benefits display stronger positive effects on overall value in public contexts; status displays stronger positive effects in private contexts. Price and risk display stronger negative effects on overall value in public contexts
Contractual vs. non-contractual
Convenience/efficiency, personalization, relational benefits, status, and effort → overall value	H5 (✓; convenience, status, and effort)	Status has a stronger positive effect on overall value in contractual contexts; convenience/efficiency has a stronger positive effect in non-contractual contexts. Effort has a stronger negative effect in non-contractual contexts. The effect of relational benefits is nonsignificant, and personalization has a stronger positive effect in non-contractual contexts
Overall value → customer outcomes	H6 (✓; WOM, and RI)	Overall value has stronger positive effects on WOM and repurchase intentions in non-contractual contexts. The positive effect on satisfaction is stronger in contractual contexts; scholars should differentiate among these three outcomes
Explored moderators	—	Excellence and novelty display stronger positive effects on overall value in contractual contexts; social benefits display stronger positive effects in non-contractual contexts. Price displays stronger negative effects on overall value in contractual contexts
Online vs. offline
Convenience, personalization, and social benefits, and risk → overall value	H7 (✓; convenience)	Convenience/efficiency has a stronger positive effect on overall value in online contexts; risk has a stronger negative effect in offline contexts. Personalization and social benefits are nonsignificant, and these effects appear to be conditional
Overall value → customer outcomes	H8 (✓; WOM)	Overall value has a stronger positive effect on WOM in online contexts; the positive effects on satisfaction and repurchase intentions are stronger offline
Explored moderators	—	Relational benefits display stronger positive effects on overall value in online contexts; excellence, status, and enjoyment/play display stronger positive effects in offline contexts. Effort displays stronger negative effects on overall value in online contexts; price displays stronger negative effects in offline contexts
B2B vs. B2C
Convenience/efficiency, excellence, and relational benefits → overall value	H9 (✓)	Convenience/efficiency, excellence, and relational benefits have stronger positive effects on overall value in B2B contexts
Overall value → customer outcomes	H10 (✓; satisfaction, WOM)	Overall value has stronger positive effects on satisfaction and WOM in B2B contexts. The effect on repurchase intentions is nonsignificant and appears to be conditional
Explored moderators	—	Price and effort display stronger negative effects on overall value in B2C contexts
Goods vs. services
Relational benefits, status, and effort → overall value	H11 (✓)	Relational benefits and status have stronger positive effects on overall value for services. Also, effort has a stronger negative effect for services
Overall value → customer outcomes	H12 (✓; WOM)	Overall value has a stronger positive effect on WOM for goods; it has stronger positive effects on satisfaction and repurchase intentions for services
Explored moderators	—	Personalization displays stronger positive effects on overall value for goods; excellence, enjoyment/play, aesthetics, and novelty display stronger positive effects for services. Risk displays stronger negative effects for services

Notes. WOM = word-of-mouth; RI = repurchase intentions.

Figure 3.

Varying importance of benefits and sacrifices on overall value.

Across nonprofit and for-profit contexts, several revealed differences can advance CPV theory. The stronger positive effect of status on overall value for customers of nonprofit organizations suggests that customers in these contexts seek to enhance their social standing by supporting these organizations (Parker and Lehmann 2011). However, the positive effects of convenience/efficiency and excellence are stronger in relation to for-profit organizations, seemingly because customers expect them to provide better services. The effects of other benefits/sacrifices on overall value also differ across these contexts, in ways we did not predict; we call on scholars to use qualitative methods to identify some underlying reasons. The stronger positive influences of overall value on WOM and repurchase intentions in nonprofit contexts also suggest that customers assess nonprofit organizations more favorably (Vock, Van Dolen, and Kolk 2013).

Between public and private contexts, we find the most differences, indicating the importance of accounting for these settings when studying CPV. Social benefits trigger overall value in a public context, but enjoyment and novelty prompt more of overall value in private contexts. The greater impact of social benefits on overall value in public settings might be attributed to the value linked to exposure and social acknowledgment. Effort appears to be perceived less negatively in public contexts, with weaker effects (Blut, Wang, and Schoefer 2016). We also observe that overall value has stronger positive effects on all customer outcomes in private contexts. It is likely that customers in private contexts focus more on CPV, which in turn leads to greater satisfaction, WOM, and repurchase intentions.

The moderating effects of contractual or non-contractual contexts also are influential, requiring scholarly attention. Status has a stronger positive effect on customers in contractual contexts, whereas convenience has a stronger effect in non-contractual contexts. Personalization, unexpectedly, show a stronger effect in non-contractual contexts, presumably, because of the greater freedom and flexibility of such contexts. We do not find any differences for relational benefits; customers in both contexts appreciate such benefits. However, we find that effort has a stronger negative effect in non-contractual contexts, perhaps reflecting customers’ expectations of the amount of effort they must perform to identify the most convenient offer (Pick and Eisend 2014). As expected, overall value enhances WOM and repurchase intentions in non-contractual contexts, where customers can switch providers easily. Overall value also appears beneficial to satisfaction among customers in contractual settings, an unexpected finding that requires further exploration.

The positive effects of convenience/efficiency are stronger in online contexts, where customers might appreciate the convenience of the offer. Yet social benefits and personalization exhibit no significant differences across contexts. Scholars should work to identify specific customer segments that might appreciate these benefits. In addition, and unexpectedly, in online contexts, effort is more detrimental to overall value than risk is, implying that customers grow impatient with browsing different websites to find an interesting offer (Harris et al. 2006). Even more surprising, we uncover inconsistent results for outcomes: Satisfaction and repurchase intentions are more strongly influenced by overall value in offline settings, whereas WOM is strongly affected by overall value in online settings. This result might reflect web characteristics that enable users to share opinions easily. Scholars could examine whether similar differences exist for other customer outcomes.

In B2B versus B2C contexts, we find relatively fewer differences that arise for the other moderators. Specifically, some benefits are more important in B2B than in B2C settings; convenience/efficiency, excellence, and relational benefits all have stronger positive effects on overall value. In some unexpected results, price and effort have stronger negative impacts on overall value in B2C contexts, perhaps due to the financial restrictions and limited resources of consumers. Overall value also has a stronger positive impact on satisfaction and WOM in B2B settings, where clients assess their relationships with suppliers very carefully (Ulaga and Eggert 2006). The effect on repurchase intentions is nonsignificant though.

Comparing goods and services, we find that most benefits and sacrifices are of greater relevance for services than for goods, such as status and relational benefits. This important finding should guide scholars’ investigations of CPV. Other benefits also are of greater relevance in service contexts, though we did not predict such outcomes. A notable exception, though, is personalization, which is unexpectedly more important for goods than services. Personalizing a physical product might create an emotional connection with the product, thus enhancing perceptions of overall value. Effort has a stronger negative effect on overall value when people consume services, reflecting the more active role of customers in service contexts (Vargo and Lusch 2004). Among the outcomes of overall value, we find, remarkably, that WOM is the only outcome that shows a more potent effect among customers of goods. Further research might investigate such differences by outcome type.

Managerial Implications

Managers can leverage CPV as a performance measure, but to do so, they should consider not only the benefits and sacrifices proposed by Leroi-Werelds (2019) but also the implications of overall value. Depending on the contexts in which managers operate, customers will have different expectations, such that certain benefits and sacrifices will be more effective for increasing overall value (Figure 3).

First, in for-profit settings, customers expect offers that make their life easier (convenience), offer high quality (excellence), and lead to better relationships with the provider (relational benefits). They also focus strongly on the price of the offer. But in nonprofit contexts, customers expect to make positive impression on others (status) and look for fun and pleasure (enjoyment). They are less willing to invest effort to interact with the firm. In contrast with arguments that nonprofit organizations should mobilize the same marketing tools as for-profit organizations, our meta-analysis reveals the different benefits and sacrifices that affect overall value in these two distinct contexts. In turn, managers of nonprofit organizations should actively work to develop offerings that generate emotional reactions and social consideration.

Second, in private contexts, overall value depends on the status, enjoyment, and novelty of the offering, whereas in public contexts (e.g., tourism, hospitality, entertainment, and education), customers expect social and relational benefits. They also appreciate utilitarian benefits (convenience and excellence) more and are more sensitive to monetary (price) and nonmonetary (risk) sacrifices. Managers working in public contexts should try to improve customer perceptions of these benefits and sacrifices.

Third, options for increasing overall value also vary across contractual and non-contractual settings. In the former context, because customers bear substantial switching costs, they expect companies to provide innovative (novelty) and high-quality (excellence) offerings, as well as distinctive status. In the latter context, customers expect offers that are useful (convenience), tailored (personalization), and capable of improving relationships with other customers (social benefits). Otherwise, they likely will take their businesses elsewhere, considering the low switching costs in non-contractual contexts.

Fourth, customers perceive greater value when digital offers on online contexts make their lives easier (convenience) and provide relational benefits. They are sensitive to how much effort they must invest though. Therefore, managers in digital contexts should provide efficient web interfaces and support for their customers. In offline contexts, customers are more likely to appreciate benefits related to excellence, enjoyment, and status. Our results suggest that companies should particularly focus on these three benefit types to increase overall value. Customers are also more price- and risk-sensitive.

Fifth, the more effective ways to increase overall value in B2B settings revolve around convenience, excellence, and relational benefits. The technical performance of a high-quality offer, combined with relational strategies (frequency of interaction, information exchange, active role for the customer and supplier) all can improve overall value in B2B contexts. In B2C contexts, our meta-analysis highlights the key roles of monetary (price) and nonmonetary (effort) sacrifices. Companies often emphasize the benefits of their offers; we recommend that they also reassure consumers about the importance of and rationale for their monetary and nonmonetary sacrifices.

Sixth, in goods contexts, customers appreciate personalization. Companies can and should use various tools to personalize their products (e.g., customization programs and product recommendations based on past choices). Managers in service contexts have more options to increase overall value, in that they can emphasize a variety of benefits associated with their services (excellence, status, enjoyment, aesthetics, novelty, and relational benefits). They also should reassure customers about the nonmonetary sacrifices (effort and risk) they must make.

Limitations and Further Research

The limitations of our meta-analysis largely stem from data availability issues. First, meta-analyses can only synthesize variables that have received sufficient attention in prior research and for which effect sizes have been reported in primary studies. Although our approach allowed for the assessment of the four most widely used conceptualizations of CPV—from unidimensional (Dodds, Monroe, and Grewal 1991) to the most integrative multidimensional (Leroi-Werelds 2019) models—we cannot test the entire set of benefits and sacrifices under consideration in this research, because too few studies reported effect sizes for some of these variables. Therefore, this study did not assess the effects of self-esteem, control, and societal and ethical benefits on WOM, nor could we test the effects of physical risk and ecological costs as sacrifices. Further research on these dimensions is needed to appraise their relative effects on satisfaction, WOM, and repurchase intentions, compared with those of other sacrifices that shape overall value.

Second, the outcomes we study are mainly relevant in the post-purchase phase. We call for more studies that explicitly differentiate between pre- and post-purchase phases and examine how CPV influences decision-making in the pre-purchase phase. Probably due to the relatively recent emergence of customer journey concepts (Lemon and Verhoef 2016), quantitative studies of value in the past 30 years have not addressed this dynamic aspect in the service experience. With a meta-analysis, we thus cannot identify whether CPV is more influential before or after the service experience.

Third, we tested six moderators that influence how overall value develops and its effects. To deliver managerial implications that can easily be implemented, we define these moderating variables to reflect macro indicators of the setting in which CPV has been examined. Our results reveal that among the three outcomes of overall value, WOM is affected by all six moderators, and satisfaction and loyalty each is affected by five, confirming the strong influence of the moderating variables. Due to the aggregated nature of the data though, differences between moderating effects within a given context might not be easily identifiable in a meta-analysis. For example, interactions with nonprofit organizations can span distinct situations, such as giving to a charity but also purchasing on a web platform of the circular economy. A more granular investigation might provide a deeper understanding of the specific conditions in which the effect of overall value is moderated. Furthermore, meta-analyses rarely include individual-level moderators like personality traits, even though such variables—while creating additional challenges for managers—may explain why a specific benefit or sacrifice varies or affects outcomes differently (Holbrook 1999). We call for more primary studies to test individual-level moderators of these relationships.

Supplemental Material

Supplemental Material - Customer Perceived Value: A Comprehensive Meta-analysis

Supplemental Material for Customer Perceived Value: A Comprehensive Meta-analysis by Markus Blut, Damien Chaney, Renaud Lunardo, Rémi Mencarelli, and Dhruv Grewal in Journal of Service Research

Supplemental Material

Supplemental Material - Customer Perceived Value: A Comprehensive Meta-analysis

Supplemental Material for Customer Perceived Value: A Comprehensive Meta-analysis by Markus Blut, Damien Chaney, Renaud Lunardo, Rémi Mencarelli, and Dhruv Grewal in Journal of Service Research

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Markus Blut

Supplemental Material

Supplemental material for this article is available online.

Note

Author Biographies

Markus Blut is professor of marketing at Durham University, UK. His primary areas of research interest are service management, retailing, and new technologies. He currently serves on the editorial review boards of Journal of Service Research (AE), Journal of Retailing, and Journal of Business Research.

Damien Chaney is professor of marketing at EM Normandie Business School, France. His research interests are consumer experience, market system dynamics, and consumer-brand relationship. He serves on the editorial review boards of European Journal of Marketing and Recherche et Applications en Marketing (AE).

Renaud Lunardo is Professor of Marketing at Kedge Business School. His expertise is in retailing. His research has received numerous awards and has been published in international journals including (among others) Journal of Retailing, Journal of the Association for Consumer Research, or International Journal of Research in Marketing.

Rémi Mencarelli is Professor at Savoie Mont Blanc University. His research focuses on perceived value and consumption experience in various contexts. He has published his research in international journals such as Journal of Public Policy & Marketing, Marketing Letters, or Journal of Business Research.

Dhruv Grewal (PhD Virginia Tech) is the Toyota Chair in Commerce and Electronic Business and a Professor of Marketing at Babson College. He currently serves on numerous editorial review boards, including JSR, JM (AE), JPP&M (AE), JAMS (AE), and he was co-editor of JR (2001–2007).

References

Arnett

German

Hunt

(2003), “The Identity Salience Model of Relationship Marketing Success: The Case of Nonprofit Marketing,” Journal of Marketing, 67 (2), 89-105.

Association of Business Schools . (2015), Academic Journal Quality Guide. Association of Business Schools.

Babin

Darden

Griffin

(1994), “Work and/or Fun: Measuring Hedonic and Utilitarian Shopping Value,” Journal of Consumer Research, 20 (4), 644-656.

Blut

Wang

Schoefer

(2016), “Factors Influencing the Acceptance of Self-service Technologies: A Meta-Analysis,” Journal of Service Research, 19 (4), 396-416.

Boksberger

Melsen

(2011), “Perceived Value,” Journal of Services Marketing, 25 (3), 229-240.

Bolton

Drew

(1991), “A Multistage Model of Customers' Assessments of Service Quality and Value,” Journal of Consumer Research, 17 (4), 375-384.

Brady

Robertson

C. J.

(1999), “An Exploratory Study of Service Value in the USA and Ecuador,” International Journal of Service Industry Management, 10 (5), 469-486.

Dodds

Monroe

Grewal

(1991), “Effects of Price, Brand, and Store Information on Buyers’ Product Evaluations,” Journal of Marketing Research, 28 (3), 307-319.

Eggert

Ulaga

Frow

Payne

(2018), “Conceptualizing and Communicating Value in Business Markets.” Industrial Marketing Management, 69 (2), 80-90.

10.

Gale

(1994), Managing Customer Value. New-York: The Free Press.

11.

Gallarza

M. G.

Arteaga

Del Chiappa

Gil-Saura

Holbrook

M. B.

(2017), “A Multidimensional Service-value Scale Based on Holbrook’s Typology of Customer Value,” Journal of Service Management, 28 (4), 724-762.

12.

Gallarza

Gil-Saura

Holbrook

(2011), “The Value of Value: Further Excursions on the Meaning and Role of Customer Value,” Journal of Consumer Behaviour, 10 (4), 179-191.

13.

Grewal

Monroe

Krishnan

(1998), “The Effects of Price-Comparison Advertising on Buyers’ Perceptions of Acquisition Value, Transaction Value, and Behavioral Intentions,” Journal of Marketing, 62 (2), 46-59.

14.

Grewal

Puccinelli

Monroe

(2018), “Meta-analysis: Integrating Accumulated Knowledge.” Journal of the Academy of Marketing Science, 46 (1), 9-30.

15.

Grove

Fisk

(1997), “The Impact of Other Customers on Service Experiences: A Critical Incident Examination of “Getting Along”,” Journal of Retailing, 73 (1), 63-85.

16.

Grönroos

Voima

(2013), “Critical Service Logic: Making Sense of Value Creation and Co-creation.” Journal of the Academy of Marketing Science, 41 (2), 133-150.

17.

Harris

Grewal

Mohr

Bernhardt

(2006), “Consumer Responses to Service Recovery Strategies,” Journal of Business Research, 59 (4), 425-431.

18.

Holbrook

(1999), Consumer Value: A Framework for Analysis and Research. London and New York: Routledge.

19.

Holbrook

(2005), “The Role of Ordinary Evaluations in the Market for Popular Culture: Do Consumers Have “Good Taste”?” Marketing Letters, 16 (2), 75-86.

20.

Hull

Lio

(2006), “Innovation in Non-profit and For-Profit Organizations: Visionary, Strategic, and Financial Considerations,” Journal of Change Management, 6 (1), 53-65

21.

Hunter

Schmidt

(2004), Methods of Meta-analysis: Correcting Error and Bias in Research Findings. Newbury Park, CA: Sage.

22.

Hutchinson

Lai

Wang

(2009), “Understanding the Relationships of Quality, Value, Equity, Satisfaction, and Behavioral Intentions Among Golf Travelers,” Tourism Management, 30 (2), 298-308.

23.

Jones

M.A

Mothersbaugh

D. L.

Beatty

S. E.

(2003), “The Effects of Locational Convenience on Customer Repurchase Intentions across Service Types,” Journal of Services Marketing, 17 (7), 701-712.

24.

Keiningham

T. L.

Rust

R. T.

Lariviere

Aksoy

Williams

(2018), “A Roadmap for Driving Customer Word-of-mouth,” Journal of Service Management, 29 (1), 2-38.

25.

Kim

Ham

Moon

Chua

B. L.

Han

(2019), “Experience, Brand Prestige, Perceived Value (Functional, Hedonic, Social, and Financial), and Loyalty Among GROCERANT Customers,” International Journal of Hospitality Management, 77 (January), 169-177.

26.

Kleijnen

De Ruyter

Wetzels

(2007), “An Assessment of Value Creation in Mobile Service Delivery and the Moderating Role of Time Consciousness,” Journal of Retailing, 83 (1), 33-46.

27.

Kulviwat

Bruner

II Al-Shuridah

(2009), “The Role of Social Influence on Adoption of High-tech Innovations,” Journal of Business Research, 62 (7), 706-712.

28.

Kumar

Reinartz

(2016), “Creating Enduring Customer Value,” Journal of Marketing, 80 (6), 36-68.

29.

Lam

Shankar

Erramilli

M. K.

Murthy

(2004). Customer Value, Satisfaction, Loyalty, and Switching Costs,” Journal of the Academy of Marketing Science, 32, 293-311.

30.

LaPlaca

P. J.

Katrichis

J. M.

(2009), “Relative Presence of Business-To-Business Research in the Marketing Literature,” Journal of Business-to-Business Marketing, 16 (1-2), 1-22.

31.

Laroche

Yang

McDougall

Bergeron

(2005), “Internet versus Bricks-And-Mortar Retailers,” Journal of Retailing, 81 (4), 251-267.

32.

Lei

de Ruyter

Wetzels

(2008), “Consumer Responses to Vertical Service Line Extensions,” Journal of Retailing, 84 (3), 268-280.

33.

Lemon

Verhoef

(2016), “Understanding Customer Experience throughout the Customer Journey,” Journal of Marketing, 80 (6), 69-96.

34.

Lemon

White

Winer

(2002), “Dynamic Customer Relationship Management,” Journal of Marketing, 66 (1), 1-14.

35.

Leroi-Werelds

(2019), “An Update on Customer Value: State of the Art, Revised Typology, and Research Agenda. Journal of Service Management 30 (5), 650-680.

36.

Leroi-Werelds

Streukens

Brady

M. K.

Swinnen

(2014), “Assessing the Value of Commonly Used Methods for Measuring Customer Value.” Journal of the Academy of Marketing Science, 42 (4), 430-451.

37.

Lin

C. H.

Sher

P. J.

Shih

H. Y.

(2005), “Past Progress and Future Directions in Conceptualizing Customer Perceived Value,” International Journal of Service Industry Management, 16 (4), 318-336.

38.

Lindgreen

Hingley

Grant

Morgan

(2012), “Value in Business and Industrial Marketing,” Industrial Marketing Management, 41 (1), 207-214.

39.

Lipsey

M. W.

Wilson

D. B.

(2001), Practical Meta-Analysis. Thousand Oaks, CA: Sage.

40.

Marwell

McInerney

(2005), “The Nonprofit/for-profit Continuum: Theorizing the Dynamics of Mixed-form Markets,” Nonprofit and Voluntary Sector Quarterly, 34 (1), 7-28.

41.

Mathwick

Malhotra

Rigdon

(2001) “Experiential Value: Conceptualization, Measurement and Application in the Catalog and Internet Shopping Environment,” Journal of Retailing, 77 (1), 39-56.

42.

Parasuraman

Zeithaml

Berry

(1985), “A Conceptual Model of Service Quality and its Implications for Future Research,” Journal of Marketing, 49 (4), 41-50.

43.

Parker

J. R.

Lehmann

D. R.

(2011), “When Shelf-based Scarcity Impacts Consumer Preferences,” Journal of Retailing, 87 (2), 142-155.

44.

Peterson

R. A.

Brown

S. P.

(2005), “On the Use of Beta Coefficients in Meta-analysis,” Journal of Applied Psychology, 90 (1), 175–181.

45.

Petrick

(2002), “Development of a Multi-Dimensional Scale for Measuring the Perceived Value of a Service” Journal of Leisure Research, 34 (2), 119-134.

46.

Pick

Eisend

(2014), “Buyers’ Perceived Switching Costs and Switching.” Journal of the Academy of Marketing Science, 42 (2), 186-204.

47.

Sánchez-Fernández

Iniesta-Bonillo

M. Á.

Holbrook

M. B.

(2009), “The Conceptualisation and Measurement of Consumer Value in Services, International Journal of Market Research, 51 (1), 1-17.

48.

Sheth

Newman

Gross

(1991). Why We Buy what We Buy: A Theory of Consumption Values. Journal of Business Research, 22 (2), 159-170.

49.

Stieler

Weismann

Germelmann

(2014), “Co-destruction of Value by Spectators,” European Sport Management Quarterly, 14 (1), 72-86.

50.

Sweeney

Soutar

(2001) “Consumer Perceived Value: the Development of a Multiple Item Scale,” Journal of Retailing, 77 (2), 203-220.

51.

Ulaga

Eggert

(2006) “Value-based Differentiation in Business Relationships: Gaining and Sustaining Key Supplier Status,” Journal of Marketing, 70 (1), 119-136.

52.

Vargo

Lusch

(2004), “Evolving to a New Dominant Logic for Marketing,” Journal of Marketing, 68 (1), 1-17.

53.

Vieira

(2013), “Antecedents and Consequences of Perceived Value: A Meta-analytical Perspective.” Journal of Customer Behaviour, 12 (2-3), 111-133.

54.

Vock

Van Dolen

Kolk

(2013), “Changing Behaviour Through Business-nonprofit Collaboration?” European Journal of Marketing, 47 (9), 1476-1503.

55.

Williams

Soutar

(2009), “Value, Satisfaction and Behavioral Intentions in an Adventure Tourism Context,” Annals of Tourism Research, 36 (3), 413-438.

56.

Zeithaml

(1988), “Consumer Perceptions of Price, Quality, and Value: A Means-end Model and Synthesis of Evidence,” Journal of Marketing, 52 (3), 2-22.

57.

Zeithaml

Bitner

(2003), Services Marketing. Boston: McGraw-Hill/Irwin.

58.

Zeithaml

Verleye

Hatak

Koller

Zauner

(2020), “Three Decades of Customer Value Research,” Journal of Service Research, 23 (4), 409-432.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.88 MB