Grocery retailer ’ s brand performances from brand personalities and marketing offerings

Purpose – Despite competition and supply-chain disruptions during Covid-19 pandemic (2019 – 2021), one grocery retailer consistently thrived and was ranked top. The sources of the sustained performances needed examination. Guided by self-congruity theory and integrating three models, the authors examined how much theretailer ’ sbrandperformances(brandloyalty,equity,preferenceandrepurchaseintentions)wereemanating frombrandpersonalitiesandmarketingofferings.Themediatingrolesofbrandloyaltyandequityweretested. Design/methodology/approach – Cross-sectional datawas collectedfrom 480 frequent customers usingan online questionnaire posted on the researchers ’ social media pages. Factor analysis was conducted to identify the dimension that best describes the grocery retailer. Partial least square – structural equation modelling (PLS-SEM) was used to test a conceptual model. Findings – Factor analysis results show that brand sincerity (28.582% variance-explained; M 5 4.1) was top (factor 1), followed by excitement (20.336% variance-explained; M 5 3.9) and then trustworthiness (18.854% variance-explained; M 5 3.87). PLS-SEM results revealed that two brand personalities (brand excitement and trustworthiness) and marketing offerings (price, place, product, promotion) impacted loyalty found to be a strong driver of brand equity. Repurchase intention and brand preference were influenced by brand equity. Brand loyalty mediated most of the relationships between brand personality dimensions, marketing offerings andbrandequity.Brandequityalsosignificantlymediatedtherelationshipsbetweenbrandloyalty,preference andrepurchaseintentions.Theintegratedmodelproducedhighexplanatorypowerswithbrandequity(67.8%),brandpreference(71.7%),brandloyalty(63.2%)andrepurchaseintentions(54.2%). Originality/value – The study extends a brand personality-loyalty model through integrating two other models that provided marketing offerings and brand equity outcomes. It demonstrates that a stream of profitable customers ’ responses awaits a retailer who holds both brand and customer mindsets by building admired brand personalities while providing desired marketing offerings.


Introduction
Retailers' brands are increasingly featuring in the list of top global brands.For example, retailers like Amazon, Zara, H&M and IKEA were rated amongst the top 40 global brands in 2020 by Interbrand (2020).In South Africa, retailers, such as Woolworths, Pick n Pay, Checkers, Mr Price and Clicks were within the top 25 South African brands in 2020.Despite the high competition amongst these retailers and in the three difficult years of the Covid-19 pandemic (2019Covid-19 pandemic ( -2021) ) with supply chain interruptions, Woolworths consistently maintained its top retailer brand rankings in 2019-2022 (BrandZ TM , 2020; BrandFinance, 2022).Even in brand value, Woolworths was the number 1 most valued grocery retailer brand in South Africa in 2019 (BrandZ TM , 2019).In 2022, Woolworths again emerged as South Africa's number 1 retailer brand in brand value (R17.3bn-$950.1m)and strength with AAA þ customer rating.Another top rated and growing grocery retailer was Checkers (BrandFinance, 2022).It is asserted that the key driver of Woolworths' brand strength is "parochial support in South Africa, where local consumers and stakeholders feel a national affinity with Woolworths" (BrandFinance, 2022).Being Africa's most developed economy, South Africa's retail market is worth R1trillion ($62bn) and consumers' love affair with their shopping centres and favourite retailers is strong (Dludla, 2022).
Loved and top brands perform well in brand loyalty, preference, and most of all, brand equity (Viet and Anh, 2021).Viet and Anh (2021, p. 74) describe brand equity as "added value uniquely generated from having a strong brand and which emanates from consumers' differential and favourable responses to the marketing mix/offerings of the brand; increased utility/preferences that the brand name adds to the products and resultant financial value enjoyed because of the loved brand".A retailer's brand equity is "a valuable and intangible asset that affects consumer behaviour and the retailer's performances" (Troiville et al., 2019, p. 73).Thus, brand equity has strategic importance to brand owners, including retailers.Troiville et al. (2019) assert that retailers earn product-market (e.g.brand preference, power against competitors, repurchase intentions) and financialmarket (e.g.share of wallets, revenue and profits) benefits from brand equity, but these brand equity outcomes are often assumed and rarely measured (Uford and Duh, 2021).What drives brand equity in the highly competitive retailing industry is another question (Willems, 2022).
In this era of digital transformation, Han et al. (2021) recommend a customer management mindset (CMM) in building brand equity.This approach seeks to understand customers' changing preferences and providing sustainable and satisfying marketing offerings (S anchez-Gonz alez et al., 2022).According to Ali et al. (2021) and Roggeveen et al. (2021), brand equity is built from a brand management mindset (BMM) that provides symbolic benefits, such that the personality of the brand creates a good consumer-brand congruence.Willems (2022) also contend that a "brand's personality may be used to appeal to consumers, differentiate the brand from competitors" and create customer-based retailer brand equity.Thus, and according to the self-congruity theory, when the personality of a brand or store is perceived by consumers as congruent with their actual or ideal self/personalities, it affects their behaviour (Sirgy, 2018), to the point of generating brand loyalty (Das, 2014) and customer-based brand equity (Willems, 2022).While acknowledging the importance of BMM in retailers' brand performances, Roggeveen et al. (2021) recommend creative and innovative merchandise offerings as a good starting point.Thus, both CMM and BMM are recommended, but considering the cost of implementing both strategies, an assessment of the greater contributor to brand performances is needed for appropriate resources allocation (Han et al., 2021).
This study is aimed at examining how much of Woolworths' CMM in terms of marketing offerings (price, promotion, product quality, distribution) and BMM in terms of brand personalities are driving its performances (brand loyalty, preference, equity, repurchase intentions).Examining contributions of both mindsets is important because according to the stimulus-organism-response framework, consumers respond to varied stimuli/triggers (e.g.marketing and brand offerings) when shopping.Insights into the varied and most impactful triggers are important for appropriate positioning and sustained retailer performance (Gupta and Mukherjee, 2022).IJRDM 51,13 2. Establishing relevant models and theory for this study and objectives Researchers (e.g.Viet and Anh, 2021;Willems, 2022) agree that retailers' brand equity is key to success.However, there is uncertainty as to the key sources of, and retailer-related outcomes/benefits of brand equity.For sources, the traditional customer-based brand equity (CBBE) model of Aaker (1991) has been predominantly used.The model postulates that brand equity emanates from brand knowledge structure (comprising brand awareness and image), perceived quality, brand loyalty and proprietary assets.Despite the widespread use of Aaker's (1991) CBBE sources, the model has been criticised for being generic, and lacking marketing and retailer-specific factors, such as the four-marketing offerings (Efanny et al., 2018;Niazi et al., 2021).Also questioned, are the contributions of a retailer's brand personalities to CBBE (Sindhu et al., 2021) and resultant outcomes (Buil et al., 2013;Uford and Duh, 2021).
In terms of CBBE outcomes, Buil et al.'s (2013) model suggests brand preference, repurchase intention and premium price payment.Before the CBBE leads to these outcomes, the model postulates that brand awareness, image and perceived quality will interrelate, impact brand loyalty, a strong driver of brand equity.However, Buil et al.'s (2013) model solely focuses on the usual four Aaker's (1991) CBBE sources (i.e.brand awareness, image, loyalty and perceived quality).Consumers patronize retailers, especially grocery retailers for their utilitarian marketing offerings (price, distribution, promotion and product quality) (Anselmsson et al., 2017;Niazi et al., 2021;Viet and Anh, 2021), as well as for metaphorical, symbolic and psychological brand attributes/positioning (e.g.brand personalities) (Kuo et al., 2022;Sindhu et al., 2021;Willems, 2022).For example, and for the brand equity success of distribution companies and retailers, S anchez-Gonz alez et al. (2022) contend that sustainable marketing offerings are important.In fact, the building of the four Aaker's (1991) CBBE sources to a point of enjoying brand loyalty and favourable and differential consumer responses (CBBE) starts with appropriate marketing offerings (Anselmsson et al., 2017;Keller, 2020).The marketing offerings as suggested by Anselmsson et al.'s (2017) model, foster retailers' brand loyalty, which Buil et al.'s (2013) model presents as a predictor of brand equity.
Brand loyalty and resultant brand equity are also driven by brand personality (Das, 2014;Willems, 2022).Aaker (1997), a seminar author of brand personality, identifies five brand personality dimensions of sincerity, excitement, competence, ruggedness and sophistication.He defines the concept (p.347) as "the set of human characteristics associated with a brand".Although Aaker's (1997) five brand personality dimensions have been widely studied, researchers (e.g.Kuo et al., 2022;Sindhu et al., 2021;Youn and Dodoo, 2021) warn that not all five dimensions are applicable for all product categories.For example, Willems (2022) reports that utilitarian grocery retailers score high in competence and sincerity, while symbolic retailers like fashion retailers score high in excitement and sophistication dimensions.In a comparative study of nostalgic and non-nostalgic brands, Youn and Dodoo (2021) found excitement, sincerity and competence as common dimensions.In a fashion retailing context, Sindhu et al. (2021) also found excitement, sincerity, competence to be prominent dimensions.Thus, this study examines these three dimensions in a grocery retailing context and adds trustworthiness as a fourth dimension considering the importance of trust in buying foodrelated items.
When the brand personality dimensions are appealing for both fashion and grocery retailers, Willems (2022) found that brand equity is developed.Whether Woolworths have developed appealing brand personalities capable of impacting brand loyalty, equity and other brand performances warrants an examination.The examination is particularly important following Willems' (2022, p. 8) recommendation that there should be follow-up studies that fills "an impactful void in the field of the self-congruity theory in retailing".The self-congruity theory posits that consumers compare their perception of a brand's personality Grocery retailer's brand performances with their self-concept (true self, idealized self) and when they find congruity, it guides their attitudes, brand preference, choice and shopping behaviour, including brand loyalty and equity (Sirgy, 2018).While Willem (2022) questions the applicability of this theory for a utilitarian brand like a grocery retailer brand, Das' (2014) retailer brand personality-loyalty model shows that for a departmental store, a retailer's brand personality significantly impacts brand loyalty.However, the dimensions of brand personality making this impact were not studied.Thus, guided by the stimulus-organism-response framework and the self-congruity theory, this study extends Das' (2014)  (1) Examine how much Woolworths' customers scored the grocery retailer in four personality traits of brand excitement, sincerity, competence and trustworthiness.
(2) Examine which of Woolworths' brand personalities and marketing offerings (price, distribution, promotion and perceived product quality) are strongly impacting brand loyalty and brand equity.
(3) Assess how much brand equity drives two outcomes: brand preference and repurchase intentions.
(4) Test the mediating role of brand loyalty in how brand personalities and marketing offerings impact brand equity.
(5) Test the mediating role of brand equity in the relationships between brand loyalty, brand preference and repurchase intentions.
The study starts with a literature review, develops a conceptual model and hypotheses, after which research methods are discussed.The study ends with the presentation and discussion of results.
3. Literature review 3.1 Creating customer-brand congruity through customer and brand management mindsets Han et al. (2021, p. 582) assert that the customer management mindset (CMM) is a retailer's marketing strategy that "views the customer as the dominant focal unitirrespective of the brand" with expectations and measurements of customer acquisition, loyalty/retention, repurchase intentions, etc.This strategy requires the studying of customers' needs to provide appropriate marketing offerings.The advantage here is an effective resource allocation and customization of the four marketing offerings.Studying elements of the CMM, Troiville et al. (2019) found that while each of the four marketing offerings makes contributions to a retailer's brand equity, the product quality and assortment, perceived value and retailer store atmosphere made strongest contributions.Are the contributions sustainable?Han et al. (2021) assert that when the marketing mix elements are appropriately customized, they will be difficult for competitors' interpretation and copying.However, they worry that if a retailer has many varied customers, "a customer-focused strategy might not be profitable because of its high variable cost per customer".Despite the cost, Roggeveen et al. (2021) contend that a retailer's marketing offerings or merchandise coupled with store atmosphere (signage, d ecor, the sensory elements) and quality of salespeople interactions can inform and shape consumers' perceptions of the retail store's brand identity.Roggeveen et al. (2021, p. 84) define brand identity as a "unique set of associations a firm wants to create for its brand . . .which impact the actual associations consumers form with the brand (i.e. the brand IJRDM 51,13 image and brand personality)".When the created brand identity becomes congruent with customers' identity and self-concept and according to the self-congruity theory, Sirgy (2018) and Roggeveen et al. (2021) suggest that brand outcomes such as customer brand engagement, brand loyalty and willingness to pay a premium price is gained.These are suggestions that a CMM can enhance a BMM for brand performances.
The BMM focuses on building, positioning and fortifying the strength of a retailer's brand in consumers' mind.Unlike the CMM requiring an adjustment of the marketing offerings for multiple market segments, with a BMM, a retailer uses an identical selling proposition to a large market segment.The advantages of a BMM as Han et al. (2021, p. 582) highlight are sustainable long-term strong brand benefits and a "centrally steered resource deployment in the context of push marketing and above-the line spending which is very efficient".For example, building a strong retailer brand equity as aforementioned leads to product-market (e.g.brand preference, power against competitors, repurchase intentions) and financialmarket (e.g.share of wallets, revenue and profits) benefits (Troiville et al., 2019).According to Roggeveen et al. (2021), these benefits are enjoyed when the consumers perceive that a retailer's brand story and personalities are congruent with their life experiences and personalities.For example, Ali et al. (2021) found that when shoppers find the personality of a mall to be congruent with their personality, especially in terms of sophistication, this provides them utilitarian and hedonic benefits.The hedonic benefit in turn improves their well-being and patronage.
South Africa has one of the greatest number and love of shopping malls and many strong retailer brands competing in the malls (Dludla, 2022).Following the stimulus-organismresponse framework, it is imperative to examine how much of both the CMM in terms of marketing offerings and BMM in terms of brand personality dimensions drive brand loyalty, brand equity and its outcomes as presented in Figure 1.
3.2 Conceptual model and hypotheses development 3.2.1 Relationships between retailer brand personality dimensions, brand loyalty and brand equity.Brand personalities are symbolic features and more enduring differentiators (Saeed et al., 2021) and can lead to consumer-brand relationships according to the brand congruence theory (Sirgy, 2018).Retailers' brand personality can be built from the appearance/dressing of their employees/salesforce, retail store environment, marketing messages and people endorsing the retailers (Bairrada et al., 2019;Willems, 2022).In addition to differentiating a brand from its competitors, brand personalities increase brand preference, build brand relationships in terms of brand love and attachment and ultimately engender brand loyalty (Shetty and Fitzsimmons, 2021).
Brand sincerity is viewed in terms of being down-to-earth, honest, wholesome and cheerful.Brand competence is intelligence and success, while brand excitement is being daring, imaginative, up-to-date and spirited (Aaker, 1997:352).Brand trustworthiness is viewed in terms of believability, integrity and reliability, especially regarding marketing messages that retailers communicate about their offerings (Dissanayake and Weerasiri, 2017).Sassaman et al. (2019, p. 17) describe trustworthiness as "the degree to which an organization is regarded as friendly, personal, attentive to people, and family oriented".Trustworthiness positively impacts brand loyalty (Anselmsson et al., 2017) and positive judgement of an organization (Sassaman et al., 2019).Brand excitement is a direct driver of retailers' brand equity (Willems, 2022).According to Sindhu et al. (2021), a fashion retailer's brand competence, excitement and sincerity will first impact brand loyalty, a strong driver of brand equity.For a luxury brand and halal brand, Shetty and Fitzsimmons (2021) and Aji and Muslichah (2022) also found a direct relationship between brand personality and brand loyalty.The brand loyalty is not only a strong predictor of brand equity, but it also mediates Grocery retailer's brand performances in the relationship between perceived quality and brand equity (Buil et al., 2013;Naizi et al., 2021).Thus, the following hypotheses are formulated: H1. Woolworths' brand sincerity positively impacts its brand loyalty.
H2. Woolworths' brand excitement has a positive impact on its brand loyalty.
H3. Woolworths' brand competence has a positive impact on its brand loyalty.
H5. Customers' brand loyalty to Woolworths positively impacts its brand equity.
H6. Woolworths' brand loyalty will mediate in the relationships between the brand personality dimensions and brand equity.evaluations, purchase decisions and brand equity (Viet and Anh, 2021).For example, while promotions and a high perceived product quality attract customers to a retailer's store, a fair price is also important.Even the product packaging adds to the perceived quality found to be directly drive retailer brand equity (Jara et al., 2017).While a high price is expected to discourage brand purchase, Shariq (2019, p. 41) assesses that the high price is a communicator of a company's intended high product/brand value and a "strong brand should command higher prices than the competition".The attraction and knowledge of retailers' marketing offerings are only made possible by the awareness created by their promotional activities which includes price deals, advertisements and sales promotions (Anselmsson et al., 2017;Shariq, 2019).When retailers promote their products in various media spaces, including social media, Mathur (2018) asserts that it can lead to brand loyalty and ultimate brand equity.
In a product distribution chain, a retailer is the ultimate supplier of products to places where consumers can have access to, at the right price and time.In line with Shariq's (2019) measurement of distribution, we construe distribution in terms of making products available at various retail outlets both in breadth and depth.Distribution significantly impacts brand loyalty, which in turn drives brand equity (Niazi et al., 2021).
In addition to distribution, Niazi et al. (2021) also found that brand promotion, perceived product quality and not price are significant drivers of brand loyalty.Viet and Anh (2021) agree that marketing mix elements are important drivers of brand loyalty, but empirically found that it was only brand awareness generated from advertising activities that significantly impacted brand loyalty.While Anselmsson et al. (2017) suggest that the brand awareness and other marketing offerings, such as pricing policy, physical store, product quality and customer service will first impact brand loyalty, Troiville et al. (2019) found that most of a retailer's marketing mix elements including distribution (access and convenience), product quality and assortments directly impacted brand equity.They also found a strong positive relationship between a retailer's brand loyalty and equity.Shariq (2019) found that the marketing mix elements impacted brand equity directly or indirectly through its dimensions like brand loyalty.This suggests a mediating role of brand loyalty.Thus, the following hypotheses are proposed: H7. Brand promotion positively influences Woolworths' (a) brand loyalty and (b) brand equity.
H8. Woolworths' distribution positively influences its (a) brand loyalty and (b) brand equity.
H9. Woolworths' product quality has a positive impact on its (a) brand loyalty and (b) brand equity.
H10. Woolworths' prices positively influence its (a) brand loyalty and (b) brand equity.
H11. Woolworths' brand loyalty will mediate in the relationships between the marketing factors and brand equity.

3.2.3
Brand equity as a mediator and its outcomes of brand preference and repurchase intentions.Brand owners spend hugely to build and manage brand equity.It is therefore recommended that its outcomes/benefits be measured.Researchers in the financial discipline measure the outcomes of brand equity in their balance sheets as brand value.In the marketing discipline, it is suggested that they be measured by assessing brand performances in terms of how consumers respond to a brand (Buil et al., 2013).A strong retailer brand equity evokes past satisfactory shopping memories in consumers' minds and thus impacts future patronage or repurchase intentions.The equity also minimizes competitors' efforts and thus creates brand preference (Willems, 2022).Brand preference is the extent to which a customer Grocery retailer's brand performances consistently favours a particular brand provided by a company, over those of competitors, even when the competitors provide similar offerings (Buil et al., 2013).Repurchase intention is the willingness to continuously buy products and services from the same seller/retailers in the future based on satisfactory past experiences (Ali and Bhasin, 2019).While Lee et al.
(2020) see brand equity as impacting repurchase intentions indirectly and through brand preference, Uford and Duh (2021) found that brand equity directly impacted brand preference and repurchase intentions.Buil et al. (2013) show that brand equity mediates in the relationship between brand loyalty, brand preference and repurchase intentions.From these findings the following hypotheses are formulated: H12. Woolworths' brand equity positively impacts its brand preference.
H14. Woolworths' brand equity will mediate in the relationship between brand loyalty and brand preference.
H15. Woolworths' brand equity will mediate in the relationship between brand loyalty and repurchase intentions.

Data collection and dealing with common method variance bias
Guided by the research problem and objectives, this study employed quantitative methods to collect and analyse data.Cross-sectional data was collected from 480 Woolworths' customers after obtaining ethics clearance with ethics number H20/04/17.Considering that cross-sectional data risk having a common method variance (CMV) bias issue, this study tried to eliminate the bias using Harman's single-factor test whereby all the constricts' items are factor analysed (Podsakoff et al., 2003).To ascertain the absence of CMV, Podsakoff et al. (2003) recommend that the items should not load onto a single factor and the unrotated factor solutions should not have a factor with a variance explained of up to 50%.This study did not have all items loading unto a single factor.Furthermore, none of the factors obtained a variance explained of up to 50%, thus revealing no problem of CMV bias.

Sample and measures
The sample size of 480 drawn in this study was good following Kyriazos' (2018) assessment that for structural equation modelling (SEM), a sample size of 200 is regarded as fair, 300 is good, 500 is very good and 1000 excellent.Due to Covid-19 pandemic, data was collected with an online questionnaire, whose link was posted on the researchers' social media pages.
Respondents snowballed the link to other respondents.Therefore, a non-probability convenience and snowballing sampling techniques were used.The questionnaire, which was developed from previously validated scales were pretested with 35 Woolworths' customers.After testing for reliability and validity, some items were deleted from some constructs to obtain acceptable composite reliability and average variance extracted (AVE) figures.Brand competence was deleted from further analyses because of poor reliability scores and its AVE score was below 0.4, even after some items were deleted.Thus, only three (i.e.brand sincerity, excitement and trustworthiness) out of four studied brand personality constructs were further analysed.The measurement of constructs started with a screening question to obtain only respondents who have frequently shopped at Woolworths.Constructs' items were measured on five-point Likert scales, with of 1 5 strongly disagree and 5 5 strongly agree.
The sincerity, excitement, competence brand personality dimensions were measured with Aaker's (1997)

Data analyses
Considering that partial least squares structural equation modelling (PLS-SEM) tests large models better and is a better method for a prediction-oriented model testing than the covariance-based SEM (Troiville et al., 2019), PLS-SEM was used to analyse data.The data analyses started with a factor analysis of 13 items used to measure brand personality.Data analyses continued with descriptive statistics, followed by measurement and structural model evaluations.

Results of brand personality dimensions best describing Woolworths
An exploratory factor analyses (EFA) using the principal component method with varimax rotation was conducted to assess which brand personality dimension was rated as Woolworths' most prominent.When testing the dimensionality of a scale, especially that derived from more than one source (e.g. from Aaker, 1997;Anselmsson et al., 2017;Sassaman et al., 2019 for this study), Gregory et al. (2022) suggest that EFA should be used.A confirmatory factor analysis can then be used to assess the validity of the freely extracted factors or components (Gregory et al., 2022) as we did later with results in Table 2.The principal component method is used when the ranking of most prominent factors is required as was the case in this study (Hair et al., 2017).
The number of items first entered for EFA was 16.Two (CM1 and CM2) of the three items representing the competence dimension of brand personality cross-loaded into the trustworthiness and excitement dimensions respectively.With these cross-loadings and low factor-loadings scores below the recommended 0.4 (Hair et al., 2017), the three competence items were deleted from further analyses.The remaining 13 items representing brand sincerity, excitement and trustworthiness were reentered for a second EFA.The results are in Table 1.
Table 1 shows that brand sincerity was the most prominent Woolworths' brand personality dimension explaining 28.582% of variance, followed by brand excitement with a variance explained of 20.336% and then trustworthiness with a variance explained of 18.854%.The total variance explained by the three-factor model was 67.772%.This is high for a three-factor model and for a consumer behaviour study (Hair et al., 2014).Even the mean ratings (Table 2) of the three brand personality dimensions showed that sincerity (M 5 4.10) was rated highest, followed by excitement (M 5 3.90) and then trustworthiness (3.87).

Descriptive statistics
The questionnaire responses received were 540, from which 480 were fully completed.The male respondents were 57.5%.Most (90.7%) of the respondents were within the age group of 26-65 years old, 7.5% were young adults within 18-25 years old, while 1.9% were 65 years and above.This indicates that most customers shopping at Woolworths are within the South

Grocery retailer's brand performances
African working ages of 18-65 years old.59.5% of the respondents indicated that they were financially well-off while 40.5% were not so well-off financially.The racial groups were 66% Blacks, 19.2% Whites, 10.4% Coloured's and 4.4% Indians/Asians.These represent the racial composition of South Africa.While most customers in South Africa may frequently shop in two or more grocery retailers, we found that up to 68.5% prefer Woolworths the most.The means of the constructs were calculated to assess respondents' level of agreement to the statements measuring the constructs on a five-point Likert scale.The means in Table 2 are all greater than 3, indicating that respondents agreed to the statements measuring the constructs.

Measurement model
The measurement model tested constructs' reliability and validity and the results are in Table 2 and Table 3.
Table 2 shows that all CA and CR figures exceeded the recommended threshold of 0.70 (Hair et al., 2017) to confirm reliability.Except for an item (PP2) in price whose factor loading was below 0.5 and was deleted from the construct, factor loadings for all other items ranged between 0.586 and 0.925.Table 2 also shows that the AVE figures were all above the recommended 0.5.Both the factor loadings and AVE figures demonstrate convergent validity (Hair et al., 2017).For discriminant validity, Benitez et al. (2020) report that the heterotrait-tomonotrait (HTMT) ratio (HTMT) is more accurate than the Fornell-Larcker criterion.However, Roemer et al. (2021) opine that HTMT is liberal, produces high correlation coefficients since items instead of constructs' average are correlated.Thus, both methods were used to test discriminant validity.The HTMT correlation coefficients (Table 3) are within the recommended threshold of <0.90 (Benitez et al., 2020).Thus, discriminant validity was achieved.This is confirmed by the Fornell-Larcker criterion test, with the square root of all the AVE figures being between 0.749 and 0.853 and exceeding the corresponding correlations (Table 3).
The measurement model was also examined in terms of the model fit.Although Hair et al. (2017) cautions against the use of model fit parameters when PLS-SEM is conducted, Benitez et al.'s (2020) suggest that the model fit can be assessed with SRMR and rms Theta.For a good fit, the recommended value for SRMR should be below 0.08 and rms Theta value should be

Grocery
retailer's brand performances below 0.12.For this study SMMR was 0.065 while rms theta was 0.103, all meeting the good fit thresholds.

Structural model and hypotheses results
The hypotheses testing was computed using a complete bootstrapping with 5000 subsamples at a 0.05 significance level.Following the recommendations of Adabre et al. (2022) who suggest that t-values of 2.54 (p ≤ 0.01), 1.96 (p ≤ 0.05) and 1.65 (p ≤ 0.09) are significant at 99%, 95% and 90% confidence levels respectively, this study accepted significance levels of 90%.
The results are in Table 4. 5.4.1 Direct effects.Hypotheses H1-H5 proposed that Woolworths' brand sincerity, trustworthiness, excitement and competence will positively impact brand loyalty, which will in turn influence brand equity.The results in Table 4 show that out of the three brand personality dimensions retained, brand excitement (H2: β 5 0.159, p < 0.05) and trustworthiness (H4: β 5 0.096, p < 0.1) significantly impacted brand loyalty, which in turn significantly influenced brand equity (H5: β 5 0.402, p < 0.01).Brand sincerity did not make a significant impact on brand loyalty with (H1: β 5 0.064, ns).Hypothesis H3 was not tested because the brand competence construct as aforementioned, was deleted from further analyses.
Hypotheses H6 and H11 were about the mediating effects of brand loyalty.The results are highlighted in the next subsection.

Grocery retailer's brand performances
Hypotheses H14 and H15 tested the mediating effect of brand equity in the relationships between brand loyalty, preference and repurchase intentions.The results in Table 4 show that brand equity significantly mediated the relationships between brand loyalty and preference (β 5 0.527, p < 0.01) and brand loyalty and repurchase intentions (β 5 0.324, p < 0.01).
The Variance Accounted For (VAF) approach was also used to confirm whether the mediation was partial or full.The VAF is calculated from the ratio of the indirect-to-total effect with the formula VAF ¼ ab ða * bÞþc , where ab 5 indirect effect, c is direct effect; a 5 coefficient between independent constructs and mediator (brand loyalty); b 5 coefficient between mediator and outcome construct (brand equity) (Hair et al., 2017).Full mediation is obtained when VAF >0.80, partial mediation happens when VAF is between 0.20 and 0.80 and when VAF is < 0.20, there is no mediation.The VAF figures are in Table 5.
According to the figures in Table 5, the VAF are between 0.20 and 0.80, all indicating partial mediation effects.
5.4.3Explanatory power and predictive accuracy of the model.We assessed the explanatory power with Chin's (2010) guidelines which state that R 2 equal to or greater than 0.67 are regarded as excellent, those above 0.33 are generally good and those below 0.19 are regarded as weak.Table 6 shows the explanatory power of the model to the outcome variables.
Following Chin's (2010) guidelines, Table 6 shows that the integrated model excellently explained brand equity (67.8%), brand preference (71.7%) and had a good explanation of brand loyalty (63.2%) and repurchase intentions (54.2%).In terms of the model predictive accuracy, Q 2 is used and its rule of thumb according to Hair et al. (2014) suggest that the decision parameters for weak predictive models are (0.02</or 5 Q 2 5 0.15), moderate predictive models are (0.15</or 5 Q 2 5 0.35) and for strong predictive models (Q 2> 0.35).Table 6 shows that our model had strong predictive accuracy for all the dependent constructs with Q 2 _predict all >0.35.This study used two theories and integrated three models to examine how much Woolworths' brand personality dimensions and marketing offerings impact brand loyalty, equity and the outcomes of brand equity.The study also assessed the mediating effects of brand loyalty and equity.It starts by examining the customers' perceptions of the grocery retailer's personalities.Out of the three brand personality dimensions studied, we found that Woolworths is perceived to having the sincerity, excitement and trustworthiness personalities.
Studying brand personality as a single dimension, Bairrada et al. (2019) did not find a significant relationship between brand personality and brand loyalty.The current study shows that when brand personality is studied multidimensionally, some dimensions (i.e.brand excitement and trustworthiness and not sincerity) do significantly impact brand loyalty.Sindhu et al.'s (2021) study also show that brand personality dimensions, i.e. brand excitement, sincerity and competence impact brand loyalty but only through brand image created.These are indications that the consumer-brand congruence that brand personality generates in line with the brand congruence theory enhances consumer-brand relationships to the point of brand loyalty as found in the current study; and brand love and brand intimacy as found by Bairrada et al. (2019).Since some and not all dimensions of brand personality influence relationship variables like brand loyalty as found in this study, brand managers should not only build a brand personality recognisable by consumers but should also regularly assess which of the dimensions are producing desirable consumer responses.This will guide appropriate positioning.It is surprising that although Woolworths' customers relatively scored them high in sincerity (M 5 4.01), it was the excitement and trustworthiness of their brand personality that generated brand loyalty.Could there a mediator between brand sincerity and brand loyalty as Sindhu et al. (2021) suggest?
To also gain brand loyalty, this study found that all Woolworths' marketing offerings are significantly contributing as did Anselmsson et al. (2017) and Efanny et al. (2018).However, while Efanny et al. (2018) found a negative relationship between price and loyalty, our study and that of Anselmsson et al. (2017) show that even when price increases, consumers will stay loyal as per the positive relationship between price and brand loyalty.We found a negative relationship between promotion and brand equity, even though significant.This could stem from the fact that during the period when data was collected (i.e.2019-2020), there was a protest against Woolworths and a call to boycott their products for trading with Israel after its capital was moved to Jerusalem.Thus, while brand loyalty was not negatively affected by the negative publicity about Woolworths, the differential response (brand equity) to the

Grocery retailer's brand performances
Woolworths brand could have been negatively impacted by customers embracing other grocery retailer brand that they may not have done ordinarily.Notwithstanding, if the brand loyalty is gained, brand equity will continuously be enjoyed as seen from the strong positive relationship between brand loyalty and equity found in this study.In the UK and Spain, Buil et al. (2013) also found a strong relationship between brand loyalty and equity even for cars and electronics.While Anselmsson et al. (2017) suggest that the loyalty is generated by retailers' marketing offerings, the current study additionally revealed that retailers' brand excitement and trustworthiness also contribute.As did Buil et al. (2013), we found that brand equity strongly drives brand preference and repurchase intentions.

Theoretical contribution
Although Luo et al. (2015, p. 379) worry that co-managing both brand and customer management mindsets are "indisputably a challenge" and difficult for retailers to implement, Han et al. (2021) contend that from an academic and theoretical standpoint it is worthwhile to conceptualize and measure how much retailers' attempts in both mindsets are yielding important outcomes.Their textual analyses of two data sources (i.e.Compustat and Nexi Uni) revealed that the two broad and major focus areas for retailers are the customer and brand strategies.The current study contributes to the fields of retail management, brand management and consumer behaviour by showing how customer and brand focus strategies concomitantly drive various consumer responses.From the brand perspective, the anthropomorphism between personalities of humans and brands has continued to be tested but the extent to which consumers' appreciation of the dimensions leads to various customer behaviours, especially for retailers has been limitedly studied (Bairrada et al., 2019).While this study contributes by exposing the stream of consumer responses that can emanate from brand personality, it also highlights that a retailer's daily marketing offerings reinforces the contributions that brand personality dimensions can make.Additionally, the use of the integrated retailer brand equity model developed in this study produced a high explanatory power of brand loyalty, equity and preference during an economically difficult period (Covid-19 pandemic period) when data was collected.Our integrated model does not only validate Anselmsson et al.'s (2017) and Buil et al.'s (2013) models in a grocery retailer context, it exposes various sources and outcomes of brand equity.The integrated models also equip managers with richer insights on several areas and how brands add value to its owners in terms of the varied consumer responses to the building and marketing of a brand.By testing our model with customers and not students like in most previous studies, our tested model can be used by other retailers to evaluate the benefits of their marketing and brand building efforts.

Managerial contributions
In the competitive retailing environment, our study exposes two routes through which brand loyalty and equity can be achieved.Specifically, it shows that if the marketing mix elements are copied, the brand personality dimensions developed and unique to a retailer cannot be copied and will continue to generate brand loyalty and other consumer-brand relationships.While Das (2014) and Willems (2022) show that brand personality is an important retailer differentiator that drives brand loyalty and equity, the current study reveals that the specific brand personality dimensions making this impact in a grocery retailing context and in an emerging market are brand excitement and trustworthiness.Woolworths' brand equity made strong impacts on brand preference (β 5 0.847) and repurchase intentions (β 5 0.736).From these findings, it can be concluded that the efforts in building and measuring retailer brand equity can be rewarding especially in terms of influencing how customers make their grocery buying choices.

IJRDM 51,13
Choices like brand preferences and repurchase intentions emanating from brand equity can, according to Chang andLiu (2009, p. 1687), enable the purchasing of "the same products or brands or by showing preference toward a particular brand, bringing firms higher market share, higher profits, or share value".All of these inform managers that it pays to build and measure the sources and outcomes of brand equity.With all the elements of the CMM (4Ps) and some elements of BMM (brand excitement and trustworthiness) impacting brand loyalty, a strong driver of brand equity, are indications that both CMM and BMM are important to even grocery retailers.Unlike Sindhu et al. (2021) and Willems (2022) who feel that the brand excitement is reserved for fashion retailers, this study shows that customers are also attracted to a grocery retailer's excitement personality.Thus, understanding these influential sources of brand equity guides further brand-building efforts.

Study limitations and areas of further research
Despite the theoretical and practical contributions of this study, there are some limitations that can be addressed by future studies.The data of this study was collected with crosssectional survey.Even though care was taken to test the constructs' reliability and validity, and Harman's single-factor test (Podsakoff et al., 2003) conducted in this study did not show a problem of common method variance bias, cross-sectional studies may also suffer from social desirability bias.Thus, further research should replicate this study with a longitudinal or experimental design to eliminate the social desirability bias issue.Although Rindfleisch et al.'s (2008) results from a cross-sectional data exhibit validity that is comparable to that obtained from a longitudinal data, Ostroff et al. (2002) found that, compared to a crosssectional study, the degree to which predictors impact outcome variables is lower by 32% when measurements are made longitudinally.For more confident generalisation, future studies should use a larger sample size, considering that this study's sample size was 480.While this sample size was large enough for all the statistical analyses (Kyriazos, 2018), its representation of all Woolworths' grocery customers should be cautiously made.Also, studying only the first top retailer (i.e.Woolworths) and not all first three retailers that thrived during the pandemic is a limitation requiring cautious generalisation to all retailers.Using a convenience sampling method is also a limitation.To overcome this limitation, replications of the study using random sampling and for other types of retailers in other emerging markets are recommended.The replicated study should also examine whether the sincerity will first impact brand image before driving brand loyalty as suggested by Sindhu et al. (2021).

Conclusion
This study integrated elements of Das' (2014) brand personality-loyalty model, Buil et al.'s (2013) CBBE response model, andAnselmsson et al.'s (2017) retailer brand equity model to develop an integrated model for the examination of the sources and outcomes of Woolworths' brand equity.While Woolworths' marketing offerings are contributing to the grocery retailers' performances, its brand personalities also have roles to play.Although the testing of the grocery retailer model produced high explanatory powers, there is room to incorporate other sources and outcomes of retailer brand equity and to test the model against other grocery retailers in South Africa, other emerging markets and even in developed markets.From the empirical testing of the model, practical and theoretical contributions were made.These contributions can be enhanced if the limitations are addressed in further studies.
Figure 1.Study's conceptual model scale, whereby respondents rated how they perceived the traits describing each brand personality dimension.The trustworthiness traits were measured with Anselmsson et al.'s (2017) and Sassaman et al.'s (2019) descriptions.The marketing offerings were measured with Anselmsson et al.'s (2017) (price and perceived quality), Efanny et al.'s (2018) (price and distribution), Jaafar et al.'s (2012) (promotion and perceived quality) scales.Brand loyalty, equity, preference and repurchase intentions were measured with Buil et al.'s (2013) and Anselmsson et al.'s (2017) scales.The items measuring each construct are in appendix.