Global Brand Identity and Local Social Meaning in Share a Coke in Australia and China

Abstract

This study examines how Coca-Cola adapted the language, personalization, and social invitation of Share a Coke between its Australian launch in 2011 and its Chinese Nickname Bottle campaign in 2013. It combines comparative qualitative content analysis with multimodal discourse analysis. The purposive corpus contains 40 traceable campaign artifacts, 20 from each market, drawn from official Coca-Cola materials, agency and award records, contemporary advertising trade coverage, and contemporaneous Chinese business reporting. Twelve artifacts with surviving images or screenshots form a close visual subsample. The analysis finds that Coca-Cola standardized a campaign grammar rather than a fixed set of words. The red-and-white palette, Coke wordmark or bottle form, replaceable label position, sharing imperative, and route from package to participation remained stable. The content of the replaceable slot changed substantially. Ten of the twelve Australian package artifacts used conventional first names, while nine of the twelve Chinese package artifacts used persona or dialect labels and three used relationship terms. Across the full corpus, personal names were the primary strategy in 13 Australian items; persona or dialect labels were primary in 13 Chinese items. The Chinese phrase “share this bottle of Coca-Cola with your…” also made attribution and relationship more explicit than the Australian first-name formula. Both campaigns connected packaging to participation, but through different media systems: Facebook, MMS, radio, a digital sign, and shopping-centre kiosks in Australia; Sina and Tencent Weibo, Renren, Baidu, influencer seeding, and a custom-label vending machine in China. The comparison shows glocalization operating through small creative and infrastructural decisions. Coca-Cola preserved recognition while changing the social identity that personalization performed.

Keywords: Share a Coke, glocalization, global advertising, personalization, brand identity, China, Australia, multimodal discourse analysis

Introduction

Share a Coke is built around a conspicuous act of subtraction. The campaign removes the corporate name from the most prominent place on a familiar package and gives that space to someone else. In the original Australian rollout, the substitute was usually a common first name. The bottle became searchable: a shopper could look for Emma, Tom, Kate, or Josh, then keep the bottle, give it away, photograph it, or circulate a digital version. The product still had to read immediately as Coca-Cola even when the script in the center of the label no longer said Coca-Cola. That tension makes the campaign a useful object for studying global communication. It asks how much a global brand can change at the point of recognition without losing the identity that makes the change valuable.

Coca-Cola launched the idea in Australia in 2011 under the internal name Project Connect. The initial packaging used 150 popular names, supported by television, radio, Facebook tools, live personalization, and an interactive digital sign. Coca-Cola later reported that more than 250 million named bottles and cans were sold during the Australian summer in a country of fewer than 23 million people. The company and its agencies then developed a toolkit for other markets, and the campaign expanded internationally (The Coca-Cola Company, 2014). Its movement across borders was not a process of literal reproduction. Local teams had to decide which identities could occupy the label, how the invitation to share should be phrased, and which media practices could turn a package into a social act.

China makes that problem visible. When Coca-Cola introduced the campaign there in 2013, it did not center the mass-market package line on a list of common personal names. Contemporary campaign accounts explain that the number and variety of Chinese names made the Australian coverage model difficult to reproduce. The Chinese team instead selected popular online nicknames, relationship terms, and social types such as 文艺青年 (wényì qīngnián, “artsy” or “literary youth”), 吃货 (chīhuò, “foodie”), 闺蜜 (guīmì, “bestie”), and 技术男 (jìshù nán, “tech guy”). The phrase above the label was structured as “share this bottle of Coca-Cola with your…,” making the term a way to describe or address another person as well as a possible self-description (Li, 2013; Zhou, 2013). The campaign also linked the bottle to the Chinese platform environment through Weibo, Renren, Baidu search, influencer seeding, and a campaign microsite (Lau, 2013).

This comparison is often reduced to a familiar cultural contrast: Australia receives individual names, while China receives collective or relational identities. That summary is too blunt. The Australian campaign was socially oriented from the beginning; the words “Share a Coke with” turned an individual name into a prompt to give, reconnect, and publicly address. Chinese labels included relationship terms, but many were also highly individualized persona labels drawn from youth and internet culture. Some labels could be flattering, teasing, ironic, or uncomfortable depending on who applied them. First-name personalization and nickname personalization therefore do not map neatly onto individualism and collectivism. They organize identity differently.

The study asks how Coca-Cola adapted the language, personalization, and social messaging of Share a Coke in Australia and China while maintaining a recognizable global brand identity. It contributes an integrated examination of package language, campaign visuals, and participatory media. Previous work has addressed personalized labels, cross-cultural branding, language choice, brand co-creation, and multimodal advertising, but few studies place the early Australian and Chinese Share a Coke systems side by side. The comparison matters because the campaign's global unit was never only the bottle. It was a repeatable relationship among a recognizable package, a replaceable identity term, an invitation to give or share, and a set of media tools through which consumers could extend the message.

Campaign and Comparative Context

Share a Coke in Australia

The Australian campaign emerged from a problem of familiarity. Coca-Cola's internal account states that young Australians recognized the brand but did not feel that it spoke to them “at eye level.” The 2011 brief asked agencies to make the brand newly noticeable during the summer and to strengthen its connection with young adults. Ogilvy Sydney's solution retained the package's red field and brand cues while replacing the central wordmark on one face with a personal name. Trademark concerns prevented the names from being printed in the exact Spencerian Coca-Cola script, so the team developed a related “You” typeface. That decision is analytically important. The name had to appear close enough to the brand's visual language to feel native to the package while remaining legally and formally distinct (The Coca-Cola Company, 2014).

The launch set contained 150 names. Coca-Cola estimated that those names directly covered only 42 percent of the population, even after generic terms such as Mum, Dad, and Mate were considered. The campaign therefore framed the package as a gift rather than a promise that every shopper would find a personal match. “Share a Coke with” changed absence into a social prompt: a bottle with somebody else's name could still be relevant because it could be given to that person. Later phases addressed demand for missing names through shopping-centre kiosks and a public vote for 50 additions (Blight, 2011; The Coca-Cola Company, 2014).

The Australian launch worked across media from the outset. Television spots used volunteered photographs to build mosaics of people who shared a name. Austereo broadcast 150 name songs. Facebook tools allowed users to make a virtual named can for a friend and assemble a campaign-style video from profile photographs. At Sydney's King's Cross, users could send a name by SMS, see it appear on the Coca-Cola sign, and receive an MMS image to redistribute. Eighteen Westfield shopping centres hosted kiosks that printed requested names on cans. The execution linked search, gifting, display, customization, and circulation rather than treating the package as a finished advertisement (Marketing Magazine, 2012).

The available outcome claims come primarily from Coca-Cola, its agencies, and award submissions, so they should be read as campaign-reported results rather than independent causal estimates. Those sources reported a 7 percent rise in young-adult consumption during the campaign, 18.3 million media impressions, an 870 percent increase in traffic to the Australian Coke Facebook page, 76,000 virtual cans shared, and roughly 378,000 personalized cans printed at Westfield locations (Interactive Advertising Bureau Australia, 2012; Marketing Magazine, 2012). These numbers establish the scale of participation claimed by the campaign team. They do not isolate Share a Coke from other sales or media influences.

International Expansion

The campaign moved from Australia to New Zealand and then into Asian and European markets. Coca-Cola's retrospective account describes the company building a toolkit after international teams expressed interest following the campaign's 2012 Cannes recognition. The toolkit preserved a simple invitation while allowing markets to change names, wording, media, and activations. By September 2014, Coca-Cola reported that the idea had reached more than 70 countries; a 2015 company release put the number above 80 (The Coca-Cola Company, 2014, 2015).

This expansion illustrates why the standardization-adaptation debate is inadequate when treated as a binary. The campaign remained recognizable because several elements traveled together: red packaging, a Coke or Coca-Cola mark, a personalized white type treatment, an invitation to share, and a route from purchase to social circulation. Yet the personalized term could change from a first name to a nickname, relationship, lyric, or other locally useful address. The platform architecture changed as well. A Facebook tool or shopping-centre kiosk was not universally transferable without considering local media use, technical infrastructure, and audience habits.

Share a Coke in China

Coca-Cola announced the Chinese “Nickname Bottle” or 快乐昵称瓶 campaign in late May 2013. National television, outdoor, retail, and print support followed in June. Contemporary sources disagree about the total number of standard labels. Adquan reported 24 designs near launch; Campaign Asia described more than 50 nicknames, and Little Black Book reported more than 60. The discrepancy may reflect different dates, bottle sizes, custom variants, or a distinction between retail and promotional labels. Because the surviving record does not resolve it, this study does not assert a definitive total. It includes only labels and executions that can be traced to a contemporary source (Adquan, 2013; Campaign Asia, 2013; Little Black Book, 2013).

The selection process was presented as research-led. Campaign China reported that social-listening analysis generated roughly 300 high-frequency terms, evaluated for volume, interaction, posting activity, and positive brand fit before a smaller set was chosen. The surviving labels combine several kinds of address: social personas such as 文艺青年 (“artsy youth”); tastes or habits such as 吃货 (“foodie”); gendered or status-bearing types such as 纯爷们 (“real man”); relationship terms such as 闺蜜 (“bestie”) and 室友 (“roommate”); and limited dialect expressions such as the Beijing term 大拿 (“expert” or “big shot”) (Campaign China, 2013; Li, 2013).

This was more than a substitution of Chinese characters for English names. The label phrase read 分享这瓶可口可乐与你的 X, literally “share this bottle of Coca-Cola with your X.” The possessive 你的 (“your”) positioned the nickname as an attribution within a relationship: your roommate, your bestie, your tech guy, your artsy youth. Some terms could also be claimed by the buyer, but the grammar made giving and playful labeling especially available. Yicai's contemporaneous report also recorded a tension in the strategy. Several labels, including 小清新, 白富美, and 文艺青年, were not neutral identity categories; speakers often used them teasingly or rejected them with a response equivalent to “you're the one who is.” The campaign borrowed vernacular whose meaning depended on tone and relationship (Li, 2013).

The rollout began on Sina Weibo with seeded packages sent to celebrities, online personalities, media workers, and selected fans. Actor Huang Xiaoming received a personalized small bottle paired with 大咖 (“big shot”) and joked publicly that a less flattering label might suit him better. The official account then revealed the mass-market packages under the hashtag #可口可乐昵称瓶#. Other elements included a four-day repost contest, nickname descriptions and sharing tools on a campaign site, Baidu search placement linking to the television creative, two commercials featuring the band Mayday, broader outdoor and retail support, and a “Happy Vending Machine” that printed a requested name (Lau, 2013; Li, 2013; Little Black Book, 2013; Sina, 2013).

Campaign China reported more than 50,000 online expressions of purchase intent during the first month, based on social listening across Sina Weibo, Tencent Weibo, and Renren. It also reported two Mayday television spots and a national outdoor, retail, and print rollout (Campaign China, 2013). As with the Australian metrics, this figure comes from campaign reporting and measures detected statements, not completed transactions. It is useful as evidence of the campaign's monitored social response but cannot establish sales effects.

Value of the Comparison

Australia and China are not proxies for two internally uniform cultures. They are bounded campaign settings separated by launch date, language, media infrastructure, and archive quality. Australia provides the originating template and the most complete surviving case documentation. China provides an early adaptation in which the identity term, syntax, and platform system were visibly reworked. The cases make it possible to identify the campaign elements that Coca-Cola treated as nonnegotiable and those it allowed local teams to reconstruct.

The comparison also separates three levels that are often collapsed. The brand level includes Coca-Cola's colors, bottle, wordmarks, and global sharing idea. The campaign level includes the replaceable label slot and the invitation to act on it. The market level includes names, nicknames, dialects, platform choices, celebrity seeding, and interaction norms. Glocalization appears in the coordination of these levels, not in a simple percentage of standardized and localized content.

Literature Review

Global Advertising Between Standardization and Adaptation

International advertising research has long asked whether companies should standardize campaigns across markets or adapt them to local conditions. Standardization promises scale, coordination, and a consistent image. Adaptation promises linguistic fit, cultural relevance, and responsiveness to local competition and media. Theodosiou and Leonidou's (2003) integrative review found that decades of research produced mixed results partly because studies defined and measured adaptation differently. Later reviews continued to argue that the decision is contingent rather than absolute: product category, market similarity, brand strategy, regulation, media structure, and organizational capacity can all change the appropriate degree of adaptation.

The binary can also hide variation within a campaign. A company may standardize its visual identity while adapting its language, or preserve a slogan while redesigning the participation mechanism. Taylor (2010) argued that the standardization question remained useful only when research moved beyond asking whether an advertisement was the same and examined which elements were standardized, under what conditions, and with what consequences. Share a Coke requires that element-level approach. Its core depends on changing a prominent piece of packaging while keeping the campaign globally recognizable.

Alden, Steenkamp, and Batra (1999) introduced global consumer culture positioning to describe advertising that associates a brand with symbols understood as transnational. Coca-Cola's red, script, bottle form, and long advertising history give it substantial global symbolic resources. Yet a global cue does not enter a market without local interpretation. A name, pronoun, or kinship term can reorganize the relationship offered by an otherwise stable visual system. Global brand identity is therefore a resource for adaptation, not merely something adaptation threatens.

Glocalization

Robertson (1995) used glocalization to describe the simultaneous presence of universalizing and particularizing processes. The concept rejects the assumption that global circulation simply erases local difference. Global forms are interpreted, used, and reconstructed in particular places. Maynard and Tian's (2004) content analysis of Chinese websites operated by leading global brands provides an advertising-specific example. Many companies retained global brand structures while changing language, imagery, institutional references, and other content for Chinese audiences. Their findings make the global and local analytically inseparable.

Glocalization is more precise than “localization” for Share a Coke because the local version depends on the recognizability of the global object. A generic red bottle bearing 文艺青年 would not produce the same meaning. The value of the nickname comes partly from its temporary occupation of Coca-Cola's branded space. Likewise, the Coca-Cola package becomes newly noticeable because local language interrupts the expected corporate name. The global form and local term create one another's salience.

Glocalization should not become a celebratory label applied whenever a company uses local words. Adaptation can flatten, stereotype, or commercially appropriate cultural practices. The Chinese nickname labels turned shifting internet vernacular into a finite product set. Their success as advertising does not prove that they represented Chinese identity comprehensively or neutrally. The concept is most useful here as a description of how scales interact, followed by close analysis of what the interaction includes and excludes.

Personalization and Consumer Identity

Personalization makes an industrially identical object appear selective. McDarby et al. (2018) tested personalized beverage labels with 404 children aged 8 to 13. Children were more likely to choose a beverage when it carried their own name, including when the personalized option was an unhealthy drink. The study does not establish how adults in Australia or China interpreted Share a Coke, but it demonstrates that a name on a package can affect choice rather than functioning only as decoration. It also raises an ethical concern: personalization can make a sugar-sweetened product more attractive by attaching identity to it.

Identity effects are not limited to self-recognition. Voyer, Kastanakis, and Rhode (2017) describe brand and stakeholder identities as reciprocally co-created. Brands offer symbolic material through which people can present themselves; consumers and other stakeholders then alter brand meaning through their use, interpretation, and circulation of that material. Share a Coke literalizes this exchange. Coca-Cola supplies the label system, but a shopper decides whether “Emma,” “Mate,” “Foodie,” or “Roommate” describes the self, another person, or a relationship worth performing publicly.

The model also helps distinguish personalization from individualization. A first name points to a person but is chosen from a mass list and may be shared by thousands. A persona label is categorical but can support a highly specific performance when used as a joke between friends. A custom-printed bottle is more individualized in production, while a public vote makes the list collectively editable. These forms distribute agency differently. Treating them all as “personalization” without separating their mechanisms obscures the communication work they perform.

Čulig Suknaić (2025) analyzes Share a Coke posters through embodiment and multimodal personification. In the English-language example, named bottles held in a toast connect verbal invitations, bodies, and the package. In the Croatian example, the localized phrase shifts toward sharing summer with a group while retaining the bottle-to-person mapping. Although the study uses only two posters and does not examine Australia and China as systems, it supports a central premise of this project: the personalized word gains meaning through its placement, typography, object, image, and implied action.

Culture Without National Caricature

Cross-cultural advertising research often uses individualism and collectivism to anticipate differences in self-concept and appeals. De Mooij and Hofstede (2010) connect national cultural dimensions to communication styles, brand roles, and advertising strategy. Their framework is useful as a source of questions: Does a message foreground uniqueness, family, peers, status, or harmony? It becomes less reliable when national scores are treated as direct causes of a particular advertisement.

Oyserman, Coon, and Kemmelmeier's (2002) meta-analysis found that commonly repeated East-West contrasts were often smaller, more variable, or more context-dependent than broad summaries suggested. Csaba (2017) similarly cautions that cross-cultural brand co-creation should treat culture as dynamic, contested, and not cleanly bounded by geography. This study therefore uses cultural dimensions as sensitizing concepts rather than coding outcomes. It does not infer a cultural value from the country label. Interpretations must be anchored in campaign language, imagery, platform use, and contemporary accounts of the creative decision.

That caution changes the reading of Share a Coke. Australian first names do not make the campaign purely individualistic because the syntax directs each name toward another person. Chinese nickname labels do not make the campaign uniformly collectivist because many terms are persona categories tied to taste, style, status, and gendered self-presentation. The more defensible comparison concerns how each campaign made identity socially usable.

Language Naming and Address

Language in international advertising does more than transmit a stable message. Krishna and Ahluwalia (2008) found that language choice among bilingual consumers could activate different associations for multinational brands. Their experiments concerned Hindi and English in urban India, not Mandarin and English, so the specific associations cannot be transferred to this study. The broader finding remains relevant: language positions the brand and the social relationship from which it speaks.

The Australian formula “Share a Coke with [first name]” uses a proper noun as an index. It directs attention toward a particular person while leaving the reason for sharing open. The Chinese formula uses a possessive and a category: “share this bottle of Coca-Cola with your [roommate/foodie/artsy youth].” The term can attribute taste, status, personality, gender, or relationship. Translation into a compact English noun often loses the pragmatic flexibility of the source. 吃货 can function like “foodie,” but it carries a playful self-mocking quality. 文艺青年 can praise cultural taste, tease pretension, or mark an aesthetic type. 闺蜜 names an intimate female friendship more specifically than the broad English word “friend.” These are not interchangeable tags.

The dialect bottles add another scale of localization. Yicai reported limited labels using Beijing 大拿 and Xiamen 少年家 (Li, 2013). Even within a campaign described as the Chinese version, the language record contains regional variation. The surviving sources do not establish the distribution or reception of these variants, so they should not support broad claims. They do show that “local” was not a single national language choice.

Participation and Brand Circulation

Share a Coke was designed to move. A bottle could be searched for, bought, gifted, photographed, customized, voted on, or reproduced as a digital object. Liu and Lopez (2016) matched Nielsen soft-drink sales data with Facebook, Twitter, and YouTube conversation and found that social-media discussion affected consumers' valuation of brand characteristics and product choice. Their model does not prove that the Share a Coke executions caused the reported campaign results. It supports treating conversation and redistribution as part of the marketing system rather than as noise surrounding it.

Participation also complicates control. Coca-Cola defined the template and filtered possible submissions, but consumers chose names, recipients, captions, and contexts. The Australian King's Cross activation reportedly required a block list of more than 5,000 terms because public submissions could turn the brand's invitation toward abuse or profanity (The Coca-Cola Company, 2014). Chinese persona labels carried meanings that the company could select but not stabilize. Co-creation describes both the value and risk of giving consumers a structured opening in the brand.

Multimodal Communication

Kress and van Leeuwen (2006) argue that visual meaning is organized through composition, salience, framing, gaze, distance, and relations among represented elements. Machin and Mayr (2012) extend multimodal analysis to the ideological and social work performed by language and image. These approaches matter because the central substitution in Share a Coke is not only verbal. The word occupies the place, scale, color, and typographic role normally associated with Coca-Cola. A name in an ordinary caption would not do the same work.

Multimodal analysis also directs attention to hands, faces, condensation, bottle groupings, platform interfaces, and settings. A bottle photographed alone can act as a personalized collectible. Two bottles touching can enact a relationship. A sports-field image can attach terms such as roommate, buddy, and rival to a scene of peer activity. A search-result placement can turn the bottle into a clickable route. The modes jointly establish who is addressed and what kind of action seems natural.

Research Gap

Existing scholarship explains important parts of Share a Coke: personalized labels can influence selection; identity can be co-created; language choice changes a multinational brand's associations; global campaigns combine standardization and adaptation; and meaning emerges through several modes. What remains less developed is an artifact-level comparison of how the originating Australian system changed in its early Chinese adaptation. Much public commentary cites the shift from names to nicknames, but that phrase alone does not show how label grammar, social categories, visual continuity, and platform participation worked together. This study addresses that gap with a traceable comparative corpus, a defined codebook, and a close reading restricted to materials that remain directly inspectable.

Theoretical Framework

The analysis uses three connected tools: glocalization, reciprocal identity co-creation, and multimodal discourse analysis. They operate at different levels and answer different parts of the research question.

Glocalization organizes the comparison between circulation and reconstruction. The analysis begins by identifying elements that travel across markets and elements that change. It does not assume that the stable elements are global while every change is local. Some apparently global features, such as the invitation to share, become meaningful only through local grammar and media practices. Some local elements, such as digital personalization, were later folded back into the wider international campaign. The framework therefore tracks relations rather than assigning each feature to a sealed scale.

Reciprocal identity co-creation explains the replaceable label as an opening between managerial control and consumer use. Coca-Cola determines the package, eligible terms, type treatment, and participation tools. Consumers determine which term is relevant, who receives it, whether it is sincere or ironic, and how it circulates. The framework prevents the analysis from describing participation as unrestricted. Both campaigns used filters, curated lists, and branded interfaces. Co-creation occurred inside a strongly designed system.

Multimodal discourse analysis examines how wording, type, color, objects, people, space, and platform interface work together. The close reading focuses on salience, composition, represented social relations, and the action offered to the viewer. It asks where personalization appears, what remains branded, who is positioned as giver or recipient, and how the artifact routes attention toward another action. The approach is interpretive. It can identify meaning potentials in the artifact but cannot establish how every viewer understood it.

National cultural dimensions are not used as a predictive theory. They enter only as questions about independence, interdependence, hierarchy, or group address when the artifact provides evidence. This decision responds to the variability identified by Oyserman et al. (2002) and the warning from Csaba (2017) that culture cannot be reduced to stable national traits. The analysis concerns campaign constructions of identity, not fixed characteristics of Australians or Chinese people.

Research Questions

The study addresses one central question:

RQ: How did Coca-Cola adapt the language, personalization, and social invitation of Share a Coke between Australia's 2011 launch and China's 2013 Nickname Bottle campaign while maintaining a recognizable global brand identity?

Three subquestions guide the comparison:

RQ1: Which verbal, visual, and participatory elements remained consistent across the two campaigns?

RQ2: What forms of identity and relationship did the personalized label construct in each market?

RQ3: How did each campaign connect the physical package to its local media and participation system?

Method

Research Design

The study combines comparative qualitative content analysis with multimodal discourse analysis. Content analysis supplies a systematic record of artifact type, personalization strategy, language, recipient orientation, and participation. Multimodal analysis provides a closer account of how selected images and interfaces produce meaning. The methods are complementary. The full corpus identifies recurring patterns, while the visual subsample tests how those patterns operate inside individual compositions.

The design is descriptive and interpretive. It does not measure campaign effects on sales, attitudes, or cultural values. Reported campaign metrics appear only as historical context and are attributed to their source. The study's findings concern the content and organization of accessible campaign materials.

Case Selection

Australia was selected because it was the originating market and established the campaign's first-name structure, visual template, and cross-platform participation model. China was selected because it was an early international adaptation that replaced the retail first-name list with nicknames and relationship terms and connected the campaign to a different writing system and platform environment. The cases offer a high-information comparison: one defines the initial campaign grammar, and the other changes the content and use of its most salient variable element.

Artifact Collection

Collection began with the sources listed in the original proposal and expanded through searches of official Coca-Cola pages, agency portfolios, advertising award records, contemporary trade publications, and Chinese business and advertising coverage. Search terms combined the campaign name, market, year, agency, medium, and Chinese titles including 昵称瓶 and 快乐昵称瓶. When an original campaign page was no longer available, a contemporary article could verify an artifact if it identified the execution, market, and launch period clearly.

The archive is asymmetric. Australia's campaign is preserved in a detailed Coca-Cola retrospective, an IAB Australia award sheet, an extended Marketing Magazine case, and visible campaign images. China's original campaign site and many 2013 social posts are no longer publicly retrievable. The Chinese record depends more heavily on contemporary Campaign China, Yicai, China Quality News, ClickZ, Adquan, Sina, and agency-trade coverage. This difference was recorded rather than concealed.

Inclusion and Exclusion Criteria

An item was included when it met four conditions: it was commissioned or distributed by Coca-Cola or a documented campaign agency; it was clearly associated with the 2011 Australian or 2013 mainland Chinese rollout; its defining verbal content or audience action could be identified; and a stable source URL or archive record remained available. Distinct label texts were counted as separate artifacts because the identity term is the campaign's central variable. A platform feature or activation was counted separately when it enabled a different action, such as voting, customizing, reposting, or projecting a name.

The study excluded later relaunches, including the 2025 campaign; fan-made packages; duplicate reposts; materials whose market could not be established; later Chinese Lyric Bottle and film-quotation campaigns; and claims about executions that could not be traced beyond an unsourced marketing summary. The two reportedly separate Mayday television commercials were treated as one campaign execution because the accessible sources document their existence but do not distinguish their content reliably enough for separate coding.

Sample

The final corpus contains 40 artifacts, evenly divided between Australia and China. Each country sample contains 12 package variants and eight cross-media executions. The same total does not imply equal archive completeness; it provides a structured comparison while retaining only items that met the inclusion rules.

Twelve items, six per market, formed the visual subsample. The Australian items were the Emma package, Mate can, Kylie and Jason television mosaic, Liam television mosaic, King's Cross name projection, and Westfield personalization kiosk. The Chinese items were the 文艺青年 package, 闺蜜 package, 室友 package, the sports-field hero creative, the Baidu search placement, and the campaign microsite. These items had surviving images or screenshots that could be inspected directly.

Unit of Analysis

The unit of analysis is one discrete campaign execution. For packages, the identity term defines the unit. For other media, the audience action or composition defines it. A gallery containing many labels is one digital execution because the interface, not each repeated bottle image, is the object being coded. A package label that also appears in the gallery remains a separate unit because it circulated as a physical product.

Coding Procedure and Codebook

The codebook recorded country, year, medium, execution, language, visible text, primary personalization strategy, recipient orientation, digital participation, visual availability, source, and translation notes. Following Neuendorf's (2017) emphasis on explicit variables and coding instructions, the primary strategy categories were defined as mutually exclusive. “Personal name” referred to a conventional given name selected for mass distribution or a name-centered broadcast execution. “Relational term” named a tie such as Mate, Bestie, 闺蜜, or 室友. “Persona label” attributed a taste, status, style, or social type such as 吃货 or 文艺青年. “Dialect persona” marked a documented regional expression. “Custom name” required user-specific printing or submission. “Crowdsourced expansion” referred to changing the mass list through public voting. “Social amplification” referred to a contest whose primary action was reposting rather than selecting an identity term.

Recipient orientation was coded when wording or documented use directed the product toward another person or relationship. Digital participation required an explicit digital action: posting, reposting, voting, submitting, creating, or redistributing. Merely appearing on a digital page did not qualify. Visual items were additionally coded for the red-and-white palette, Coca-Cola/Coke wordmark or contour cue, prominence of the personalized slot, representation of a human relationship, and participation cue. Appendix A provides the full definitions and borderline rules.

Translation Procedure

Chinese text was retained in its original script, followed by pinyin and a conservative English gloss where useful. Translation emphasized the function of a term in the artifact rather than claiming a perfect equivalent. For example, 吃货 is glossed as “foodie” with a note that its tone can be playful and self-mocking. 文艺青年 is glossed as “artsy” or “literary youth” rather than reduced to “hipster.” 闺蜜 is rendered as “bestie” while noting its conventional reference to an intimate female friendship.

No independent bilingual translator completed a second verification, so the study does not claim bilingual intercoder validation. Meanings were checked against the surrounding Chinese reporting and English-language campaign coverage. The least secure dialect item, 少年家, is translated conservatively as “young person,” with the source's identification of it as Xiamen usage retained. Translation uncertainty is included in Appendix D.

Multimodal Analysis

The close analysis examined verbal salience, typography, color, composition, represented people and objects, gaze where visible, spatial grouping, setting, and platform context. Each artifact was read first on its own terms and then against the matching market pattern. Claims about a visual feature were restricted to the twelve-item visual subsample. Text-only records informed content findings but not image composition.

Comparative Analysis

Analysis proceeded in three stages. First, artifacts were grouped within each market by medium and primary strategy. Second, the package subset and digital-participation subset were compared across markets. Third, recurring continuities and adaptations were tested against the visual subsample. Counts summarize the purposive corpus and are not estimates of the full campaign population.

Reliability Validity and Consistency

One researcher completed the coding. The proposed second coder was not available, so no Cohen's kappa or other intercoder statistic is reported. After the initial pass, every row was checked again against the operational definitions and source record. The mutually exclusive primary-strategy variable was recalculated from the final dataset. Visual claims were separated from textually documented claims, and campaign-reported performance metrics were not treated as independent findings.

Construct validity was supported by defining personalization categories before comparison and retaining the source language. Source triangulation was used when possible. For example, the Australian kiosk appears in the Coca-Cola retrospective and Marketing Magazine case; the Chinese nickname strategy appears across Campaign China, Yicai, China Quality News, ClickZ, and agency reporting. Disagreement among sources, including the number of Chinese labels, is reported rather than resolved by assumption.

Ethics

The study uses publicly available corporate, agency, trade, and news materials. It did not recruit participants or collect private data. Individual social-media users were not sampled, and no nonpublic posts were reproduced. Public figures appear only when a documented seeding execution is necessary to explain campaign strategy. Copyrighted advertisements are described and linked in the dataset rather than reproduced in the paper.

Methodological Limits Built Into the Design

The sample is purposive and archive-dependent. It favors executions that left a durable record and cannot represent every label, post, retail display, or audience response. The equal country totals aid comparison but do not mean the archives are equally complete. The analysis is strongest when describing visible campaign structure and weakest when inferring reception. These limits are addressed again after the discussion.

Results

The Sample Preserved the Campaign Structure Across Different Media

Both market samples extended beyond packaging. Twelve package artifacts in each market were joined by eight cross-media executions. Five items in each country required explicit digital participation under the codebook definition. The equal count should not be read as evidence that digital participation had equal scale. It shows that both campaign systems made a deliberate route from the physical package into digital action.

The platform mix differed. Australia's digitally participatory items included the King's Cross submission and MMS return, Facebook virtual cans, Facebook user-made television spots, the Facebook vote, and Facebook-linked name songs. China's included the official Weibo reveal, Weibo repost contest, Baidu placement, campaign microsite sharing tools, and influencer seeding designed for social posting. Both campaigns used live customization, through Westfield kiosks in Australia and a Happy Vending Machine in China.

Personalization Changed From Proper Names to Social Classification

The strongest difference concerns the content of the personalized slot. Personal names were the primary strategy in 13 of 20 Australian artifacts. Persona or dialect labels were primary in 13 of 20 Chinese artifacts. No conventional personal name served as the focal strategy in the Chinese mass-market package subset, although custom names were available through the vending machine and personalized seeding.

The package subset sharpens the contrast. Ten of twelve Australian packages used first names and two used relationships, Mate and Bestie. Nine of twelve Chinese packages used persona or dialect labels, while three used relationships: 闺蜜 (“bestie”), 室友 (“roommate”), and 神对手 (“formidable rival”). The Chinese mass-market package did not become less personal. It shifted from identifying a particular person through a proper noun to identifying a socially recognizable type or tie.

The categories also offered different kinds of fit. A first name either matched or did not match a shopper or intended recipient. A persona label required interpretation. A person could accept “foodie,” assign it to a friend, use it ironically, or reject it. Relationship labels required a dyad: somebody is a roommate, bestie, or rival in relation to somebody else. The Chinese labels therefore made classification and attribution part of the campaign action.

Recipient Orientation Remained a Global Constant

All 40 artifacts were coded as recipient-oriented because each was organized by the invitation to share with, label, nominate, or create something for another person. This does not mean every consumer used the product as a gift. It means the campaign's grammar positioned another person as available.

In Australia, the preposition “with” linked an individual name to an open social act. In China, the possessive 你的 (“your”) made the relationship more explicit: share this bottle with your X. Even persona labels were grammatically routed through someone else's perspective. A “tech guy” or “artsy youth” appeared as the addressee selected by the giver. The continuity lies in recipient orientation; the adaptation lies in how the recipient was named.

Visual Brand Cues Remained Stable in the Directly Inspectable Items

All twelve items in the visual subsample retained the Coca-Cola red-and-white palette and a Coke/Coca-Cola mark, contour bottle, or both. All foregrounded a replaceable personal term or the package containing it. These results are limited to the selected surviving visuals, but the pattern is consistent across package, television, outdoor, search, retail, and website contexts.

The brand did not disappear when the wordmark was displaced. It moved to the cap, reverse label, corner lockup, bottle form, red field, or surrounding interface. The personalized term received the central position while the brand supplied recognition from the edges and from the object itself. The visual system created enough redundancy that Coca-Cola could surrender its most obvious word without surrendering the package.

One Australian item, the Liam face mosaic, was coded without an explicit relationship cue because its still centers a single name and face, though it belonged to a sharing campaign. Every Chinese visual item contained a relational cue through the label grammar, paired terms, people, or platform sharing functions.

Participation Was Curated Rather Than Open

Both campaigns invited audiences to contribute while retaining strong controls. Australia's name list was centrally selected, user submissions were filtered, and the second wave was produced from a managed vote. China's terms were selected through social listening and internal review for frequency, interaction, positivity, and brand fit. The Weibo contest offered a choice among approved nicknames. The custom machines widened access but still operated within a branded interface and printable-content rules.

This pattern supports a constrained model of co-creation. Consumers could choose, assign, vote, customize, and circulate. They could not redesign the visual system or guarantee that a requested term entered mass production. The campaign's effectiveness as a participatory object depended on the opening being noticeable and usable while the boundaries remained corporate.

The Chinese Localization Worked at Several Scales

Chinese adaptation occurred at the national language, youth-vernacular, relationship, regional, celebrity, and platform levels. Simplified Chinese made the package locally readable, but the campaign did not stop at translation. Internet terms positioned Coca-Cola inside current youth discourse. Relationship labels mapped the bottle onto peer ties. Dialect variants suggested regional play, though the archive is too limited to measure their scope. Celebrity seeding attached a label to a public persona and invited reinterpretation. Weibo, Renren, and Baidu made the package circulate through services used in the market.

These layers are evidence against treating localization as a single switch. The Chinese campaign was not localized once. Different executions localized different components, sometimes with unresolved tensions. A term selected as positive could still be used sarcastically. A nationally distributed campaign could include regionally specific words. A mass persona system could coexist with custom personal names.

Multimodal Analysis

The Australian Name Package

The Emma package shows the campaign's central visual transfer. “Share a Coke with” appears in small white letters above Emma, while the first name occupies the position and scale normally associated with the corporate script. The letters borrow the fluidity of Coca-Cola's visual language without duplicating the protected Spencerian mark. The red field, white type, cap branding, bottle shape, and neighboring packages complete recognition. The hierarchy makes the individual name the newest information and Coca-Cola the assumed context.

This arrangement creates a double address. The package can seem to speak to Emma because her name is present, but the imperative addresses someone who knows Emma. The shopper is positioned as a finder and potential giver. The first name is intimate enough to suggest a person but common enough to support mass production. Personalization is therefore an effect of placement and encounter, not one-to-one manufacture.

The Mate can modifies the same frame. “Mate” is not a proper name and does not identify a unique person. It specifies a relationship in ordinary Australian English. Placed in the same typographic slot, it demonstrates that the campaign's globalizable unit was already broader than names. Coca-Cola could insert any socially legible addressee that completed “Share a Coke with.” The Australian origin contained the logic later expanded in China.

Australian Television Mosaics

The Kylie and Jason still contains two large faces built from many smaller portraits. The pair are positioned at equal scale, creating a visual relationship before the viewer reads the names. The mosaic technique joins uniqueness and repetition: one named face emerges from a crowd of people who share that name. The television execution solves a problem created by mass personalization. A common name can feel personal while openly belonging to many people.

The Liam still uses one composite face, a centered name, and a block of explanatory text. Its relationship cue is weaker in isolation, but the mosaic still turns a personal name into a collective visual object. The red campaign panel, white logo, and package imagery stabilize the frame. The personalized word and face receive attention; Coca-Cola owns the visual environment in which they appear.

King's Cross and the Westfield Kiosk

The King's Cross execution changed the city sign into a temporary naming surface. The submitted name appeared at monumental scale beside the fixed Coca-Cola mark. Its visibility made private address public, while the MMS return converted public display into a portable digital image. The participant moved through several roles: sender, spectator, recipient of branded content, and potential distributor through Facebook or email. The execution did not merely ask people to post an existing advertisement. It produced an advertisement from their submission and returned it to them.

The Westfield kiosk brought customization into retail space. Campaign images show the red kiosk as a compact production site where the participant watches a generic can become personally labeled. The activation resolves the exclusion built into a list of 150 names, but it also makes the act of printing visible. Personalization becomes a performance and a queue-worthy event. The standardized can remains unchanged except for the controlled variable slot.

The Chinese Nickname Package

The Chinese package retains the red label, white type, cap, and bottle shape, but its verbal organization differs. The line 分享这瓶可口可乐与你的 (“share this bottle of Coca-Cola with your”) leads into a category such as 文艺青年. The focal term does not point to one legal or everyday personal name. It asks the giver to recognize a person as an artsy youth. That recognition may be affectionate, teasing, or self-conscious. The bottle performs an attribution.

The 闺蜜 bottle makes the relational structure still clearer. 闺蜜 conventionally refers to an intimate female friend. The term does not merely describe an isolated identity; it presupposes a relationship from which the giver can speak. Its English gloss “bestie” is useful but incomplete because the Chinese term is more gendered in common use and carries a particular intimacy. The packaging converts that relationship into a purchasable address while keeping the Coca-Cola system visually constant.

The Sports Field Hero Creative

The Chinese hero image places three chilled bottles in ice before a fenced sports field. Their labels read 室友 (“roommate”), 老兄 (“bro” or “old buddy”), and 神对手 (“formidable rival”). Figures playing in the background provide a social scene without assigning any one bottle to a visible face. The bottles occupy the foreground and remain larger than the people. This composition keeps the product as the agent that organizes relationships.

The selected terms form a small peer network. Roommate names shared living; buddy names familiarity; rival names competition. The field gives 神对手 a plausible setting and allows rivalry to remain friendly. The Open Happiness lockup retains the wider brand platform. Rather than depict a national cultural value, the ad builds a local scene in which several forms of male-coded peer address can be exchanged through the product.

Baidu Placement and Campaign Microsite

The Baidu screenshot shows a red nickname bottle inserted into a search-results page. The bottle is both image and route. A user searching on the platform encounters the television creative and a link to the campaign site. The white page gives the red object high contrast, while the familiar bottle shape carries recognition at a small digital scale. Localization here is infrastructural: the campaign inhabits the search service through which users in China could find and enter it.

The microsite presents nicknames as a browsable gallery with tools for redistribution through Chinese social platforms. Bottles appear as selectable identity objects rather than a single finished message. The interface encourages comparison: Which label fits me, and which fits someone I know? The gallery turns the campaign's finite vocabulary into a small social taxonomy. Sharing buttons translate classification into circulation.

What the Visual Comparison Shows

Across the twelve images, the campaign protects recognition through repetition and moves local meaning into a controlled central slot. The red-white system works like a frame around variable language. Australia often anchors the slot to a proper name and then expands it through faces, live display, and custom printing. China anchors it to a type or relationship and then expands it through a platform gallery, search placement, and contextual scenes.

The comparison also shows why “names versus nicknames” is incomplete. Both campaigns use the label to position a giver in relation to another person. Australia's proper names require the viewer to supply the relationship. China's categories often name the relationship or characterize the person. The campaigns differ in the semantic work delegated to the label, not in whether they are social.

Comparative Discussion

Coca-Cola Standardized an Interaction Script

The most stable element was not a sentence translated word for word. It was an interaction script: recognize the Coke; notice an unexpected identity term; connect that term to a person; and move the object or its image toward that person. Color, typography, bottle form, and wordmark made the first step nearly automatic. The replaceable label created surprise. The share imperative supplied direction. Digital and live tools extended the action.

This finding refines the standardization-adaptation debate. Coca-Cola standardized the relations among components while allowing the components themselves to change. A first name, relationship, persona, dialect term, or user-entered word could occupy the same branded slot. Facebook, MMS, Weibo, Baidu, or a vending machine could supply the route outward. The campaign remained coherent because the parts performed comparable functions.

Personalization Did Not Mean the Same Thing

In Australia, mass personalization was primarily indexical. A first name pointed toward a person. Its persuasive force depended on recognition: that is my name, or I know someone with that name. The package's social value often began with a match.

In China, mass personalization was primarily attributive. The label characterized someone through taste, status, style, gender, peer relation, or regional language. Its social value could begin with judgment: this sounds like me, this is my friend, or this is funny precisely because it does not fit. The fit was negotiated rather than literal.

This difference complicates the assumption that names express individuality while Chinese labels express collectivism. The Chinese persona terms could support conspicuous self-branding. Australian first names were embedded in a gifting command. The better distinction is between proper-name identification and socially negotiated classification. Both are relational once the package is used.

The Brand Gave Up Textual Centrality but Kept Structural Control

Share a Coke is often described as Coca-Cola handing the brand to consumers. The evidence supports a narrower claim. The company gave consumers and local teams a controlled portion of the label. It kept the red field, object, campaign syntax, eligibility rules, production system, and platform design. Even user-generated names passed through filters and interfaces.

That controlled opening was still meaningful. A corporate name temporarily yielding to Emma or 文艺青年 changes the social voice of the package. Consumers can use the object for affection, teasing, collection, or display. Reciprocal identity co-creation occurs because those uses feed back into what Share a Coke means. Yet the corporation remains the architect and beneficiary of the exchange.

Media Localization Was as Important as Verbal Localization

The Australian campaign's participation system reflected the media environment and campaign capabilities of 2011: Facebook creation, SMS submission, MMS return, radio songs, a landmark digital sign, and mall kiosks. The Chinese campaign used Sina and Tencent Weibo, Renren, Baidu, celebrity seeding, and a campaign site designed for local sharing. Those differences affected who could encounter, modify, and redistribute the label.

Platform choice is therefore part of cultural adaptation, though it should not be mistaken for culture itself. Services have technical affordances, commercial relationships, governance, and user conventions. A campaign becomes locally usable by entering those systems. Translation without platform adaptation would have preserved wording while limiting circulation.

Glocalization Happened Through Specific Creative Decisions

The comparison supports Robertson's (1995) account of universalizing and particularizing processes occurring together. Coca-Cola's global identity made local occupation of the label valuable. Local language made the global package newly noticeable. Maynard and Tian's (2004) observation that brands adapt Chinese digital communication while retaining global identity appears here across packaging and participation as well as websites.

The evidence also sets a limit on celebratory accounts of glocalization. Local vernacular was curated according to brand positivity and market relevance. Some labels carried gender, beauty, status, or class implications. A fixed product list simplified a changing digital vocabulary. Glocalization made the campaign usable, but it also commercialized particular ways of classifying people.

Discussion

The study contributes to global advertising research by shifting attention from whether a campaign is standardized to what kind of structure travels. Share a Coke's portability came from a modular design. Brand recognition was distributed across several cues, so the central word could change. The campaign invitation defined a function for that word, so local teams could select different identity resources without losing the action. The physical-digital connection was open enough to map onto different platforms.

This modularity explains why the campaign could localize deeply while appearing simple. A bottle label seems like a small surface, but it coordinates brand identity, language, product design, retail encounter, interpersonal address, and media circulation. The strongest adaptation in China was not the use of Chinese characters by itself. It was the decision to make the label a social classification drawn from platform vernacular and to phrase it as something one person assigns to another.

The findings support Voyer et al.'s (2017) account of reciprocal identity co-creation, with an important qualification. Consumers received resources for identity work, but those resources were curated and brand-safe. The Australian vote and Chinese social listening made audience language visible inside campaign development, yet Coca-Cola retained selection authority. Co-creation is best understood here as bounded reciprocity.

The results also complicate culture-based predictions. De Mooij and Hofstede's (2010) questions about self and social relation remain useful, but the artifacts resist a national binary. “Mate” and “Bestie” were present in Australia; custom personal names were still available in China. Chinese persona labels could be self-expressive, while Australian names were organized around giving. Oyserman et al.'s (2002) warning about overgeneralized individualism-collectivism contrasts is borne out by the campaign content.

For practitioners, the case suggests that global consistency can be protected through a system of redundant cues rather than identical copy. A global team can specify the campaign's recognition assets and interaction logic while local teams select language, relationships, and platforms. The creative risk lies in assuming that a translatable noun has a stable local meaning. The Chinese labels worked partly because their meanings were socially alive; that same quality made them hard to control.

The public-health implication is smaller but real. McDarby et al. (2018) show that personalized labels can change children's beverage choices. Share a Coke's identity appeal should therefore be analyzed as persuasion, not only as clever participation. The campaign attaches recognition and social connection to a sugar-sweetened product. This study does not measure consumption effects, but its findings show how the package was designed to make selection and gifting personally meaningful.

Limitations

The largest limitation is archive survival. The Australian case is documented by long-form agency and company records with accessible images. China's original microsite, many Weibo posts, and campaign videos are incomplete or unavailable. Contemporary Chinese sources preserve wording and execution descriptions, but a textual record cannot support the same visual claims as a surviving artifact. The study addressed this by restricting close multimodal analysis to twelve directly inspectable items.

The sample is purposive, not exhaustive or random. Counts describe the corpus and should not be generalized to every execution. Counting distinct package labels as artifacts gives the central language variable appropriate attention but also increases the weight of packaging in the dataset. Equal country totals support comparison but do not correct unequal source availability.

Only one researcher coded the material. The second-coder and kappa procedure proposed earlier did not occur, so no reliability statistic is presented. The item-by-item audit improved consistency but does not substitute for independent coding. Chinese translations were not independently verified by a bilingual reader. Original wording and conservative glosses reduce the risk of false equivalence but cannot capture every regional or pragmatic shade.

The study analyzes campaign content, not audience reception. It cannot determine whether consumers understood a label sincerely, ironically, positively, or critically. Campaign-reported metrics were not independently audited and cannot establish causality. The analysis also cannot represent cultural variation across all regions, language communities, ages, or social groups in either country.

Finally, the cases launched two years apart in different platform environments. Some differences may reflect time, agency practice, technology, or archive conventions as much as national market. The comparison identifies how these two campaign systems differed; it does not isolate culture as the single cause.

Conclusion

Coca-Cola adapted Share a Coke in China by changing the identity work performed by its most visible word. Australia's campaign usually placed a proper name in the branded slot and invited the shopper to find a person. China's campaign usually placed a social persona, relationship, or dialect expression in that slot and invited the shopper to characterize a person. The distinction altered the grammar of personalization while preserving the act of sharing.

What remained stable was a campaign architecture: Coca-Cola red and white, the bottle and wordmark, a replaceable label, a recipient-oriented imperative, and a path from package to circulation. Those elements made the campaign recognizable even when the central language and platform system changed. The global brand did not maintain consistency by repeating the same message. It maintained consistency by repeating the same communicative problem and allowing local executions to solve it differently.

The comparison also shows why global campaigns should be studied below the level of the slogan. Glocalization occurs in typography, pronouns, label categories, platform links, filtering rules, and participation tools. These choices determine who can recognize themselves, how they can address another person, and what the brand permits them to do next. Share a Coke succeeded as a global idea because its identity was firm at the edges and deliberately unfinished at the center.