Perceived Ad Customization

Abstract

The Perceived Ad Customization scale, developed and validated by Freya De Keyzer, Nathalie Dens, and Patrick De Pelsmacker (2022), is a concise, psychometrically robust three-item self-report instrument designed to quantify an individual’s subjective appraisal of the degree to which an advertisement is personalized, tailored, and customized to their specific identity, interests, and preferences. Adapted from foundational literature in interactive marketing and personalized communication, the instrument assesses a core cognitive-affective appraisal that operates as a critical mediating variable between objective algorithmic targeting and downstream consumer behavior. Administered on a seven-point Likert response format ranging from 1 (strongly disagree) to 7 (strongly agree), the unidimensional measure yields a composite score obtained by computing the unweighted arithmetic mean across the three items. Psychometric validation conducted across a pretest and two extensive empirical studies demonstrates exceptional internal consistency (Cronbach’s $\alpha$ routinely exceeding .90, composite reliability $\text{CR} > .90$), robust construct and convergent validity (Average Variance Extracted $\text{AVE} > .80$), and clear discriminant validity from related constructs such as perceived privacy invasion, intrusive surveillance, and general advertising relevance. Confirmatory factor analysis confirms an invariant unidimensional structure with standardized factor loadings typically exceeding .85. The instrument has established substantial utility in digital advertising, human-computer interaction, consumer psychology, and privacy research, enabling investigators to isolate subjective perception from objective targeting manipulation.

Keywords

Perceived ad customization, personalized advertising, tailored advertising, consumer psychology, psychometric scale, cognitive appraisal, advertising effectiveness, brand attitude, click intention, word-of-mouth intention, privacy calculus, digital marketing

Authors

The scale was formalized and validated by an academic research team specializing in marketing communications, consumer behavior, and interactive media at the University of Antwerp, Belgium:

• Freya De Keyzer, Ph.D. — Department of Marketing, Faculty of Business and Economics, University of Antwerp, Antwerp, Belgium. Her research focuses on personalized advertising, social media marketing, consumer empowerment, and digital communication strategies.
• Nathalie Dens, Ph.D. — Full Professor of Marketing, Faculty of Business and Economics, University of Antwerp, Antwerp, Belgium. Her research interests encompass advertising effectiveness, digital and interactive marketing communications, brand management, and consumer decision-making.
• Patrick De Pelsmacker, Ph.D. — Emeritus Professor of Marketing, Faculty of Business and Economics, University of Antwerp, Antwerp, Belgium. A leading European scholar in marketing communications, advertising psychology, and consumer ethics, he has published extensively on digital media effects and advertising responsiveness.

Corresponding author inquiries regarding the foundational empirical validation are historically directed through the Department of Marketing at the University of Antwerp or via the publication registry of the Journal of Advertising.

Purpose

The primary purpose of the Perceived Ad Customization scale is to assess the subjective, phenomenological perception of commercial messaging tailored specifically to an individual consumer. In the modern programmatic advertising ecosystem, algorithmic systems continuously deploy behavioral targeting, geotargeting, and psychographic profiling to deliver targeted promotional content. However, an enduring theoretical and empirical challenge in communication science and consumer psychology is that objective personalization—such as the algorithmic injection of user search keywords, demographic markers, or localized data—does not automatically correspond to a user’s subjective experience of customization. Consumers frequently misattribute, overlook, or fail to recognize targeted elements; conversely, highly accurate recommendations may occasionally be perceived as coincidental rather than customized.

Consequently, researchers require an instrument that captures this subjective cognitive appraisal cleanly, reliably, and without construct contamination. De Keyzer, Dens, and De Pelsmacker (2022) developed this scale to serve as both an experimental manipulation check and a core psychological mediator. In experimental paradigms, the scale confirms whether experimental manipulations of customization (e.g., highly personalized versus non-personalized ad variants) are successfully perceived by participants. In explanatory structural models, the scale serves as an essential cognitive antecedent that triggers downstream attitudinal and behavioral outcomes, such as brand attitude ($A_{brand}$), click-through intentions, and word-of-mouth (WOM) intentions.

Furthermore, the instrument is critical for studying the "personalization paradox" or "privacy-personalization paradox." This phenomenon occurs when increased customization simultaneously drives consumer engagement via heightened perceived relevance while eliciting psychological reactance, perceived surveillance, and defensive behavior via privacy concerns. By isolating the precise extent to which an ad is perceived as tailored to one’s interests, researchers can establish empirical boundary conditions, moderation thresholds, and non-linear associations governing consumer responses to big-data-driven marketing campaigns.

Psychological Construct

The construct of Perceived Ad Customization is defined as the degree to which a consumer subjectively evaluates an advertisement as being specifically designed, adapted, or configured to reflect their unique individual preferences, interests, needs, or identity. Grounded in cognitive psychology and information processing theory, it represents a conscious cognitive appraisal formed during or immediately following exposure to a stimulus.

The construct encompasses three closely related semantic dimensions that are integrated into a single, cohesive, unidimensional factor:

• Personalization: The attribution that the message is oriented toward the self rather than a generic or mass audience. This facet engages self-referent cognitive processing, wherein the individual recognizes their own life, background, or identity markers mirrored within the commercial message.
• Tailoring: The perception that the informational content, visual framing, or product offerings have been modified or filtered to align directly with the user’s specific tastes, needs, or behavioral tendencies.
• Individual Customization: The overarching assessment that the communicative artifact was uniquely assembled "for me," creating an experience of one-to-one dialogue between the sender (the brand or platform) and the recipient.

Importantly, perceived ad customization must be conceptually distinguished from adjacent advertising constructs. It is distinct from perceived ad relevance, which describes whether an advertisement is useful, interesting, or timely to a consumer, regardless of whether they believe the advertiser intentionally configured it for them. An ad for winter coats shown on a snowy day may have high perceived relevance without possessing high perceived customization. Similarly, it is distinct from perceived ad intrusiveness or privacy infringement, which represent negative affective and cognitive reactions to perceived boundary violations. While perceived customization often covaries with perceived privacy threat, it remains a neutral-to-positive cognitive appraisal of stimulus fit.

When perceived customization is activated, it triggers heightened cognitive elaboration, encourages self-referencing (connecting ad content to personal autobiographical memory), and increases perceived diagnostic value, ultimately fostering favorable brand evaluations when boundary conditions such as privacy fears are managed.

Theoretical Framework

The operationalization of perceived ad customization is anchored in several prominent psychological and communication frameworks:

1. Cognitive Appraisal Theory

According to Cognitive Appraisal Theory, formulated by Richard Lazarus and Susan Folkman, individuals continuously appraise external environmental stimuli regarding their personal significance, valence, and relevance to personal well-being. Perceived ad customization reflects an essential primary appraisal: the individual evaluates whether the communicative encounter is personally salient, beneficial, or congruent with personal goals. When a message is appraised as customized, it is categorized as high in goal congruence, eliciting positive cognitive and affective reactions.

2. The Elaboration Likelihood Model (ELM)

Under the Elaboration Likelihood Model of persuasion developed by Richard E. Petty and John T. Cacioppo, an individual’s motivation and ability to scrutinize persuasive arguments dictate whether attitude change occurs via the central or peripheral route. Perceived ad customization serves as a potent motivational trigger. When consumers recognize that an advertisement is customized to their unique preferences, self-relevance increases dramatically. This heightened self-relevance enhances motivation to process message arguments, driving deeper cognitive elaboration via the central route and leading to more enduring brand attitudes and stronger behavioral intentions.

3. The Persuasion Knowledge Model (PKM)

Marian Friestad and Peter Wright’s Persuasion Knowledge Model posits that over time, consumers develop intuitive theories about the tactics, motives, and goals of marketers. In the context of customization, perceived ad customization can activate persuasion knowledge. Consumers may recognize targeting tactics (e.g., "This ad reflects my recent browsing history"), which can prompt skepticism or resistance if the customization is perceived as manipulative, creepy, or excessive. Thus, perceived ad customization sits at the intersection of processing facilitation and persuasion knowledge activation.

4. Privacy Calculus Theory

Within interactive digital media, consumers constantly engage in an implicit cost-benefit analysis known as the privacy calculus. Perceived ad customization represents the primary tangible benefit received in exchange for personal data disclosures. If the degree of customization provides tangible utility, personal relevance, and cognitive efficiency, it can offset perceived privacy risks, fostering positive consumer engagement.

Validity

The Perceived Ad Customization scale possesses robust empirical validity across multiple rigorous experimental and cross-sectional studies:

Construct and Convergent Validity

In the scale development and validation studies by De Keyzer et al. (2022), convergent validity was evaluated using Confirmatory Factor Analysis (CFA). Across both exploratory pretests and primary experimental studies, the standardized factor loadings ($lambda$) for all three items loaded cleanly onto a single latent factor with values consistently exceeding .85 (often falling between .88 and .95, $p < .001$). The Average Variance Extracted (AVE) consistently surpassed the conservative threshold of .50 proposed by Fornell and Larcker, commonly reaching values between .80 and .88. This establishes that the vast majority of variance in the observed indicators is directly attributable to the latent construct of perceived customization rather than measurement error.

Discriminant Validity

Discriminant validity was established against structurally proximate and theoretically competing constructs, including perceived ad relevance, brand familiarity, perceived ad intrusiveness, and privacy concern. Using the Fornell-Larcker criterion, the square root of the AVE for perceived ad customization exceeded its highest bivariate correlation with any other latent construct in the measurement model. Furthermore, contemporary assessments using the Heterotrait-Monotrait ratio of correlations (HTMT) remained well below the conservative .85 threshold, demonstrating that the scale measures a construct psychometrically distinct from general message involvement or perceived manipulation.

Predictive and Nomological Validity

The nomological network was validated through structural equation modeling (SEM) examining direct and indirect pathways to consumer responses. Perceived ad customization demonstrated significant positive predictive validity for brand attitude ($A_{brand}$), click intentions, and electronic word-of-mouth (eWOM) intentions. De Keyzer et al. (2022) revealed that perceived ad customization mediated the effect of algorithmic personalization on downstream marketing outcomes, while also interacting with privacy concerns and social media platform contexts. These replicable structural relationships substantiate the nomological validity of the measure in digital contexts.

Reliability

The three-item Perceived Ad Customization scale demonstrates exceptional internal consistency reliability across varied empirical samples and experimental conditions:

• Cronbach’s Alpha ($lpha$): Across the pretest and two formal experimental studies reported by De Keyzer et al. (2022), Cronbach’s alpha values ranged from .91 to .94, well above the conventional benchmark of .70 (and the stricter .80 threshold for applied research).
• Composite Reliability (CR): Raykov’s composite reliability and McDonald’s omega ($\omega$) metrics similarly surpassed .92 in structural equation evaluations, verifying that the scale items maintain strong internal consistency without suffering from excessive, redundant item bloat.
• Stability Across Subsamples: Multigroup invariance tests confirmed that the scale’s internal consistency remains stable across diverse experimental conditions (e.g., high vs. low personalization manipulations, varying levels of social endorsement) and respondent demographic groups.

Because the measure consists of three highly cohesive items, it achieves optimal parsimony, avoiding respondent fatigue while maintaining the minimal item count required for just-identified or over-identified latent variable modeling in structural equation packages (e.g., AMOS, Mplus, lavaan).

Factor Analysis

Psychometric evaluations employing Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) demonstrate that the Perceived Ad Customization scale is definitively unidimensional:

Exploratory Factor Analysis (EFA)

Initial principal components and maximum likelihood factor analyses with unconstrained extraction criteria uniformly identify a single factor with an eigenvalue substantially exceeding 1.0 (typically $\lambda_1 > 2.50$). Scree plots show a single sharp drop followed by an immediate leveling off, with the primary factor explaining between 82% and 88% of the total item variance. No secondary factors emerge with eigenvalues exceeding 0.35, verifying strict unidimensionality.

Confirmatory Factor Analysis (CFA)

When evaluated as an isolated latent measurement model or embedded within larger structural models, the three-item specification yields an excellent fit to empirical data. In a saturated three-indicator model, degrees of freedom ($df = 0$) require evaluation in the context of broader measurement batteries. When estimated alongside multi-item measures of brand attitude, click intention, and privacy concerns, the measurement models demonstrate outstanding fit indices:

• $\chi^2/df < 2.5$
• Comparative Fit Index (CFI) $ge .98$
• Tucker-Lewis Index (TLI) $ge .97$
• Root Mean Square Error of Approximation (RMSEA) $le .05$ (with 90% confidence intervals spanning .00 to .08)
• Standardized Root Mean Square Residual (SRMR) $le .03$

All three items exhibit high, statistically significant standardized factor loadings ($\lambda_1 \approx .90$, $\lambda_2 \approx .92$, $\lambda_3 \approx .89$, all $p < .001$), confirming that each statement serves as a potent indicator of the latent construct.

Instrument / Measurement Tool

The scale is structured as follows:

• Construct Assessed: Perceived Ad Customization
• Instrument Type: Self-report psychometric scale
• Administration Format: Paper-and-pencil, computer-assisted, or online web-based survey
• Target Population: Consumers, internet users, and experimental participants exposed to commercial messages
• Total Items: 3
• Response Format: 7-point Likert scale (1 = strongly disagree, 7 = strongly agree)
• Estimated Completion Time: Under 1 minute (approximately 20–30 seconds)
• Scoring Procedure: Items are averaged to create an overall composite index of perceived ad customization. Higher numerical scores reflect greater perceived ad customization.
• Reverse-Scored Items: None; all three items are framed positively.
• Interpretation: Mean scores from 1.00 to 2.99 indicate low perceived customization (generic ad perception); 3.00 to 4.99 indicate moderate or ambiguous customization; 5.00 to 7.00 reflect high perceived customization (strong alignment with personal profile).

Permissions & Fee and Test Year

The Perceived Ad Customization scale was published in 2022 by Freya De Keyzer, Nathalie Dens, and Patrick De Pelsmacker in the Journal of Advertising (Volume 51, Issue 1, pages 39–56). The scale items are publicly documented in the published literature for academic, scholarly, and non-commercial educational research purposes under standard academic citation practices. No administrative or licensing fee is required for non-commercial academic research. Researchers planning commercial deployments, inclusion in proprietary diagnostic platforms, or large-scale corporate consumer insight testing should consult the copyright policies of the publisher (Taylor & Francis / Routledge) and seek direct guidance from the authors.

References

• De Keyzer, F., Dens, N., & De Pelsmacker, P. (2022). How and when personalized advertising leads to brand attitude, click, and WOM intention. Journal of Advertising, 51(1), 39–56. https://doi.org/10.1080/00913367.2021.1981496
• Friestad, M., & Wright, P. (1994). The persuasion knowledge model: How people cope with persuasion attempts. Journal of Consumer Research, 21(1), 1–31. https://doi.org/10.1086/209380
• Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. Springer Publishing Company.
• Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. https://doi.org/10.1016/S0065-2601(08)60214-2
• Shanahan, K. J., Tran, T. P., & Taylor, E. C. (2019). Overcoming extreme consumer skepticism to generate personal relevance and favorable consumer outcomes: How personalized ads optimize cognitive processing and lead to meaningful outcomes. Journal of Advertising, 48(2), 168–185. https://doi.org/10.1080/00913367.2019.1585308

Items of the Scale