Experiential branding in the digital age: evaluative and experiential pathways from brand experience to customer satisfaction

Purpose This study examines the psychological mechanisms linking brand experience and customer satisfaction. Drawing on the stimulus–organism–response (S-O-R) framework and appraisal theory, it investigates the mediating roles of brand authenticity and brand engagement, treated, respectively, as an evaluative (cognitive-appraisal) and a multidimensional experiential-investment organismic pathway. Design/methodology/approach Data were collected from 304 smartphone consumers in Haryana NCR (India) using a structured questionnaire and a non-probability convenience sample. Partial least squares structural equation modelling (PLS-SEM, SmartPLS 4) was used to test the model, which incorporates two reflective higher-order constructs and two mediators modelled in parallel (a formative Mode B re-estimation of the higher-order constructs, a sensitivity analysis with and without one affection item and a bootstrap contrast of the two specific indirect effects were added as robustness checks and a formal comparison). One indicator (AFFBENG3; loading 0.376) was removed post hoc; the structural results are the same when it is retained. Findings Brand experience is positively related to customer satisfaction, brand authenticity and brand engagement. Brand engagement ( β = 0.424) and brand authenticity ( β = 0.128) are both positively related to satisfaction, with engagement showing the numerically larger coefficient. Both specific indirect effects are significant (through engagement β = 0.107; through authenticity β = 0.048) alongside a significant direct path (β = 0.136), a pattern consistent with complementary partial mediation through two mediators modelled in parallel; because the direct path is small, this classification is treated cautiously. A bootstrap contrast showed that the difference between the two specific indirect effects (Δ β = 0.059) is not statistically significant (95% bias-corrected CI [ − 0.013, 0.155]; p = .166) and the authenticity → satisfaction association is very small ( f 2 = 0.019). Retaining AFFBENG3 or specifying the higher-order constructs formatively (Mode B) gave substantively the same direct and indirect effects. The parallel structure is consistent with the data but was not tested against serial alternatives. Brand experience explains limited variance in engagement ( R 2 = 0.064) and authenticity ( R 2 = 0.142) and out-of-sample predictive power is modest (Q 2 predict = 0.049–0.122). Because the design is cross-sectional and the sample is a convenience sample, the results describe associations and not causal effects. Research implications The study refines the explanation of the brand experience–satisfaction link by separating, within one model, an evaluative appraisal of the brand (authenticity) from the consumer’s own experiential investment in it (engagement), in the smartphone category, where mediation evidence has so far focused on brand equity and repurchase intention. Both pathways are statistically significant; the engagement pathway has the numerically larger coefficient, whereas the authenticity pathway is significant but very small. The difference between the two pathways is not statistically significant, so the data do not separate them. The contribution is limited to the simultaneous associations of the two mediators with satisfaction and does not claim superiority over established mediators. Practical implications Brand engagement has the highest importance in the IPMA and is the more promising lever for managers, although its advantage over authenticity is not statistically significant (bootstrap contrast, p = 0.166). Brand authenticity, already the best-performing construct, adds a very small contribution, and improving brand experience itself, which has the lowest performance score, offers a further route to satisfaction gains. Originality The study simultaneously tests an evaluative-appraisal mediator and a multidimensional experiential-investment mediator within one S-O-R model with higher-order constructs and combines bootstrap mediation analysis, PLSpredict, CVPAT and IPMA to compare the empirical weight of the two pathways.