1PB IMPROVING BRAND LINKAGE EFFECTIVENESS: CUSTOMER RATINGS ABSTRACT In the brand linkage literature, assessing consumers’ brand ratings is an important area. Despite this, there has been comparatively little research executed to analyse the pre and post brand linkage experience of customers from an African perspective. The study offers practical guidance for improving brand linkage effectiveness – and the competitive advantage for brand assessment strategy – in an increasingly complex world. Pretest-posttest control group experiment design was employed to examine the impact of host brand on invited brand and vice versa using a card scoring method. The four groups had an effect on brand linkage ratings, and the partner brands were significant to changes in ratings, confirming brand linkage as a viable strategy for partner brands. Keywords: brand linkage; brand communication; co- branding; brand alliance; brand ratings; marketing commu- nication INTRODUCTION Brand linkage effectiveness can be measured in many ways, but in retail it generally is considered to be the brand assessment achieved for a given brand perception rating. Brand linkage involves the planned alliance or integration of independent brands into a short- or long- term partnership with the aim of influencing customers. Brand linkage effectiveness refers to consumer response to a brand linkage. It has become increasingly popular with brands of various categories, and brand linkage strategies used by marketers are commonly found in both developed and developing countries. The combination of two or more brands into brand linkage or alliance is increasingly common in business markets (Yan & Cao 2017). Although this study uses the term brand linkage, authors define it differently, some label the term as co-branding or brand alliance. Authors differ in “brand alliance classification. Thus, co-branding, co- promotion, co-advertising and similar terms are all part of brand alliances” (Voss et al. 2012:929). Other well-known brand linkages include Taco Bell and Doritos’ popular Doritos Dr Abyshey Nhedzi Department of Strategic Communication, Faculty of Humanities, University of Johannesburg, Johannesburg, South Africa Email: abyshey1@gmail. com ORCID: https://orcid. org/0000-0002-2438- 2686 DOI: https://dx.doi. org/10.18820/24150525/ Comm.v25.5 ISSN 2415-0525 (Online) Communitas 2020 25: 1-20 Date submitted: 24 November 2020 Date accepted: 03 December 2020 Date published: 29 December 2020 © Creative Commons With Attribution (CC-BY) mailto:abyshey1@gmail.com mailto:abyshey1@gmail.com https://orcid.org/0000-0002-2438-2686 https://orcid.org/0000-0002-2438-2686 https://orcid.org/0000-0002-2438-2686 https://dx.doi.org/10.18820/24150525/Comm.v25.5 https://dx.doi.org/10.18820/24150525/Comm.v25.5 https://dx.doi.org/10.18820/24150525/Comm.v25.5 https://creativecommons.org/licenses/by/2.0/za/ https://creativecommons.org/licenses/by/2.0/za/ 32 Nhedzi Locos Tacos, and Quaker Oats and Tyson (Yan & Cao 2017). The marketing paradigm of competitive corporates is different from that of complementary corporates in that the products of one brand lose sales to the other brand rather than benefit from each other’s sales. In this article, the focus is on competitive corporates and it considers a new cooperative to derive optimal strategies for business marketers when they plan to form a brand linkage. Despite research on brand linkage growing rapidly, little is known about how brand linkages are measured and how beneficial they are (Nhedzi et al. 2016). Such research forms a useful conduit between brand linkage theory and practice. However, less research has looked at how brand linkage manifests within African countries, and there is a need to understand the practice better within these countries where there is a growing trend of promoting brands in international set-ups. Since the end of apartheid and the democratic elections, South African businesses have been growing. However, these open opportunities to conduct business with the world markets increased competition with foreign business. On the one hand, it accelerated South Africa’s integration into multinational businesses, which has changed the domestic market isolation, and on the other hand, foreign companies’ entry intensified the competition of the domestic market. A brand goes beyond programmes, services, products and displays; rather, it encompasses everything else about an organisation, including reputation, culture and core values. As such, it is not simply advertising or marketing. Branding has been referred to as a messaging instrument that helps a business reach its goals and it encompasses the promotion of everything associated with the business (Prindle 2011:32). New brand launches are risky endeavours for marketers, as many fail to attract a sustainable customer base. Hence, in order to enter into the larger African market, South African enterprises joined forces with foreign enterprises for positive and effective expansions. Brand linkage is a strategic intent to add a composite of two or more separately owned brands to leverage brand value while targeting the same market. For instance, South African companies formed partnerships with foreign brand such as Simba with Frito-Lay of the United States of America (USA) in the snacking food industry and Robertson’s with Best Foods in the USA (Munzhelele 2014). In addition, Pick n Pay owns a 49% share of a Zimbabwean supermarket business, TM Supermarkets. Marketers use the brand alliance strategy to create brand value for growth, image, lifestyle and popularity in the minds of stakeholders. However, combining two or more brands may change brand meaning in ways that were never intended. Therefore, brands seek to pursue collaboration and joint ventures instead of a single strategy, faced with enormous direct competition. Consequently, the study considered the following two questions: 1) What is the impact of brand linkages on consumers’ ratings of brands? and 2) What is the impact of a brand linkage campaign on consumers’ ratings of brands? The following hypotheses were offered: H1: There is high brand rating of the host brand in brand linkage compared to other competitors H2: A brand linkage campaign has a positive relationship with the host brand 32 Improving brand linkage effectiveness H3: The brand ratings of brand linkage have a positive effect on the host brand H4: The brand ratings of brand linkage have a positive effect on the host brand, more so for experimental than for competitors’ single retailer brands EFFECT OF BRAND LINKAGE Brand linkage can occur in the form of a co-brand. Co-branding refers to any combination of two or more brands in a marketing context, such as new product introduction, co- advertising, co-sponsoring, or joint promotions (Grossman 1997). This implies any pairing of brands in a collaborative marketing effort, including advertisements, services, promotions, public linkages or distribution outlets (Besharat & Langan 2014). Besharat and Langan (2014) further state that co-branding means more than joint ventures, strategic alliance programmes, and value endorsement. In distinguishing co-branding from other brand partnerships, they argue that co-branding differs from other brand partnerships, including brand alliance, ingredient branding, co-promotion/advertising, dual branding, co-sponsorship and cause-related marketing (Besharat & Langan 2014). Brand alliances are defined as a strategic leverage association of brands to improve their images, combining a firm’s brand identity with that of its partner(s) to create a new, stronger competitive position in the market (Besharat & Langan 2014). A brand linkage may exist without new product formation. This is common in retail sector co-branding or alliances. Thus, co-branding is also called brand alliance by Newmeyer et al. (2014). Besharat and Langan (2014) conclude that there is a lack of clarity and consensus on the term co-branding. Early studies classified co-branding into three categories, namely market share, brand extension, and global branding (Park & Srinivasan 1994). The first level is where one company joins another in the form of a merger or acquisition, or otherwise to penetrate the market. The second level is to grow the brand and its product line at times using vertical or horizontal integration, considering the company’s current standing in the market. The third level helps a brand to become a global identity through the unity of two individual brands. With fundamental trends in co-brand research (Lei et al. 2008), partnering with other brands can be risky, as it can have negative consequences (Thomasson & Janusonis 2012; Washburn et al. 2004). According to McKee (2009:3), “if [the customer] is not positive – even if it is the other brand’s fault − it may reflect negatively on a partnering brand”. Hence, previous studies have focused on important factors affecting consumer attitudes towards co- brands (also referred to as brand alliances), such as the familiarity (Levin & Levin 2000), perceived quality (Helmig et al. 2007) and relative brand equity (Washburn et al. 2004) of the partnering brands. Many forms of brand alliances have been documented in marketing literature. However, the literature provides evidence of the existing definition complexity on brand alliances and how the inconsistency among various scholars (see Nhedzi et al. 2016; Anslinger & Jenk 2004; Gammoh & Voss 2011; Laforet 2010; Leuthesser et al. 2003; Simonin & Ruth 1998; Washburn et al. 2004) is difficult to resolve. Some brand linkage efforts 54 Nhedzi take various forms, including ingredient brand alliance, joint venture brand alliance, and multiple sponsorship brand alliances. Acknowledging the wealth of knowledge accumulated on the success of brand alliances, the trend for brand alliances has grown since the increase in internet access from the year 2000 (McCarthy & Von Hoene 2014). Bigger or more successful rivals have acquired many start-ups, as in the case of Overture, which bought AltaVista in 2003 for only 10% of its market value in 2001. A growing concern about the effectiveness of brand alliances permeates existing marketing and advertising theories and practices, as evidenced in McCarthy and Von Hoene (2014: 6); if “market trends are any indication, the use of [co-brand] will only continue to grow and expand between companies and brands”. While these case studies inform practitioners on the importance of brand alliances, they are silent on the strengthening of weak or indifferent ones, and weak brand alliances might be obtained in brands from emerging and developing countries. Without a practical accounting of effectiveness in brand alliances, marketing communication frameworks will remain misleading and incomplete. Fournier and Alvarez (2013: 254) argue, “A contemporary marketing environment laced with distrust, hyper-criticism, and increased consumer power exacerbates [brand] problems and encourages malevolent [consumer] brand relationships. Strong positive brand relationships can easily overnight turn into hateful and vengeful associations.” The intention of brand alliance strategies is primarily to leverage brand quality, which can enhance positive consumer-brand relationships. Most studies have addressed fictitious (unreal) brand linkages or co-brands based on familiarity (Simonin & Ruth 1998). However, real-world examples of brand alliances (noteworthy examples include Engen and Woolworths’ convenient food stores; Pick n Pay and TM) may not fit precisely into such patterns. Singh et al. (2014) found consumer perceptions of the positioning strategies of partner brands to be significant determinants of the positioning perceptions of a co-brand and found some evidence for spillover effects on partner brand positioning. Wason and Charlton (2015) found consumer perceptions of positioning strategies of parent brands to be a significant factor influencing their view of the co-brand. This is similar to other studies as it relies on “a hypothetical pairing of two real brands” (Wason & Charlton 2015:10). The current study offers an important opportunity to re-examine and build an understanding of existing knowledge about brand linkage effects (Nhedzi et al. 2016) in the retail supermarket domain. This explanatory study seeks to examine how a composite brand linkage strategy would add to or detract from the brand value of partnering brands. Challenges of the single organisations or brands Big retail supermarket brands in emerging markets face many challenges: unreliable government policies; shrinking consumer-brand relationships; pricing pressures; increased competition; culture fit; high promotional expenses; and hostile competition from other supermarkets, both national and international. These challenges overlay a changing landscape in the retail supermarket industry characterised by industry consolidation and a youthful population driving market growth; a prevalence towards direct-to-consumer promotion (at least in the African market); open internet access to 54 Improving brand linkage effectiveness affordable products; different suppliers and “all knowing” consumers; and easy access to other developing economies. Retail industry in the African context International retail joint venture or brand linkage conceptualisation has inconsistency, with new terms being introduced often because of a lack of a theoretical base. Not all international retail brand linkages are similar as they might be distinctive in character and yet interrelated in other ways. This is an inevitable configuration because the form is a result of the complex formation and operation of relationships between brand linkages. Sub-Saharan Africa continues to make inroads into the global index ranking (GRI), with three countries (Angola, Botswana and Nigeria) in the top 30 (Kearney 2012). South African brands and retailers have been at the forefront of African expansion but global competition might arrive soon. For South Africa’s major retail players to keep their competitive advantage, sound strategies and investment plans are required (Kearney 2012). Pick n Pay is one of the largest Mass Grocery Retail (MGR) companies in Africa, with a market share of 30% in South Africa. The group operates 794 outlets, comprising hypermarkets, supermarkets and family stores. It employs approximately 55 000 people, according to the 2019 Pick n Pay annual report. Moreover, it operates in seven African countries outside of South Africa (Kearney 2012: 6). The group offers three private label ranges, which are No name brand, Pick n Pay brand and Finest. Conceptual framework The model depicted in Figure 1 shows four key constructs: brand linkages, brand communication, brand ratings, and redefinition of brand relationships. Clearly, brand linkage research details how consumers’ quality perceptions (Rao et al. 1999; Voss & Gammoh, 2004; Wason & Charlton 2015) and general product evaluations or attitudes (Helmig et al. 2007; Simonin & Ruth 1998) spill over from parent brands to alliances, and vice versa. 76 Nhedzi FIGURE 1: BRAND LINKAGE ASSESSMENT MODEL Source: Nhedzi et al. 2016: 61 Brand linkage In this setting, two independent firms (brand A and B) team up and form a brand linkage or alliance to cooperate. When the two firms form a brand linkage, they act together to maximise their joint profits and names. The brand linkage is a cooperation that aims at a synergy where each firm’s benefits from the alliance will be greater than what it could achieve from individual effort, but each firm remains an independent organisation. Brands constantly seek ways to distinguish themselves in today’s highly competitive market. The brand linkage is of particular interest because many firms enter an agreement to benefit from each other’s resources, capabilities, and core competencies (Nhedzi et al. 2016). The following are some of the opportunities available to partnering brands (Fournier & Alvarez 2013; McCarthy & Von Hoene 2014): ♦ Accelerate growth ♦ Access critical capabilities ♦ Enter new markets ♦ Reduce costs or capacity ♦ Accelerate rediscovering of products or services ♦ Build a critical skills base In recent years, there has been rapid growth in brands cooperating with other brands across industries. It has become more difficult for companies to differentiate their 76 Improving brand linkage effectiveness products because of the increased number of competitors and product offerings. Companies face a new challenge of how to preserve or improve their brand popularity (Jevons et al. 2002; Laforet 2010). According to Besharat and Langan (2014), these challenges include the high cost of establishing new brands, which suffer failure rates of between 80 and 90 percent, the expense of investments required to enter new markets, and the intense competition in today’s dynamic marketplace. To meet these challenges, companies have adopted innovative strategies, such as creating brand extensions, joint promotions, cooperative advertising and forming strategic alliances (Jevons et al. 2002; Lanseng & Olsen 2012). A variety of terms has been used interchangeably from 1985 to 2004 for international retailing brand linkages. The terms include cooperative arrangements, contractual joint ventures, cross- shareholdings links, collaboration, cooperative linkages, equity joint venture, franchise joint ventures, interfirm linkages, conglomerate joint ventures, business partners and interfirm alliances (Palmer & Owens 2006). Brand communication For the purpose of this study, TM and Pick n Pay’s brand linkage advertisement campaign was used as stimulus to test customers’ ratings of TM. Brand ratings Brand ratings refer to consumers’ scoring cards of six brands. This is the main method for brand linkage assessment. The brand linkage is of particular interest because many brands enter an alliance to benefit from each other’s resources, capabilities, and core competencies. A non-verbal measuring instrument, known as card scoring, represents the monetary value allocation of consumers’ preferences for supermarket brands (Nhedzi et al. 2016). This scale of measurement had 99 cards at the disposal of the participant. Brand relationship In this setting, two independent firms (brand A and B) team up and form a brand relationship to develop a linked brand product, instead of developing their own products. When the two brands form a brand linkage to develop the cobranded product, they act together to maximise their joint profits. The brand alliance is a cooperation that aims for a synergy where each firm’s benefits from the alliance will be greater than what it could achieve from individual efforts, but each brand remains an independent organisation. Jevons et al.’s (2002) brand-relationship-interaction matrix offers a rare classification of brand relationships. In practice, many well-known brands, including some familiar brands that are treated as single brands, engage in direct or indirect alliances with other brands in the form of strong or weak, or distant or close partnerships. This is justifiable, because no single brand can satisfy the numerous needs of stakeholders; hence, some brands exploit multiple brand lines by formal or informal cooperation or linkages to leverage their brands. To the best of the author’s knowledge, no study has investigated the impact of brand linkage and brand ratings on retail supermarket brands across Africa. 98 Nhedzi METHODOLOGY Sample The aim of this study was to investigate the impact of brand linkages on consumers’ ratings of African retail supermarket brands. In the study, the experiment was conducted in Harare and Masvingo in Zimbabwe. The city of Harare was chosen for having a greater projection in the growth of such services in Zimbabwe and because it is the nation’s capital. The majority of the sample had adult customers as the target sample. Usually, they have more money, their lifestyle is typical of the consumption situation of most consumers, and they have more experience buying most consumer products. These similarities may resemble their perceptions, motivations, and preferences in the real world. This study aimed to establish whether either brand significantly added value to or detracted value from the other; also, to establish spontaneous awareness of brand linkages. A true experiment design with card scoring was conducted to test the propositions posited by this study. The experimental method is used in marketing communication research as a primary research method for establishing causation, but the card scoring method is rarely applied for empirical research. The researcher examined the impact of brand linkage using experimenting between May 2018 and March 2019. An industry expert with more than 20 years’ experience in advertising marketing evaluated the initial version of the experiment. Operationalisation of measures The experiment had a single factor (brand linkage) between subjects’ design. Nhedzi et al. (2016) used an experiment method to test the effect of brank linkage on brand relationships. Similarly, a pretest-posttest control group design was used to evaluate whether there were significant differences in brand ratings between the (experimental) groups, who were aware and unaware of the linkage, and the (control) groups, who were aware and unaware of the linkage. The participants were randomly divided into four groups and each participant received 99 cards. At pretest, the respondents were given envelopes labelled with the names of six major supermarket brands and a set of 99 cards. They were asked to allocate these cards to each brand in accordance with their general assessment of the brands by placing cards in each envelope. The process of sorting the cards into six supermarkets named envelopes was repeated at posttest. The four groups: (1) 28,6% (N=143) aware experiment (AE) group, in which the participants who mentioned TM being linked to Pick n Pay and shown the advertisement; (2) 29% (N=145) aware control (AC) group, in which the participants were aware of the brand linkage and then no campaign was shown; (3) 21,6% (N=108) unaware experiment (UE) group, in which the participants did not mention TM being linked to Pick n Pay and were shown the brand linkage campaign; and 98 Improving brand linkage effectiveness (4) 20,8% (N=104) an unaware control (UC) group, in which the participants were not shown any campaign. The study ensured sufficient statistical power; a size of 30 to 40 participants per experimental condition seemed to be an adequate size, with a total number of 500 respondents (Geuens & De Pelsmacker 2017; Van Voorhis & Morgan 2007). It was “big enough” to find effects of scientific relevance to be statistically significant. However, the population of interest was heterogeneous in terms of age, level of education, and profession, and all these were relevant characteristics in the context of the study, then randomly assigning individuals to small experimental subgroups necessarily led to subgroup equivalence in terms of several of these characteristics. More males (51,6%) than females (48,4%) were part of this study. In the age range, 63,6% were 21 to 30 years, 15,8% 31 to 40 years, 14,4% 18 to 20 years, 4,8% 41 to 50 years, and 1,4% were older than 50 years. In the racial and ethnic make-up, 98,4% were black African, 0,8% Coloured, 0,6% white, and 0,2% Indian or Asian. The employment make-up was as follows: 47,6% were employed (self-employed or working for a company), 33,4% were unemployed, and 19% were students. As described earlier, the researcher had two points of measurement at different times in the study. After the first step of pretest single blinding non-disclosure, the respondents in all conditions were asked to indicate their favourite brand through the card-scoring method. Blinding helped in preventing biased ascertainment of outcomes and reduced the chance of co-interventions. This was an exploratory, quantitative study based on a Pretest-Posttest Control Group Experimental Design (Campbell & Stanley 1963; 1966; Nhedzi et al. 2016). This study sought to replicate and extend the brand linkage study; thus, it used a similar design. However, there are two distinct differences between the method used in Nhedzi et al.’s (2016) study and this one. The current study did not use two local brands. Of the brands used for this analysis, one is local (present in one country), the TM brand, and the other, the Pick n Pay brand, is an international brand. Experiment – main effects This experiment aimed to demonstrate the direct effects of brand linkage and brand perception through brand rating and brand assessment (Hypotheses 1 and 2). The author introduced an existing brand as an experimental stimulus, rather than a fictitious one, to generate the reality knowledge of the brand. Single blinding or masking was employed to all respondents to control for possible confounding based on prior knowledge of the brand. In a single-blind trial, only the participants were unaware of which intervention they received. Supermarkets were selected as the product category because they have been used widely as a target product manipulation in the branding literature. The results for six supermarket brands are presented in Figure 2. The total number of votes allocated was 49 500. If these votes had been randomly distributed (that is, no differentiation between brands), each brand would have received 8 250 votes. TM’s score of 13 590 was therefore 65% higher than a random score. The 1110 Nhedzi average score for TM, per respondent, was 27,2 votes (13 590/500) compared to an average random score, per respondent, of 16,5 votes.   4221 13590 4508 14134 5096 7944 3912 15835 4093 13595 4801 7264 0 2000 4000 6000 8000 10000 12000 14000 16000 18000 Food World TM Mahomed Mussa OK Zimbabwe Choppies Spar Series1 Series2 FIGURE 2: RESULTS FOR SIX SUPERMARKET BRANDS Then, the respondents were asked whether they were aware of these brands being associated in any way with any other brand or brands. Responses were recorded in a short questionnaire. There were 288 respondents, who were spontaneously aware of the TM/Pick n Pay brand linkage. These “aware” respondents allocated 8 840 votes to TM, which on average rated TM 30,5 votes, compared to a random allocation score of 16,5 votes, and the TM average of 27,2 votes. The spontaneously aware results showed that the respondents of this linkage were more likely to rate TM’s salience much higher than average, except for OK Zimbabwe. Therefore, the clinical results gave a clear indication that the brand linkage itself (after exposure to the linked brand campaign) resulted in noticeably higher ratings of TM as a linked brand by those who were aware of the linkage. DATA ANALYSIS AND RESULTS Statistical procedures used to test the hypotheses included correlation analysis, linear regression analysis, and analysis of variance (ANOVA). Data analysis consisted of validity and reliability assessments, and covariate analysis using Statistical Package of Social Science (SPSS 26). An alpha level of p = ,05 was required for significance in all statistical procedures (Pallant 2013). Whilst generalisation may be limited, certain meaningful results may be constructed from the findings, which can direct continued research in this complex arena. 1110 Improving brand linkage effectiveness The researcher conducted an experiment with a high degree of internal validity, which reduced the potential influence of extraneous variables to such an extent that the independent variable is the most likely cause of the observed change in the dependent variable. The study focused on seven classes of extraneous variables that could undermine the internal validity of an experiment design – history, maturation, instrumentation, selection of subjects, testing, statistical regression and experimental mortality (see Table 1). TABLE 1: INTERNAL VALIDITY THREATS Threat How it was dealt with History No unanticipated event occurred while the experiment was in progress that affected the dependent variable Maturation Manifested equally in experimental and control groups as participants in both groups changed (“mature”) at the same rate due to random assignment Instrumentation The use of a fixed measuring instrument (card sorting) ensured that this was not a problem Selection Random assignment, all participants had an equal chance of being in treatment or comparison groups, and the groups were equivalent Testing Both groups were exposed to the pretest and so the difference between the groups was not due to testing Statistical regression If this was a problem, it would have manifested equally in the experimental and control groups due to randomisation Experimental mortality The same number of participants made it through the entire study in both the experimental and comparison groups Source: Bordens & Abbott (2014: 326); Campbell & Stanley (1963: 7-9) The study was a randomised controlled trial (RCT), which is a true experiment, which was considered for this study (Campbell & Stanley 1963). The threat of maturation and testing should be equally manifested between the experimental and the control group; statistical regression, mortality, and selection interaction threats are protected by the random assignment of participants, occurring probably equally across the two groups. Randomisation is known to ensure control of extraneous variables by equally distributing them among the groups (Campbell & Stanley 1963, 1966: 16), whereas parallel grouping provided and minimised control over relevant stimulus factors, because all four groups were exposed to a similar experimental process (Kirk 2013). The pretest would sensitise both the control group and the experimental group to the posttest in a like manner; therefore, presenting no internal threat to validity. This platform of prior exposure was typically used as a screen for asking questions that are 1312 Nhedzi more detailed about advertising effectiveness, such as message take-out or likeability (Dubow 1994). This was to avoid the potentially confounding effects of previous exposure or experience with existing brands (Schneider et al. 2015). According to Geuens and De Pelsmacker (2017: 86), “Another solution is to use a control group that is not exposed to any of the experimental stimuli”. Existing brand names (TM and Pick n Pay) had to be used for purposes of comparing the effects of existing brand linkage versus single brands to depict a real-world setting. In this study, existing brands were carefully selected based on their suitability for the research objective at hand. Differences between the four groups after intervention To determine whether any between-group differences were found at the end of the study due to the intervention of participants across groups, inter-group comparisons were firstly conducted to determine whether there were significant differences between the four groups’ posttest scores following exposure to the intervention. TABLE 2: DIFFERENCES ACROSS FOUR GROUPS (POSTTEST) N Mean Std. deviation Std. error Min. Max. F Sig. Post_TM Aware, Intervention (campaign) 143 39,50 20,271 1,695 0 94 Aware, No Intervention 145 30,54 17,162 1,425 0 73 15,950 ,000 Unaware, Intervention (campaign) 108 30,52 18,265 1,758 0 78 Unaware, No Intervention 104 23,66 16,189 1,587 0 72 Total 500 31,67 18,952 ,848 0 94 Significance: (p<.05); Not significant: (p>.05); Confidence level: 95% Post hoc tests revealed statistically significant differences across the four groups and how respondents rated TM (aware intervention, n =143: aware no intervention, n = 145: unaware intervention, n = 108: unaware no intervention, n = 104), x4 (4, n = 500) = 15,95, p = ,000 after the intervention) (see Table 2). The aware intervention group recorded a higher mean score (M =39,50) than the other three groups, which recorded mean values of 30,54, 30,52 and 23,66. 1312 Improving brand linkage effectiveness TABLE 3: MULTIPLE COMPARISONS − POST HOC TEST RESULTS (I) Group (J) Group Mean Difference (I-J) Std. Error Sig. 95% Confidence Interval Lower Bound Upper Bound Aware Experimental Group Aware Control Group 8,959* 2,139 ,000 3,44 14,47 Unaware Experimental Group 8,985* 2,314 ,001 3,02 14,95 Unaware Control Group 15,840* 2,339 ,000 9,81 21,87 Aware Control Group Aware Experimental Group -8,959* 2,139 ,000 -14,47 -3,44 Unaware Experimental Group ,026 2,307 1,000 -5,92 5,97 Unaware Control Group 6,881* 2,333 ,017 ,87 12,89 Unaware Experimental Group Aware Experimental Group -8,985* 2,314 ,001 -14,95 -3,02 Aware Control Group -,026 2,307 1,000 -5,97 5,92 Unaware Control Group 6,855* 2,494 ,031 ,43 13,28 Unaware Control Group Aware Experimental Group -15,840* 2,339 ,000 -21,87 -9,81 Aware Control Group -6,881* 2,333 ,017 -12,89 -,87 Unaware Experimental Group -6,855* 2,494 ,031 -13,28 -,43 *. The mean difference is significant at the 0,05 level. Post hoc tests shown in Table 3 reveal that the aware intervention (M = 39,50, SD =20,27) is significantly different from the aware no intervention (M = 30,54, SD = 17,16; p = 000), compared to unaware intervention (M = 30,52, SD = 18,27; p = ,001) and to unaware no intervention (M = 23,66, SD = 16,19; p = ,000). Also, the aware no intervention (M = 30,54, SD = 17,16) was significantly different from the unaware no intervention (M = 23,66, SD = 16,19; p = ,017). Similarly, there was also statistically significant difference in mean scores between unaware intervention and unaware no intervention (p = ,031). However, there was no statistically significant difference in mean scores between aware no intervention and unaware no intervention (p = 1,000). 1514 Nhedzi Differences between the four groups over time (intra-group comparisons) It was also necessary to investigate each of the four groups separately and compare for significant differences between the pre-intervention and post-intervention ratings of TM for each group. TABLE 4: DIFFERENCE BETWEEN PRE-INTERVENTION RATING AND POST-INTERVENTION RATING FOR EACH GROUP ON TM Mean N Std. Deviation Std. Error Mean t df Sig. Aware, Intervention (campaign) Pre_TM 32,76 143 19,988 1,672 -4,473 142 ,000 Post_TM 39,50 143 20,271 1,695 Aware, No Intervention Pre_TM 28,50 145 18,100 1,503 -2,129 144 ,035 Post_TM 30,54 145 17,162 1,425 Unaware, Intervention (campaign) Pre_TM 22,58 108 17,049 1,641 -4,998 107 ,000 Post_TM 30,52 108 18,265 1,758 Unaware, No Intervention Pre_TM 22,43 104 15,527 1,523 -1,224 103 ,224 Post_TM 23,66 104 16,189 1,587 In Table 4, a paired sample t-test was conducted to compare the number of scores on TM for each group’s pretest and posttest. Firstly, there was a significant difference in the scores for TM aware intervention pretest (M = 32,76, SD = 19,99) and aware intervention posttest (M = 39,50, SD = 20,27) conditions, t(142) = -4,473, p = ,000. Secondly, there was a significant difference in the scores for TM aware no intervention pretest (M = 28,50, SD = 18,10) and aware no intervention posttest (M = 30,54, SD = 17,16) conditions, t(144) = -2,129, p = ,035. Thirdly, there was a significant difference in the scores for TM unaware intervention pretest (M = 22,58, SD = 17,05) and TM unaware intervention posttest (M = 30,52, SD = 18,27) conditions, t(107) = -4,998, p = ,000. Finally, the results showed no statistically significant difference between pretest unaware no intervention (M = 22,43, SD = 15,53) and posttest unaware no intervention (M = 23,66, SD = 16,19) conditions, t(103) = -1,224, p = ,224. However, this was expected, as Group 4 was not given an intervention condition. There were significant differences among the three groups before and after intervention on TM scores. Overall, the results confirmed that a brand linkage campaign was strongly 1514 Improving brand linkage effectiveness associated with the brand linkage perception of TM in a positive manner. The results are summarised in Table 4. DISCUSSION In sum, the findings from the study on real-life brands confirmed the proposed impact of brand linkages on consumers’ ratings of brands. The impact of a brand linkage campaign on consumers’ ratings of brands highlighted that the two different brands had unique positive effects on brand perceptions, in support of Hypotheses 1 and 2. Consistent with brand linkage assessment and the brand ratings, the results confirmed that brand linkage advertisement exposure through experiment produced higher TM brand perceptions; whereas retail brands advertisement exposure through experiment produced higher brand-rating perceptions. The findings of this research produced implications for both brand linkage theory and practice, with more practical implications for marketing managers in retail brand marketing. In its totality, this research adds clarity and support for a brand linkage rating in retail brand marketing. The theoretical significance of this research is the extension of the brand linkage assessment method, the card sorting to retail brand marketing. This research answers the call in the literature to provide empirical evidence of consumer responses to brand linkage in marketing. Furthermore, this research adds to the growing body of empirical research on brand assessment in retail brand marketing in the literature. Brand linkage assessment is predicted theoretically and proven empirically (Nhedzi et al. 2016), but few studies have examined brand linkage types (an exception is Gammoh & Voss 2011; Jevons et al. 2002). In this cross-African retail market study, there is further clear evidence of the effect of brand linkage; albeit, one brand (OK Zimbabwe) scored highest in pretest. The brand linkage campaign effect was demonstrated in several different ways in this study. This corroborates previous work showing that brand linkage can significantly enhance brand recognition, brand attitude, brand image, and the perceived quality of partnership brand (Ma et al. 2018). The practical significance of this research is found in its confirmation of the fact that “famous brand helps firms improve the assessment of an unknown brand and improves the perception of quality” (Ma et al. 2018: 23). To marketers, the choice of brand linkage becomes a strategic one, with implications for several consumer responses, which leads to the adoption of a brand linkage strategy to improve the perceived value of their own brands and market position, and to obtain a competitive advantage. First, in support of H1-4, the findings suggest that the benefits to marketers of a brand linkage is the spill-over effects of brand linkage that enables alliance partners to achieve these goals with more favourable changes in attitude toward the host brand. Given experiment exposure to these retail brand linkage ads, it would seem as if the host brand is more successful at creating a positive perception with the brand linkage. The present study demonstrated that brand linkage shapes the positive ratings and consumer perceptions of brands (Ma et al. 2018). It also supports the previous studies highlighting that normally firms, in building brand linkages, achieve more than they can 1716 Nhedzi on their own (Yan & Cao 2017). The analyses revealed significant differences among the manipulation groups concerning the variables of brand linkage perceptions. Many of the limitations in this study can be addressed with future research. First, previous knowledge of the brand linkage and ad campaign may influence respondents’ answers. The use of an existing brand linkage (TM-Pick n Pay) in the stimuli could provide an explanation for the full mediation effect of ad attitude between group condition and brand attitude. In this experimental study on brand message effects, people were directly exposed to one version of an ad, which is in one dimension. In turn, future work could also expand on the findings by including more stimuli – testing a variety of brand categories and different types of characters. Future research could expand on this conjecture to determine whether brand and ad campaign familiarity plays a role in affecting consumer responses to the brand linkage ad execution styles examined in this study; and, a comparison of all three brand awareness measures, namely top of mind, aided and spontaneous or recall. The collected data allowed the author to conclude that South African retail units have been accepted relatively easily by desperate consumers under challenging economic conditions, gaining their trust, sympathy and awareness. It will be interesting to determine how consumers’ perceptions will change when the number of stores of each retailer increases and other competitors penetrate the market. These limitations aside, this research holds potential for future research on brand linkage in supermarket marketing. This research shows the power of brand linkage in the age of complexity and an overload of consumer marketing, where two brands are better than one brand, which is expected to create more organic interactions with customers well into the future. CONCLUSION This study provided an opportunity to examine the impact of a brand linkage on brand ratings in a new context, the retail supermarket industry. The study can provide useful insights on the behaviour and decisions of consumers in real-life cobrand situations. The research has benefits for academic advancement of theory, methodology and managerial insights. Ultimately, the experiment in this study has a practical mission: to deliver objective, actionable, research-based insights that improve brand linkage assessment. Firstly, the originality of the work consists in connecting retail giant supermarket brand alliances from emerging (South Africa) and developing (Zimbabwe) countries, using a brand assessment methodology. While these important post-modern aspects of brand alliance, such as co-branding, have been analysed previously, they have not been investigated from an African perspective. Secondly, it will enrich the strategic brand alliance literature, adding new contexts in which no empirical evidence exists. To the researcher’s knowledge, brand linkage strategies have never been studied in the retail supermarket sector, apart from brands from developed countries. Moreover, the study contributes to the test-retest reliability of a brand assessment scale of an experimental study by Nhedzi et al. (2016) and card scoring scale methodology. 1716 Improving brand linkage effectiveness The retail space in Africa has changed rapidly due to the complex macro-environmental complexities that retailers have to deal with when conducting business. Until recently, when Walmart entered the market in 2012, internationalisation of retailing in Africa had not been driven by huge global retailers, but rather by South African actors. This includes retailers such as Massmart, Metcash Trading Africa, Pick n Pay, Shoprite, and Spar, which have established operations in other African countries (Dakora 2012). Pick n Pay has 56 supermarkets in Zimbabwe, 40 trading as TM, and 16 trading as Pick n Pay. Zimbabwe provides an example of a country that has witnessed a transformation of the supermarket space, which was dominated by the actors mentioned previously. These actors have been subjected to a changing economic and political landscape. The country has gone through key phases in its economic and political cycles, which have transformed the retail space, in particular the demise of the supermarket brands. This empirical research will advance marketing communication knowledge by expla- natory consumer perceptions and brand ratings of co-brands. It will also answer a call for further understanding of consumers’ attitude towards the partnering brands from an African perspective. Moreover, this unique brand rating study will offer definition in the context of retail supermarket brands. The pre- and post-evaluation of constituent brands and consumer learning about brand linkage was the impetus for this study on brand partnerships. REFERENCES Anslinger, P. & Jenk, J. 2004. Creating successful alliances. Journal of Business Strategy 25(2): 18-22. https://doi.org/10.1108/02756660410525362 Besharat, A. & Langan, R. 2014. Towards the formation of consensus in the domain of co-branding: Current findings and future priorities. Journal of Brand Management 21(2): 112-132. https://doi.org/10.1057/bm.2013.25 Bordens, K.S. & Abbott, B.B. 2014. Research design and methods: A process approach. (Ninth edition). New York: McGraw-Hill Education. Campbell, D.T. & Stanley, J.C. 1963. Experimental and quasi-experimental designs for research on teaching. In: Gage, N.L. (ed.). Handbook of research on teaching. Chicago, IL: Rand McNally. Campbell, D.T. & Stanley, J.C. 1966. Experimental and quasi-experimental designs for research. Boston: Houghton Mifflin. Dakora, E.A.N. 2012. Exploring the fourth wave of supermarket evolution: Concepts of value and complexity in Africa. International Journal of Managing Value and Supply Chains 3(3): 25-37. https://doi.org/10.5121/ijmvsc.2012.3303 Dubow, J.S. 1994. Point of view: Recall revisited: Recall redux. Journal of Advertising Research 34(3): 92-106. Fournier, S. & Alvarez, C. 2013. Research dialogue relating badly to brands. Journal of Consumer Psychology 23(2): 253-264. https://doi.org/10.1016/j.jcps.2013.01.004 https://doi.org/10.1108/02756660410525362 https://doi.org/10.1057/bm.2013.25 https://doi.org/10.5121/ijmvsc.2012.3303 https://doi.org/10.1016/j.jcps.2013.01.004 1918 Nhedzi Gammoh, B.S. & Voss, K.E. 2011. Brand alliance research: In search of a new perspective and directions for future research. Journal of Marketing Development and Competitiveness 5(3): 81-93. Geuens, M. & De Pelsmacker, P. 2017. Planning and conducting experimental advertising research and questionnaire design. Journal of Advertising 46(1): 83-100. https://doi. org/10.1080/00913367.2016.1225233 Grossman, R.P. 1997. Co-branding in advertising. Journal of Product Brand Management 6(3): 191-201. https://doi.org/10.1108/10610429710175709 Helmig, B., Huber, J. & Leeflang, P. 2007. Explaining behavioural intentions toward co- branded products. Journal of Marketing Management 23(3-4): 285-304. https://doi. org/10.1362/026725707X196387 Jevons, C., Gabbott, M. & De Chernatony, L. 2002. A taxonomy of brand linkages: The brand-relationship-interaction (BRI) matrix. White paper. University of Birmingham, Birmingham Business School. Kearney, A.T. 2012. The retail industry on the rise in South Africa. Gauteng Provincial Treasury Quarterly Bulletin (April-June). Kirk, R.E. 2013. Experimental design: Procedures for the behavioral sciences (Fourth edition). Thousand Oaks, CA: Sage. https://doi.org/10.4135/9781483384733 Laforet, S. 2010. Managing brands: A contemporary perspective. London: McGraw-Hill Education. Lanseng, B. & Oslen, L. 2012. The attitudinal response to alternative brand growth strategies. European Journal of Marketing 46(1/2): 177-191. https://doi. org/10.1108/03090561211189293 Lei, J., De Ruyter, K. & Wetzels, M. 2008. Consumer responses to vertical service line extensions. Journal of Retailing 84(3): 268-280. https://doi.org/10.1016/j. jretai.2008.05.001 Leuthesser, L., Kohli, C. & Suri, R. 2003. 2+2=5? A framework for using co-branding to leverage a brand. Brand Management 11(1): 35-47. https://doi.org/10.1057/ palgrave.bm.2540146 Levin, I.P. & Levin, A.M. 2000. Modeling the role of brand alliances in the assimilation of product evaluations. Journal of Consumer Psychology 9(1): 43-52. https://doi. org/10.1207/s15327663jcp0901_4 Ma, B., Cheng, F., Bu, J. & Jiang, J. 2018. Effects of brand alliance on brand equity. Journal of Contemporary Marketing Science 1(1): 22-33. https://doi.org/10.1108/ JCMARS-08-2018-0007 McCarthy, K. & Von Hoene, S. 2014. Co-branding: A sweet business strategy. Westlaw Journal. Computer and Internet 31(18): 3-6. McKee, S. 2009. The pros and cons of cobranding. [Online]. Available at: http://www. businessweek.com/smallbiz/content/jul2009/sb20090710_255169.htm [Accessed on 19 August 2020]. Munzhelele, M. 2014. An employee’s perspective of co-branding separation on brand commitment. Johannesburg: University of the Witwatersrand. https://doi.org/10.1080/00913367.2016.1225233 https://doi.org/10.1080/00913367.2016.1225233 https://doi.org/10.1108/10610429710175709 https://doi.org/10.1362/026725707X196387 https://doi.org/10.1362/026725707X196387 https://doi.org/10.4135/9781483384733 https://doi.org/10.1108/03090561211189293 https://doi.org/10.1108/03090561211189293 https://doi.org/10.1016/j.jretai.2008.05.001 https://doi.org/10.1016/j.jretai.2008.05.001 https://doi.org/10.1057/palgrave.bm.2540146 https://doi.org/10.1057/palgrave.bm.2540146 https://doi.org/10.1207/s15327663jcp0901_4 https://doi.org/10.1207/s15327663jcp0901_4 https://doi.org/10.1108/JCMARS-08-2018-0007 https://doi.org/10.1108/JCMARS-08-2018-0007 http://www.businessweek.com/smallbiz/content/jul2009/sb20090710_255169.htm http://www.businessweek.com/smallbiz/content/jul2009/sb20090710_255169.htm 1918 Improving brand linkage effectiveness Newmeyer, C.E., Venkatesh, R. & Chatterjee, R. 2014. Cobranding arrangements and partner selection: A conceptual framework and managerial guidelines. Journal of the Academic Marketing Science 42(2): 103-118. https://doi.org/10.1007/s11747- 013-0343-8 Nhedzi, A., Overton-de Klerk, N. & Bester, T. 2016. Brand linkages: Winners, losers and how to measure these. Communitas 21: 59-72. https://doi.org/10.18820/24150525/ Comm.v21.5 Pallant, J. 2013. A step by step guide to data analysis using the SPSS program: SPSS survival manual (Fifth edition). Maidenhead, Berkshire: Open University Press. Palmer, M. & Owens, M. 2006. New directions for international retail joint venture. The International Review of Retail, Distribution and Consumer Research 16(2): 159- 179. https://doi.org/10.1080/09593960600572183 Park, C.S. & Srinivasan, V. 1994. A survey-based method for measuring and understanding brand equity and its extendibility. Journal of Marketing Research 31(2): 271-288. https://doi.org/10.1177/002224379403100210 Prindle, R. 2011. A public relations role in brand messaging. International Journal of Business and Social Science 2(18): 32-36. Rao, A.R., Lu, Q. & Ruekert, R.W. 1999. Signalling unobservable product quality through a brand ally. Journal of Marketing Research 36(2): 258-268. https://doi. org/10.1177/002224379903600209 Schneider, L., Systems, B. & Cornwell, T.B. 2015. Cashing in on crashes via brand placement in computer games: The effects of experience and flow on memory. International Journal of Advertising 24(3): 321-343. https://doi.org/10.1080/02650 487.2005.11072928 Simonin, B.L. & Ruth, J.A. 1998. Is a company known by the company it keeps? Assessing the spill-over effects of brand alliances on consumer brand attitudes. Journal of Marketing Research 35(1): 30-42. https://doi.org/10.1177/002224379803500105 Singh, J., Kalafatis, S. P. & Ledden, L. 2014. Consumer perceptions of cobrands: The role of brand positioning strategies. Marketing Intelligence & Planning 32(1): 145-159. https://doi.org/10.1108/MIP-03-2013-0055 Thomasson, O. & Janusonis, A. 2012. Managing a demerger process: A case study of a corporate divorce. [Online]. Available at: https://gupea.ub.gu.se/ bitstream/2077/29688/1/gupea_2077_29688_1.pdf [Accessed on 19 August 2020]. Van Voorhis, C.R. & Morgan, B.L. 2007. Understanding power and rules of thumb for determining sample sizes. Tutorials in Quantitative Methods for Psychology 3(2): 43-50. https://doi.org/10.20982/tqmp.03.2.p043 Voss, K.E. & Gammoh, B.S. 2004. Building brands through brand alliances: Does a second ally help? Marketing Letters 15(2/3): 147-159. https://doi.org/10.1023/ B:MARK.0000047390.01552.a2 Voss, K.E., Gammoh, B.S. & Fang, X. 2012. How does a brand ally affect consumer evaluations of a focal brand? Psychology and Marketing 29(12): 929-940. https:// doi.org/10.1002/mar.20575 https://doi.org/10.1007/s11747-013-0343-8 https://doi.org/10.1007/s11747-013-0343-8 https://doi.org/10.18820/24150525/Comm.v21.5 https://doi.org/10.18820/24150525/Comm.v21.5 https://doi.org/10.1080/09593960600572183 https://doi.org/10.1177/002224379403100210 https://doi.org/10.1177/002224379903600209 https://doi.org/10.1177/002224379903600209 https://doi.org/10.1080/02650487.2005.11072928 https://doi.org/10.1080/02650487.2005.11072928 https://doi.org/10.1177/002224379803500105 https://doi.org/10.1108/MIP-03-2013-0055 https://gupea.ub.gu.se/bitstream/2077/29688/1/gupea_2077_29688_1.pdf https://gupea.ub.gu.se/bitstream/2077/29688/1/gupea_2077_29688_1.pdf https://doi.org/10.20982/tqmp.03.2.p043 https://doi.org/10.1023/B:MARK.0000047390.01552.a2 https://doi.org/10.1023/B:MARK.0000047390.01552.a2 https://doi.org/10.1002/mar.20575 https://doi.org/10.1002/mar.20575 PB20 Nhedzi Washburn, J.H., Till, B.D. & Priluck, R. 2004. Brand alliance and customer-based brand equity effects. Psychology and Marketing 21(7): 487-508. https://doi.org/10.1002/ mar.20016 Wason, H. & Charlton, N. 2015. How positioning strategies affect co-branding outcomes. Cogent Business & Management 2(1): 1-12. https://doi.org/10.1080/23311975.201 5.1092192 Yan, R. & Cao, Z. 2017. Is brand alliance always beneficial to firms? Journal of Retailing and Consumer Services 34(1): 193-200. https://doi.org/10.1016/j.jretconser.2016.10.009 https://doi.org/10.1002/mar.20016 https://doi.org/10.1002/mar.20016 https://doi.org/10.1080/23311975.2015.1092192 https://doi.org/10.1080/23311975.2015.1092192 https://doi.org/10.1016/j.jretconser.2016.10.009 _Hlk40491191 _Hlk40491211 _Hlk40491235 _Hlk40491286 _Hlk40491378 _Hlk40491418 _Hlk40491442 _Hlk40491546 _Hlk40491560 _Hlk40491751 _Hlk40491793 _Hlk40491874 _Hlk40491899 _Hlk40491920 _Hlk40491944 _Hlk33734625 _Hlk34825201 _Hlk40492137 _Hlk40492161 _Hlk40492181 _Hlk40492210 _Hlk40492228 _Hlk37904357 _Hlk37969289 _Hlk37970658 _Hlk38668209 _Hlk40492319 _GoBack