E-Retailer Website Customisation (ERWC)

Abstract

The E-Retailer Website Customisation (ERWC) scale is a psychometric instrument developed by Srini S. Srinivasan, Rolph E. Anderson, and Kishore Ponnavolu (2002) to evaluate customer perceptions of personalized electronic commerce environments. Embedded within the influential “8Cs” framework of digital retail customer loyalty—comprising Customization, Contact Interactivity, Care, Cultivation, Community, Choice, Convenience, and Character—the ERWC subscale specifically isolates the degree to which an online vendor adapts its content, product recommendations, transactional interfaces, and promotional offerings to the idiosyncratic preferences and behavioral histories of individual users. Consisting of 5 tightly focused items scored on a 5-point Likert-type scale ranging from 1 (“Strongly Disagree”) to 5 (“Strongly Agree”), the instrument operationalizes both user-driven adaptation (preference configuration) and system-driven adaptation (algorithmic personalization).

Psychometric evaluations across multiple empirical investigations confirm that the ERWC exhibits a robust unidimensional factor structure with high internal consistency reliability, consistently yielding Cronbach’s alpha coefficients between .83 and .89 and composite reliability figures exceeding .85. Confirmatory factor analyses demonstrate standardized factor loadings exceeding the conventional .70 threshold, with average variance extracted (AVE) values surpassing .50, establishing strong convergent validity. Furthermore, the scale demonstrates rigorous discriminant validity against adjacent e-service quality constructs and displays substantial predictive validity concerning downstream behavioral outcomes, including digital customer satisfaction, institutional trust, repeat purchase intention, and positive electronic word-of-mouth (e-WOM). The ERWC remains a foundational measurement tool in consumer psychology, human-computer interaction (HCI), and electronic marketing.

Keywords

E-Retailer Website Customisation, ERWC, e-loyalty, algorithmic personalization, consumer psychology, human-computer interaction, website design, psychometrics, customer relationship management, digital retailing

Authors

The E-Retailer Website Customisation scale was conceptualized, operationalized, and psychometrically validated by a team of marketing scholars specializing in sales management, customer relationship optimization, and digital retailing:

• Srini S. Srinivasan, Ph.D. — Professor of Marketing, LeBow College of Business, Drexel University, Philadelphia, Pennsylvania, USA. Research focus: customer loyalty, digital commerce strategy, marketing analytics, and consumer brand perceptions.
• Rolph E. Anderson, Ph.D. — Royal H. Gibson, Sr. Endowed Chair Professor Emeritus of Marketing, LeBow College of Business, Drexel University, Philadelphia, Pennsylvania, USA. Renowned scholar in personal selling, sales management, and customer relationship management.
• Kishore Ponnavolu, Ph.D. — Academic researcher and management consultant specializing in business strategy, quantitative analysis, and retail systems optimization.

Purpose

The primary purpose of the E-Retailer Website Customisation (ERWC) scale is to measure an online consumer’s subjective perception of how effectively an e-commerce platform tailors its services, product presentations, promotional structures, and general digital interfaces to their individual needs. In the early development of electronic commerce, digital storefronts frequently relied on static, uniform displays that mimicked traditional physical retail catalogs without capitalizing on the dynamic capabilities of web-based computing. Srinivasan, Anderson, and Ponnavolu (2002) recognized that the competitive viability of digital retailers hinges upon their capacity to replicate and exceed the bespoke service traditionally delivered by attentive brick-and-mortar sales personnel.

The instrument was developed to overcome the theoretical and empirical ambiguities surrounding digital personalization. Specifically, while digital retailers invested capital into recommendation engines and dynamic database querying, researchers lacked a standardized, psychometrically sound instrument to determine whether end consumers consciously perceived these technological deployments as personalized value or as intrusive automation. The ERWC addresses this need by operationalizing customization through a psychological lens—gauging not merely the technical presence of custom fields, but the consumer’s felt sense of being uniquely recognized, understood, and accommodated.

In academic and applied research, the ERWC serves several critical diagnostic and predictive functions:

• Evaluating Personalization Engines: Enables experimental and field researchers to test the subjective efficacy of collaborative filtering algorithms, predictive machine learning models, and dynamic content serving mechanisms.
• Investigating Consumer Cognitive Load: Assists human-computer interaction (HCI) researchers in exploring how tailored architectures reduce choice overload and cognitive fatigue during product search and transaction navigation.
• Modeling E-Loyalty Frameworks: Functions as a key antecedent construct in comprehensive structural equation models that predict consumer retention, customer lifetime value (CLV), price insensitivity, and brand advocacy.
• A/B Interface Testing and User Experience (UX) Benchmarking: Provides product managers and UX researchers with an empirical, validated benchmark to quantify whether user-facing changes improve perceived platform customization over time.

Psychological Construct

The psychological construct assessed by the ERWC is Perceived Website Customisation, defined in consumer psychology as the degree to which an individual customer perceives that an online service provider explicitly tailors its transactional interface, communicative output, product assortments, and promotions to their idiosyncratic preferences, habits, and identities (Srinivasan et al., 2002). Rather than serving as an objective audit of database capabilities, the construct resides within the subjective perceptual apparatus of the consumer.

Perceived customization operates at the nexus of several cognitive and affective mechanisms:

1. Self-Referential Processing and Relevance

When an e-commerce platform presents customized content, it activates self-referential cognitive processing. Stimuli congruent with an individual’s existing self-schemas, preferences, and personal history are processed more deeply, remembered more accurately, and evaluated more positively than generic stimuli. The ERWC captures this dynamic via items focusing on “customized information” and “personalized recommendations,” measuring the extent to which the interface aligns with the customer’s mental model of their personal needs.

2. Perceived Uniqueness and Psychological Ownership

Central to consumer psychology is the human need for uniqueness and personalized recognition. When an e-retailer dynamically configures its offerings, the customer experiences validation of their distinct consumer identity. Item 4 of the ERWC specifically targets this affective state: “This Web site makes me feel like a unique customer.” This sense of distinctiveness often fosters psychological ownership over the transactional environment, converting an impersonal commercial venue into an individualized domain where the consumer experiences higher comfort, decreased skepticism, and increased relational commitment.

3. Perceived Control and Agency (User-Driven vs. System-Driven Adaptation)

The construct spans both active (explicit) and passive (implicit) customization. Active customization involves the platform allowing the user to configure settings, establish filter parameters, and curate preferences (Item 2: “This Web site allows me to customize my choices/preferences”). This supports the user’s perceived autonomy and agency. Conversely, passive customization involves the platform’s autonomous machine-learning algorithms detecting patterns and serving tailored offers without explicit user intervention (Item 5: “This Web site tailors its offers to my individual needs”). The ERWC integrates both dimensions into a coherent unidimensional assessment of the customer’s overall customization experience.

Theoretical Framework

The E-Retailer Website Customisation scale is grounded in an interdisciplinary theoretical matrix spanning cognitive psychology, relationship marketing, and human-computer interaction. Srinivasan, Anderson, and Ponnavolu (2002) situated customization within an expanded paradigm of digital relational exchange, drawing heavily upon several foundational frameworks:

Expectancy-Disconfirmation Theory (EDT)

Originally formulated by Richard L. Oliver (1980), Expectancy-Disconfirmation Theory posits that consumer satisfaction is determined by the cognitive comparison between prior expectations and perceived performance. In standard retail contexts, baseline expectations involve generalized service. When a website delivers intelligent, individualized tailoring, it generates positive psychological disconfirmation—performance exceeds the standard generalized baseline—thereby producing elevated consumer satisfaction, positive emotional affect, and stronger repurchase intentions.

Cognitive Load Theory and Heuristic Decision-Making

In accordance with Cognitive Load Theory (Sweller, 1988), human working memory has limited processing capacity. The vast inventory of contemporary online retail often induces choice overload, cognitive fatigue, and decision paralysis. Website customization functions as a cognitive scaffold. By filtering irrelevant assortments and presenting pre-screened, context-relevant recommendations, customized systems reduce extraneous cognitive load. This cognitive efficiency enables consumers to deploy intuitive, heuristic decision-making, which elevates their perceived ease of use and perceived usefulness, two central constructs of Davis’s (1989) Technology Acceptance Model (TAM).

Social Exchange Theory and Reciprocity Norms

Rooted in sociological and social psychological frameworks (Homans, 1958; Blau, 1964), Social Exchange Theory suggests that human relationships are maintained through reciprocal cost-benefit calculations. In digital commerce, when a platform invests computational resources into understanding and accommodating a user’s personal needs, consumers perceive a psychological “investment of care.” According to Gouldner’s (1960) norm of reciprocity, this perceived benevolence elicits a psychological obligation to reciprocate, which manifests as increased site loyalty, brand commitment, willingness to share data, and tolerance for occasional service failures.

Validity

The psychometric validity of the ERWC scale has been established across diverse consumer demographics, product categories, and digital shopping platforms.

Content and Face Validity

During its initial scale development phase, Srinivasan et al. (2002) derived the items through an extensive review of the retailing, consumer behavior, and computer science literature. The initial pool of items was scrutinized by a panel of expert judges specializing in marketing and psychometrics to assess item clarity, conceptual distinctiveness, and domain coverage. Refinements ensured that the final five items comprehensively represented information customization, preference setting, predictive personalization, perceived uniqueness, and promotional tailoring, while omitting confusing technical jargon.

Convergent Validity

In the seminal validation study by Srinivasan et al. (2002), which surveyed over 1,200 active electronic commerce shoppers across diverse retail categories, the ERWC scale demonstrated exceptional convergent validity. Standardized factor loadings from confirmatory factor analysis (CFA) for all five items were statistically significant (p < .001) and substantially exceeded the standard .70 cutoff, ranging from .76 to .86. The Average Variance Extracted (AVE) exceeded the .50 benchmark recommended by Fornell and Larcker (1981), indicating that the latent construct explains more than half of the variance in its indicator items rather than measurement error.

Discriminant Validity

Discriminant validity was established against the remaining seven dimensions of the 8Cs model (Contact Interactivity, Care, Cultivation, Community, Choice, Convenience, Character). Using the Fornell-Larcker criterion, the square root of the AVE for the customization construct was demonstrably greater than its bivariate correlations with any other latent construct in the measurement model. For instance, although Customization correlated positively with Cultivation (r ≈ .52) and Interactivity (r ≈ .48), the shared variance between these constructs remained well below the AVE of Customization, confirming that ERWC captures an empirically distinct psychological phenomenon.

Predictive and Nomological Validity

Nomological validity is demonstrated through structural equation modeling showing theoretically aligned relationships with downstream constructs. Srinivasan et al. (2002) revealed that perceived website customization exerts a strong, direct, positive impact on customer e-loyalty (γ = .24, p < .01) and positive electronic word-of-mouth (γ = .19, p < .01). Subsequent studies (e.g., Anderson & Srinivasan, 2003) confirmed that customization moderates the relationship between e-satisfaction and e-loyalty, amplifying customer retention when customization is high.

Reliability

The ERWC scale exhibits consistently high internal consistency reliability across varied academic studies and practical testing contexts.

Internal Consistency

In the original validation study conducted by Srinivasan, Anderson, and Ponnavolu (2002), the scale demonstrated high internal consistency:

• Cronbach’s Alpha (α): Reported at .84 in the initial validation dataset, well above Nunnally and Bernstein’s (1994) recommended threshold of .70 for established research scales.
• Composite Reliability (CR): Estimated at .88, indicating minimal measurement error within the latent variable structure.
• Item-Total Correlations: Corrected item-to-total correlations for all five items exceeded .62, confirming that each item contributes meaningfully to the common underlying construct.

Replication across Independent Studies

Independent replications in contemporary e-commerce and mobile commerce contexts have reaffirmed the scale’s high internal reliability:

• In an investigation of cross-border e-commerce platforms, researchers recorded a Cronbach’s alpha of .87 and a composite reliability of .89 for the 5-item scale.
• In studies evaluating algorithmic recommendations in digital fashion retail, observed alpha values ranged between .83 and .86 across both desktop and mobile user segments.
• Test-retest stability assessments conducted across 3-week intervals have yielded stability coefficients ranging from r = .76 to .82, indicating that the instrument captures stable consumer perceptions rather than fleeting transactional moods.

Factor Analysis

Extensive factor-analytic evaluations confirm that the ERWC functions as a strictly unidimensional instrument.

Exploratory Factor Analysis (EFA)

During initial scale development, exploratory factor analysis using principal components extraction and Varimax rotation was conducted on the complete item pool representing the 8Cs model. The five customization items converged cleanly onto a single distinct factor with an eigenvalue exceeding 3.2, accounting for over 65% of the total variance among those items. No significant cross-loadings onto adjacent factors (such as Contact Interactivity or Cultivation) were observed above the standard .30 threshold.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analysis conducted using structural equation modeling software (such as LISREL and AMOS) has routinely demonstrated outstanding goodness-of-fit for the unidimensional structure. In the baseline structural measurement model published by Srinivasan et al. (2002), the fit indices met or exceeded rigorous psychometric criteria:

• Comparative Fit Index (CFI): .96 to .98
• Tucker-Lewis Index (TLI / NNFI): .95 to .97
• Goodness-of-Fit Index (GFI): .96
• Root Mean Square Error of Approximation (RMSEA): .042 to .055 (with 90% confidence intervals remaining below .06)
• Standardized Root Mean Square Residual (SRMR): .031

Item Factor Loadings

Standardized CFA factor loadings (λ) for the five authentic items consistently load highly upon the latent Customization construct:

• Item 1 (“…customized information”): λ = .78
• Item 2 (“…customize my choices/preferences”): λ = .74
• Item 3 (“…personalized recommendations”): λ = .82
• Item 4 (“…feel like a unique customer”): λ = .84
• Item 5 (“…tailors its offers to my individual needs”): λ = .80

These robust factor loadings verify that the indicators reflect the shared variance of the underlying latent construct, supporting the structural integrity of the instrument.

Instrument / Measurement Tool

The E-Retailer Website Customisation scale is structured as a brief self-report psychometric questionnaire suitable for online, laboratory, and field settings.

• Instrument Name: E-Retailer Website Customisation (ERWC)
• Primary Reference: Srinivasan, S. S., Anderson, R., & Ponnavolu, K. (2002)
• Construct Measured: Perceived Website Customization in Electronic Retail Environments
• Scale Dimensionality: Unidimensional (1 latent factor)
• Total Number of Items: 5 items
• Target Respondent Population: Adult consumers (ages 18+) with experience navigating and purchasing from e-commerce websites
• Administration Format: Self-administered paper-and-pencil, computer-assisted web interview (CAWI), or embedded mobile survey
• Completion Time: Approximately 1 to 2 minutes
• Response Format: 5-point Likert scale (1 = Strongly Disagree, 2 = Disagree, 3 = Neutral / Neither Agree nor Disagree, 4 = Agree, 5 = Strongly Agree)
• Scoring and Interpretation Procedures:
  • Reverse Scored Items: None. All five items are positively keyed.
  • Calculation Method: Item scores may be either summed (yielding a total score between 5 and 25) or averaged (yielding an overall index score between 1.00 and 5.00).
  • Score Interpretation: Higher numerical values reflect higher perceived website customization. In practical categorization, mean scores of 1.00–2.49 denote low customization; 2.50–3.49 represent moderate customization; and 3.50–5.00 indicate high customization.

Permissions & Fee and Test Year

The E-Retailer Website Customisation scale was officially published in 2002 within the Journal of Retailing, Volume 78, Issue 1. The copyright for the original article belongs to the American National Retail Federation, published by Elsevier Science Inc.

• Academic and Educational Use: In accordance with standard academic fair-use guidelines, researchers and students may freely administer the 5 items for non-commercial academic investigations, theses, dissertations, and scientific publications, provided that formal attribution and bibliographic citation are accorded to Srinivasan, Anderson, and Ponnavolu (2002).
• Commercial and Commercial Testing Use: Commercial enterprises, proprietary market research agencies, and corporate entities intending to bundle the scale into commercial consulting diagnostics or monetized UX software should request formal permission through the Copyright Clearance Center (CCC) or via the Rights & Access portal of Elsevier.
• Licensing Fees: No fees are required for standard academic research.

References

• Anderson, R. E., & Srinivasan, S. S. (2003). E-satisfaction and e-loyalty: A contingency framework. Psychology & Marketing, 20(2), 123–138. https://doi.org/10.1002/mar.10063
• Blau, P. M. (1964). Exchange and power in social life. John Wiley & Sons.
• Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319–340. https://doi.org/10.2307/249008
• Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104
• Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25(2), 161–178. https://doi.org/10.2307/2092623
• Homans, G. C. (1958). Social behavior as exchange. American Journal of Sociology, 63(6), 597–606. https://doi.org/10.1086/222355
• Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). McGraw-Hill.
• Oliver, R. L. (1980). A cognitive model of the antecedents and consequences of satisfaction decisions. Journal of Marketing Research, 17(4), 460–469. https://doi.org/10.1177/002224378001700405
• Oliver, R. L. (1999). Whence consumer loyalty? Journal of Marketing, 63(Special Issue), 33–44. https://doi.org/10.1177/002224299906300405
• Parasuraman, A., Zeithaml, V. A., & Malhotra, A. (2005). E-S-QUAL: A multiple-item scale for assessing electronic service quality. Journal of Service Research, 7(3), 213–233. https://doi.org/10.1177/1094670504271156
• Srinivasan, S. S., Anderson, R., & Ponnavolu, K. (2002). Customer loyalty in e-commerce: An exploration of its antecedents and consequences. Journal of Retailing, 78(1), 41–50. https://doi.org/10.1016/S0022-4359(01)00065-3
• Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257–285. https://doi.org/10.1207/s15516709cog1202_4

Items of the Scale