Algorithmic Branding

A New Era of Marketing Management and Platform-Based Services

Context

The central terminology surrounding technoculture, technocultural fields, platform assemblages, affordances, algorithms, and desire networks establishes the foundations for a deeper conceptual understanding of the elements of algorithmic branding. Algorithmic branding has transcended the mere association of specific “mythical” qualities with a product or experience, becoming a multidimensional process of using media to manage communication. The goal of marketing professionals is to employ engagement, algorithmic activation, amplification, personalization, and connectivity practices to draw consumers deeper into the brand spiral, entangling them in brand-related networks of desire.

This points to the growing importance of platform service brands and the consequent transfer of brand power to technology companies. It also raises a series of ethical and pragmatic questions that marketing professionals, researchers, and policymakers can examine. 

Introduction

In just a few years, platforms such as Amazon, Google, and Facebook have achieved unprecedented levels of market dominance. New forms of marketing involving devices such as home computers and mobile phones have become ubiquitous and essential. The Internet of Things, service robots (Wirtz et al., 2018), lateral exchange marketplaces (Perren & Kozinets, 2018; Wirtz et al., 2019), and augmented/virtual reality are rapidly gaining ground (Gäthke, 2020). Consumer tastes have been radically transformed by access to vast amounts of information and choice, as well as by the influence of online influencers, online crowds, and word-of-mouth (Keiningham et al., 2018; Kozinets et al., 2008).

Enormous amounts of detailed information about billions of consumers are widely available. The amount of available information is measured in the order of quintillions of bytes of data. Analyzing this information to gain sophisticated insights into even the smallest market behaviors has become a vital source of competitive advantage (Field et al., 2018). The importance of these capabilities has led to a new arms race based on surveillance capitalism and big data analytics (Breidbach & Maglio, 2020), resulting in a field in which brand management has merged with operations management and computer science in the pursuit of advanced applications for processing consumer data. Meanwhile, services and products are being delivered by virtual influencers, chatbots, service robots, and voice-activated AI virtual assistants such as Google Home and Amazon Alexa (Sidaoui et al., 2020).

New AI chatbots, such as Microsoft's Xiaobing, are beginning to embed automated and branded personalities into these technologies.

For marketers, Amazon, Google, Facebook, Twitter, Netflix, and other platform brands are clearly service brands, as they offer an intangible product in which “the company [and in this case the platform] is the primary brand” (Berry, 2000). Moreover, as discussed in services marketing, platform service brands involve “human thinking, conduct, organization, and expression within the logic of big data and large-scale computing,” which Striphas (2015) defines as “algorithmic culture” and considers “a movement that changes how the category of culture has long been practiced and understood.”

The term algorithmic branding is a convenient way of capturing the new reality in which a relatively recent type of service brand—technology-based platform service brands—has radically altered and continues to transform the world of branding. Algorithms are not the only element of this transformation; they are fundamental mediators and performative elements. Algorithms are part of the co-coding of assemblages that encompass all the complexities involved in contemporary branding, including consumers, stakeholders, platforms, technologies, brands, processes, dynamics, and practices involved in branding, as well as their externalities.

Algorithmic branding unfolds the practice of thinking about, organizing, and implementing branding through the logic of big data and large-scale computing. In this process, brand managers, retailers, consumers, platform developers, influencers, policymakers, regulators, and others actively participate in branding.

Algorithms are already part of brand and service research, including studies investigating how chatbot interviews with consumers can be enhanced through AI using sentiment-analysis extraction algorithms (Sidaoui et al., 2020), as well as research developing e-commerce recommendation systems (Schafer et al., 1999). However, brand experiences are holistic and broad concepts (Brakus et al., 2009; Hoffman & Novak, 2018). Brand experiences increasingly occur within contexts involving information and communication technologies and are culturally experienced.

Providing a conceptual explanation of the role of platform service brands in contemporary brand experience requires the incorporation of broader concepts.

One important concept is that of assemblages. Assemblage theory is a comprehensive socio-material theory of social complexity that emphasizes what emerges, becomes stabilized, and becomes destabilized through interactions between ontologically equivalent human and non-human actors (Hoffman & Novak, 2018, p. 1180). A fundamental application of assemblage theory to branding is to consider the broader interrelationships among the various components of contemporary brand experience, from platforms to locations, consumers, and products, to devices, functions, language, skills, competencies, and goals. 

Technoculture

Technocultures are “the various identities, practices, values, rituals, hierarchies, and other sources and structures of meaning that are influenced, created, or expressed through the consumption of technology” (Kozinets, 2019, p. 621). Contemporary technoculture includes familiar manifestations such as selfies, emojis, avatars, memes, augmented reality, online word-of-mouth (Keiningham et al., 2018), chatbots (Sidaoui et al., 2020), Instafame (Marwick, 2015), and Zoombombing (Lee, 2021).

Marketers and consumers communicate with one another because they are mutually embedded within a technocultural field composed of platforms and their content, but also of meanings, identities, and values communicated through that content. Content differs for each individual within the technocultural field and may consist of text, audio (e.g., podcasts, Discord, Clubhouse), prerecorded audiovisual content (e.g., YouTube unboxing videos), livestreams (e.g., Twitch, Facebook Live), graphic images, and combinations of all of these (e.g., Reddit, Facebook, or Instagram).

One of the primary ways in which brand experiences occur today is through platform assemblages. Technology platform companies such as Facebook, Twitter, Amazon, and Google constitute the core of these assemblages. 

Affordances

Affordances are potentials for behavior that emerge from the relationship between an object (e.g., the TikTok or Sina Weibo platform) and one or more goal-oriented actors and are associated with achieving a concrete, immediate outcome (Bygstad et al., 2016; Forte et al., 2014). One example would be a feature that allows consumers to rate a seller, as found on many e-commerce websites such as Amazon or eBay. Whether the company uses a five-star system or a rating from one to ten, it is an affordance—an opportunity to evaluate that is shaped by the technical feature.

The absence of affordances on a platform can prevent a user, whether a company or consumer, from achieving a specific outcome, such as when a website does not allow commercial discussions (Reddit) or does not allow consumers to rate a product, organize a boycott, or leave a detailed written review of a supplier (Kozinets et al., 2021). Affordances shape the consumer's experience of the brand and the platform. And because social media are social, platform features shape consumers' social and cultural experiences.

xQc, currently considered Twitch's leading streamer, generated nearly US$2 million in revenue in 2020 through content on his platform—a form of “servitization” (Field et al., 2018; Zeithaml et al., 2014) for the video game and technology industries, and for brands such as Xbox, PlayStation, Call of Duty, Animal Crossing, Logitech, among others. Another important concept is that of “produsers” (Bruns, 2008), which combines production and use, capturing the blurred distinction between consumption and active creation.

Inherent in the operations of platform assemblages and their possibilities are programming algorithms, which are typically invisible to users (making them opaque) and under the control of technology platforms (making them impossible for marketers to manipulate). Algorithms are complex concepts that “can be conceived of in many ways—technically, computationally, mathematically, politically, culturally, economically, contextually, materially, philosophically, ethically—but are best understood as contingent, ontogenetic, and performative in nature, and embedded within broader sociotechnical assemblages” (Kitchin, 2017, p. 14).

As many authors have written, hidden algorithms affect search engine results, calculate credit ratings, suggest insurance prices, help determine who will enter which school and which jobs they will obtain, and influence many other important aspects of services (Breidbach & Maglio, 2020; O'Neil, 2016). They also have important ethical consequences that affect society. As services become datafied and subject to algorithms, they can result in surveillance, loss of privacy, and “total manipulation and limited self-determination” (Breidbach & Maglio, 2020, p. 180).

Based on Deleuze and Guattari's (1983) assemblage theory, “networks of desire” combine algorithmic technologies, consumers, energized passion, and virtual and physical objects into interconnected systems that produce and amplify consumer interest among network actors, as well as within parts of the broader social system integrated into the network (Kozinets et al., 2017, p. 667).

Within these ubiquitous networks, a brand such as Disneyland or Gucci may become an object of desire, blending with many other related products, brands, and consumption practices within technocultural fields. For example, Billabong or Hurley may be more strongly embedded in networks of desire associated with surf culture than through their own specialized brand communities. 

Algorithmic Branding in the Technocultural Context

Three main actors are emphasized in the figure: platforms, which control their own technology and algorithms; marketers from companies or organizations, whose task is to implement branding in support of sales or other organizational objectives; and consumers, who influence and affect brands and are more participatory and productive than in previous branding models. In addition, brand stakeholders, influencers/content creators, the public, and regulatory and legislative actors and institutions also play important roles.

Information and communication become content that can be repurposed in multiple ways and across different formats and platforms. A vital part of this development and circulation of content is the work of personal brands competing within online attention economies (Smith & Fischer, 2021), often associated with content creators, influencers, and producers.

As consumers interact with platforms, they leave online traces that become valuable when treated as data for investigating questions such as consumer needs or demands (Kozinets, 2020, p. 16). Because of their proximity to this data, platform companies possess more information about consumers, leveraging it for their own benefit and building business models based on selling their insights to third parties.

However, a smaller brand such as Ole Henriksen may depend on Instagram for promotion and on Amazon or Facebook Live for sales. Sellers of collectible books, vintage clothing, and comics, for example, are highly dependent on eBay for their sales. Many brands are discovering that they are dominant exchange partners (Fischer & Rueber, 2004) with service platforms.

Consider the hypothetical example of a sportswear merchant who becomes highly dependent on Amazon as a retail channel. The more the marketer depends on Amazon for sales, the more likely changes in Amazon's search algorithms—for example, changes favoring a new Amazon sportswear brand—are to affect the marketer” (Kozinets & Gretzel, 2021, p. 158). Platform dependence is a marketing vulnerability that must be protected against, as well as an asset to be leveraged.

Outside platforms, companies still have their own websites, but the overall influence of these sites, without the support of paid search and social media to attract consumers, is declining, as is the influence of other forms of organic owned or shared media, such as organic brand communities. “The increasing use of AI by marketers creates power imbalances and makes them more vulnerable to algorithmic changes” (Kozinets & Gretzel, 2021, p. 2). 

The Brand Desire Spiral

Brands have been described as “sophisticated networks of information, association, and feelings” (Berthon et al., 2003). These networks have become even more sophisticated in a world where information and communication networks are ubiquitous and where media platforms are service brands that create value for consumers, marketers, influencers/content creators, audiences, and others.

Evolution of Spiral Bubbles: Brand Atmosphere + Ecosystem + Moments (Touchpoints)

The relationship between brands and media is fundamental. Historically, brand management evolved from notions based on the transmission of meaning associations and persuasion toward practices of consumer co-creation, participation, mutual storytelling, and engagement within a set of intrinsically interconnected “networked narratives” (Kozinets et al., 2010).

Today, “‘everything is media’ in the sense that [media denotes] anything—from living bodies to material objects—that can capture, channel, store, process, or display information can potentially be incorporated into the structure of the brand” (Arvidsson, 2006).

Branding has transcended “attaching specific ‘mythical’ qualities to a product or experience,” whether those qualities involve psychological associations or cultural ideologies, and has become a more complex matter: “the multidimensional process of using media to manage communication in general” (Carah & Brodmerkel, 2020, p. 8).

Brand desire is depicted as a spiral composed of numerous circles of different colors, patterns, and sizes. Each of these circles represents a branding incident or moment—an event in which a brand is mentioned, used, or even thought about.

The ongoing goal of marketers is to take consumers deeper into the brand spiral, entangling them in brand-related networks of desire found in the spiral's innermost regions. The goal is ongoing because the marketer's work is to meaningfully incorporate the constant flow of new products, brands, and services into the spiral, while continuing to deepen relationships with existing brands, products, and services.

Algorithms, AI, and platforms are mobilized, often in real time, to associate the brand with particular consumer identities, needs, and activities in increasingly experiential ways. For example, a parental and family identity may be targeted by Disney for its Disney+ streaming service. A Marvel or Star Wars comic-book fan identity may also be targeted by Disney for the same streaming service.

The goal of branding efforts in this example is to create a digitally augmented “brand atmosphere.” This atmosphere is multimodal, complex, and multidimensional—and increasingly includes the metaverse. It attracts target consumers—wherever they are and whoever they may be—to increasingly comprehensive and co-created brand experiences. This occurs within a competitive field.

For example, Disney would be positioned against other streaming services such as Netflix and Paramount, which could offer their own family entertainment streaming services and universes of science-fiction emotions. A crucial way for brand managers to achieve the orchestration necessary to gain a competitive position is by capitalizing on the technical capabilities, richness of information, multimedia elements, connectivity, and mobility of digital platforms, digital televisions, computers, smartphones, wearables, and other devices.

Brand messages are amplified, for example, through reinforcement or repetition via algorithmically informed paid media when they attract customers or engagement—a measurable correlate of attention and interest. Messages are translated as they move into community conversations, such as through an influencer whose posts may be whitelisted and promoted in the feeds of similar individuals, as determined by algorithms and AI.

This is where the influential power of crowds, content creators, and fan groups can intensify interest. Communications, products, and even brands can be personalized in response to relevant patterns in brand engagement metrics and other relevant KPIs detected by algorithms and AI.

Brand activations can occur in specific contexts, such as promoting brands or brand-related behaviors within the Disney app while a consumer is at Disneyland using the location-based application. The ubiquitous connectivity of mobile applications and devices creates opportunities to redirect consumers' purchases and interests back toward brands.

All these movements—and many more—are supported by individual brand moments, sustained by platforms, companies, online groups, and consumers acting together, coordinated to take consumers toward increasingly deeper moments of engagement with brand desire networks.

In this way, the brand spiral reveals how the use of social media and other forms of networked connectivity is “reconstituting the brand,” marking a “dramatic shift” in which the brand has become increasingly involved in “feeding short-term cultural phenomena” (Fournier & Avery, 2011, p. 206; Hampton, 2016).

Because of the ubiquity of social media, collective and individual experiences become blurred—and the brand is increasingly involved in this fusion. We can consider algorithmic branding within the brand spiral as drawing upon and developing a type of “brand omniconnectivity,” in which consumers may remain engaged for extended periods in “uninterrupted flows of attention and data sharing” related to the brand (Carah, 2017, p. 398).

Algorithms enable not only more precise advertising targeting, but also “an expansion of the facets and qualities of audience attention,” making them effective as a modality for carrying and responding to human experience (Carah, 2017, p. 397). Algorithmic branding enables a more complex, varied, interactive, engaging, and precise interconnection of experiences among marketer, consumer, and brand than ever before. 

Implications for Branding Practice and Research

Several practical research directions could address the many gaps in these initial conceptualizations. It may be useful to begin cataloging the variety of brand-related behaviors that occur within and through platform assemblages.

Working in an era preceding social media, Schau et al. (2009) provide a valuable set of general processes for “collective value creation in brand communities.” However, their categories of brand use and community engagement need to be updated for a world of algorithm-driven platform branding rather than the earlier era of organic “online brand communities.”

They must consider that “social networking” and “community engagement” functions have increasingly been integrated into service platforms and are strongly affected by their affordances and algorithms. The exchange of practices, sharing of repertoires, and even social interaction are under the control of platforms whose objectives are often based on controlling and monetizing interactions rather than fostering a sense of community.

Schau et al.'s ideas of community-driven value creation urgently require modernization for a context in which the affordances, algorithms, and AI of service platforms measure, alter, and determine much more of consumers' online experience than before. These changes alter the practice of community management. The old ideals of community dissemination and open conversation are subsumed into data-driven metrics and profit-maximization objectives that constrain both.

Today, researchers have the opportunity to closely study how algorithmic branding is implemented through far-reaching, consumer-centered brand behaviors such as consumer reviews, influencer posts, and online activism.

Online, consumers evaluate products and companies using the possibilities and constraints of platforms. They also use brand-related features playfully for a variety of cultural purposes, including unexpectedly transforming product reviews into a form of humorous entertainment (Kozinets, 2016).

Influencers and word-of-mouth marketers are employed to translate marketing information into relevant and seemingly authentic community messages (Kozinets et al., 2010). However, if these messages are not amplified as paid media to reach larger audiences, they may be left to wither because of corporate algorithms.

Other consumers act as “institutional entrepreneurs” seeking market change through social media activism (Scaraboto & Fischer, 2013) and may similarly be treated as content creators to be promoted or silenced.

With the flip of a switch, an attractive influencer or a compelling activist can be hijacked and silenced. Replacing the old community manager model will be a new generation of specialists skilled at navigating the new waters of algorithmic branding.

This will include industry-focused social platform lobbyists, algorithm snake charmers, powerful influencer relations agencies, professional digital activists, as well as increasingly specialized brand, segment, and industry data analysts and other new professions that are still emerging. 

Implications for Branding Practice and Research

Breidbach and Maglio (2020) provide the example of insurance companies and medical services monitoring images of junk food posted on a consumer's Instagram account and using them to set rates and provide advice. Service brands in the financial, insurance, and healthcare sectors already hold enormous amounts of private consumer data and use platform services that know a great deal about individual consumers' preferences, locations, social networks, and personal information.

Beyond simple information sharing, there is potential for manipulation—also described and discussed by Breidbach and Maglio (2020). Scholars such as Zuboff (2019, p. 375) have been particularly harsh in their descriptions of unethical marketers who use platforms, algorithms, and AI to modify or shape consumer behavior: “the behavioral market regime of surveillance capitalism finally has at its disposal the instruments and methods that can impose [B. F.] Skinner's behaviorism through the varied domains of everyday life and into our depths, now conceived as the global laboratory of capital.”

Individual consumers or users are compared with other users and grouped based on: (1) the social contexts in which their data emerge, (2) the content they use, produce, and consume, and (3) the values assigned to them by various individuals and agencies, including software algorithms, based on those contexts and contents.

Personas are replaced by profiles. Research can examine how algorithmic outputs, human experience, complex data flows, and diverse interpersonal factors combine quantitative clustering with qualitative identity-based labels for market segmentation, producing the lens through which marketers identify and monitor those who consume their products and services (Kotliar, 2020).

Would algorithmic branding—with its advantages in terms of engagement, manipulation, and activation—be welcomed by service consumers within a given context? 

Conclusion

Platform brands are powerful and important service brands upon which marketers and consumers depend. Platform brands establish algorithmic branding, which involves the practice of thinking about, organizing, and implementing branding through the logic of big data and large-scale computing.

In practice, algorithmic branding uses the decision-making logic of algorithms and artificial intelligence, together with the ubiquity of mobile devices, to “tune” consumer experiences in much the same way that automated algorithms in slot machines are designed to modulate players' activity (Schüll, 2012).

For example, Facebook continuously experiments with its news feed and advertising algorithms to increase brand engagement in a manner analogous to how casino-machine algorithms attempt to hook players (Justice-Leibrock, 2013; Van Dijck, 2013; Zuboff, 2019).

Algorithmic culture can be understood as the use of “computational processes” to classify and rank people, ideas, and other things and concepts, as well as “the habits of thought, conduct, and expression that emerge in relation to these processes” (Hallinan & Striphas, 2014, p. 3).

How do these computational processes set in motion the connections among brands, people, desires, companies, and cultures?