Abstract
Internet meme marketing is a digital marketing practice in which marketers leverage internet memes to promote their brand or product. Despite its growing adoption in brand communications, academic understanding of meme marketing remains in its infancy. This research theoretically develops and empirically tests how, why, and when meme marketing is effective. The authors find that meme marketing enhances digital advertising effectiveness by fostering shared psychological ownership of the marketing message and subsequently a strengthened self–brand connection. In addition, the effectiveness of meme marketing diminishes when the leveraged memes are not in their maturity stage (i.e., during the introduction, growth, or decline stages) or when applied to promote niche products. Six studies—including a large-scale secondary data analysis (N = 900,139 posts), a field experiment (N = 423,565 impressions), and four controlled experiments (N = 3,958 participants)—provide robust and converging evidence for these propositions. The effectiveness of meme marketing is demonstrated across both behavioral outcomes (e.g., likes, click-through rates, conversion rates) and attitudinal responses (e.g., purchase intention, likelihood of liking the post). This research advances theoretical understanding of meme marketing and offers actionable insights for practitioners seeking to leverage internet memes in their brand digital marketing communications.
Keywords
Meme marketing is rapidly emerging as a popular digital marketing strategy. For example, the “It feels like a movie” meme trended recently after pop star Harry Styles remarked, “My favorite thing about the movie is, like, it feels like a movie.” Following the trend, Delta Air Lines posted a tweet on X that replaced “movie” with “airplanes,” 1 and DreamWorks Animation shared a similar post using the word “animation” in place of “movie” 2 (see Web Appendix A for more examples). According to an industry report, on average, firms have been allocating 5%–10% of their marketing budgets to meme marketing, with some startups investing as much as 30%–40% (Dey 2022). Considering that global digital advertising spending reached $470 billion in 2024 and is projected to rise to $558 billion by 2026 (Statista 2024), meme marketing may already account for a substantial and growing portion of marketing investment. While meme marketing shows significant promise for enhancing digital communications, academic understanding of this practice remains in its early stages. There is a lack of fundamental understanding of what meme marketing is, as well as systematic empirical evidence of its effectiveness and the underlying mechanism that drives such effectiveness. We also know very little about how firms can leverage this novel strategy to fully harness its benefits. The current research aims to fill these gaps to provide actionable insights into this increasingly popular marketing practice.
Meme marketing, by its very nature, is a marketing-oriented “remix” or adaptation of an established meme. We thus conceptualize meme marketing based on memetic process theory (Shifman 2013, 2014), in which memes are characterized as collective creations of culturally resonant content with template-based variation and wide dissemination. We further map these memetic processes onto the communication model (Lasswell 1948; Peng, Eisend, and Chen 2025; Shannon and Weaver 1949), which specifies four core elements in the communication flow (i.e., source, message, channel, and receiver). Through this theoretical lens, we are the first to identify four critical aspects of meme marketing: creator as communication source, format and content as communication message, and diffusion as communication channel and receiver. Accordingly, we define meme marketing as (1) a cocreation between brands and netizens (creator) (2) that uses an established and recognizable meme template (format) (3) to integrate marketing messages with culturally resonant memes (content), which is (4) capable of leveraging the diffusion trend of memes to consumers (diffusion).
Building on this systematic definition of meme marketing, we investigate several important research questions: (1) Is meme marketing an effective digital advertising practice, like generating social media engagement and financial returns? (2) If so, why is it effective, given its uniqueness in creator, format, content, and diffusion? That is, what underlying mechanisms drive consumers’ favorable reactions? (3) How might contextual factors associated with meme marketing’s unique aspects or communication model moderate its effectiveness so that firms can better implement strategy to enhance their benefits? These research questions are closely related to the challenges that practitioners face when executing meme marketing. According to a survey (see Web Appendix B) of 100 U.S. digital marketers using Qualtrics Research Services, marketers often face the challenge of selecting the right memes to leverage, given that some may no longer be effective (e.g., “People don’t like [this meme] because it's overused”), and it may be difficult to pair memes with the right product (e.g., “It is childish and not related at all to what I do”). The answers to these questions can offer a nuanced understanding of how, why, and when meme marketing influences digital advertising effectiveness.
Figure 1 depicts the framework of the current research. We answer the aforementioned research questions by adopting a multimethod approach across six studies: a large field study using secondary data, a field experiment employing industry-relevant metrics, and four controlled experiments. To better contextualize our findings and provide guidance for managerial actions and future research, we also propose a comprehensive research landscape by synthesizing our findings, insights from prior literature, and potential opportunities for future studies. The framework encompasses the antecedents, mechanisms, outcomes, and moderating contextual factors of meme marketing. Beginning with the antecedents, we decompose meme marketing into two main elements: product or brand information and existing internet memes. By remixing their product or brand information into existing internet memes, brands engaging in meme marketing become cocreators of meme variations, effectively leveraging shared cultural knowledge and memetic diffusion to promote their products.
Research Framework, Mechanisms Identified in Prior Literature, and Future Directions.
Moving to the mechanisms and outcomes, we theorize and find that consumers develop a sense of shared psychological ownership (Kirk, Peck, and Swain 2018; Peck et al. 2021) toward the meme marketing message—a feeling that the marketing message is jointly owned by the brand and the netizens who collectively created the meme. This sense of shared psychological ownership strengthens the self–brand connection and, in turn, boosts digital advertising effectiveness, as measured by the number of likes in real social media (Study 1), real purchase behavior and other digital advertising performance metrics (e.g., click-through rate [CTR], conversion rate [CR], return on advertising spend [ROAS]; Study 2a), likelihood of liking the post (Study 2b), and purchase intention (Studies 2b, 2c, 3 and 4). As shown in Figure 1, our findings complement those of alternative mechanisms (e.g., societal mood; Jung and Jeong 2021) or more general mechanisms (e.g., engagement; Malodia et al. 2022) identified in the literature. While our empirical analysis demonstrates the positive impact of meme marketing on a wide range of outcomes, its effect on additional downstream consequences remains open for future research, such as post longevity and diffusion rate at the post level and referral and repurchase intention at the brand level.
Furthermore, Figure 1 highlights the possible moderating contextual factors structured along the four elements of the communication model (Lasswell 1948; Shannon and Weaver 1949). For the source, factors that may influence the effect of meme marketing include the type of products (e.g., mainstream vs. niche products that are remixed in meme marketing; tested in Study 4) and general brand features (e.g., identities, positioning, social media personas). For the content, factors influencing its effect could be the meme life cycle (e.g., introduction, growth, maturity, and decline stages; tested in Study 3), special memes (e.g., political memes), the compatibility of the meme with the overall company communication strategy, and synergy with other digital marketing strategies. For the channel, platform policy (e.g., censorship, intellectual property protection) and platform infrastructure (e.g., built-in templates) might serve as moderators. For the receiver, the consumer's sense of internet community (tested in Supplementary Study C) and existing consumer–brand relationship are considered important. These contextual factors highlight the contribution of the current research and provide a landscape for future research directions, as described in the “General Discussion” section.
This research framework makes important theoretical and managerial contributions. First, it enriches the digital marketing literature (Farace et al. 2026; Shankar et al. 2022) with a new definition and investigation of meme marketing based on its four core aspects (i.e., creator, format, content, and diffusion). The definition also distinguishes meme marketing from other commonly used digital marketing practices (e.g., viral marketing, influencer marketing, electronic word-of-mouth [eWOM] marketing) (see Web Appendix D for a comparison). Due to the unique features of internet memes, meme marketing is best positioned as a message type or content style, distinct from influencer marketing or eWOM marketing, which emphasize the initiator (creator). These other digital marketing practices can use the content nature of memes to better facilitate their effectiveness. Second, our research contributes to the nascent field of meme marketing by identifying shared psychological ownership as a novel consequence. Compared with mechanisms identified in prior research, our proposed mechanism is uniquely attributable to meme marketing, as it encompasses all four core aspects of meme marketing. Third, our research builds on the thriving practice of digital marketing to offer specific recommendations for executing meme marketing. The proposed moderators—meme life cycle stage and product type—provide valuable insights for effectively implementing meme marketing.
Theoretical Development
Internet Memes, Memetic Process, and Meme Marketing
Derived from Greek word “mimema” and shortened by Dawkins (1976) to rhyme with “gene,” a meme was defined as the basic unit of any cultural idea or concept transmitted from one mind to another, analogous to the biological transmission of genes. Scholars have since adapted this concept to encompass internet memes, defined as “a group of digital content units with common characteristics, created with awareness of each other, and circulated, imitated, and transformed by many users” (Shifman 2013, p. 367). While a single text, image, or video may not qualify as a meme on its own, internet memes typically manifest as recurring clusters of similar texts (e.g., “Therapist: No”), images (e.g., “Galaxy Brain”), or videos (e.g., “An Idiot Sandwich”) (see Web Appendix A for more examples). A specific meme series or collection is generated through a collective creation process by netizens and is easily recognizable at a first glance because of a template-based format (Shifman 2014). The prevalence of memes has sparked a stream of research into their applications within the business context (see Table 1).
Internet Meme Related Research Under Business Context.
The process through which memes replicate, spread, and evolve across individuals and societies is known as the memetic process, which has several key components: creation, replication, variation, and selection (Shifman 2013, 2014). Memes begin with creation, where an idea, image, or cultural artifact is shaped into “spreadable media”—content designed to facilitate sharing and engagement (Jenkins, Ford, and Green 2013). In the replication phase, spreadable media is transmitted from one person to another, often through replication or imitation with a defined and easily recognizable template. As memes spread, they may undergo variation, where netizens modify them based on individuals’ perspectives and experiences, allowing memes to adapt to different contexts or audiences while remaining within recognizable templates (Wiggins and Bowers 2015). Finally, in the selection phase, certain emergent memes may grow to maturity with notable popularity or prevalence in society due to their appeal, relevance, or resonance with cultural values, while others fade into obscurity (Jenkins 2009). Through such a deliberate selection process, mature memes normally reflect or align with mainstream societal views and are easily recognized by most netizens (Zulli and Zulli 2022). The memetic process helps define internet memes as a unique form of online communication through four core aspects: (1) collective creation by netizens (creator), (2) use of easily recognizable and repeatable templates (format), (3) conveyance of shared knowledge or cultural resonance (content), and (4) wide diffusion through continuous remixing and replication (diffusion).
Digital marketers can actively participate in the memetic process by combining brand or product information with existing meme templates to generate “meme marketing posts,” a special type of meme variation. This practice is termed “meme marketing” (Razzaq, Shao, and Quach 2023, 2024), and the generated meme marketing posts can be regarded as marketing-oriented remixes or adaptations of existing memes. Incorporating the four core aspects of memes, we define meme marketing as a distinguishable digital marketing practice (see the full definition in the second paragraph of introduction).
Marketing posts using memes—under the premise that those memes can be recognized—could easily be used to engage netizens for several reasons: first, netizens would appreciate the effort of digital marketers to join the conversations among netizens; second, netizens would feel intimacy when they see the familiar meme templates in the meme marketing posts; third, the contents of the memes are usually appealing and have cultural resonance. Finally, the powerful meme diffusion network helps the meme marketing message reach a large audience. Therefore, these meme marketing posts, compared with generic firm-generated marketing posts, would be more likely to attract engagement and interactions from consumers on the internet. We thus hypothesize:
Shared Psychological Ownership and Self–Brand Connection
We follow a two-stage approach to probe into the psychological underpinnings of the increased effectiveness of meme marketing and conceptualize the psychological mechanism. First, building on the framework of psychological ownership from prior research (Kirk, Peck, and Swain 2018; Peck et al. 2021), we postulate that the four core aspects of memes (i.e., creator, format, content, and diffusion) help generate a sense of psychological ownership among netizens by synergistically enhancing the three contributors to psychological ownership: investing the self into the target, controlling the target, and coming to know the target intimately. First, being creators of memes, netizens not only invest themselves in the collective creation process but also gain intimate knowledge and understanding of memes (creator). Second, with the specific and recognizable meme templates, netizens control the “production lines” of different memes (format). Third, as the most popular meme variations have great cultural resonance, intimately knowing the meme's content inevitably boosts the perceived psychological ownership of the meme (content). Finally, during meme diffusion, netizens invest themselves in continuously remixing and controlling the diffusion process (diffusion). We conducted a pilot study to provide evidence for these important theoretical foundations (see Web Appendix C). Specifically, 392 U.S. netizens indicated a feeling of psychological ownership toward the internet memes that were popular around the time of the data collection during the pilot study. An analysis of an open-ended question, in which participants elaborated on what characteristics of the memes influenced their sense of ownership, revealed the four key unique aspects of memes—creator, format, content, and diffusion—in accordance with our conceptualization of what contributes to the psychological ownership perception. Importantly, we found that netizens tend to develop a general sense of psychological ownership toward internet memes, regardless of whether they directly participated in the meme creation by creating or remixing their own version. This aligns with prior research on crowdsourcing and cocreation, which shows that even without direct contribution, shared experience or marginal involvement in collaborative processes can foster a sense of ownership over jointly developed outcomes (Gerber and Hui 2013; Torfing et al. 2021). Despite this, please note that the recognition of memes is a necessary precondition.
Second, we extrapolate the “psychological ownership” generated toward memes into a “shared psychological ownership” of meme marketing—a combination of memes created by netizens and marketing messages added by marketers. Shared psychological ownership refers to the feeling of joint possession or responsibility over a particular object, idea, or content among two different groups—a feeling that “we” share a specific object with “them” (Storz et al. 2022). In the context of meme marketing, when a brand incorporates a popular meme into its marketing messages, consumers may feel a sense of shared psychological ownership toward the meme marketing message because they feel a psychological ownership toward the meme in the first place. When brands use the template of internet memes to disseminate their marketing messages, netizens will develop a greater perception of shared ownership toward these messages than when firms use only firm-generated content without incorporating meme content. Notably, the constructs of psychological ownership and shared psychological ownership are correlated yet different. In the context of memes and meme marketing, psychological ownership refers to a general feeling among netizens that internet memes are collectively owned by the online community. Since individuals identify as part of this group, they feel a personal sense of psychological ownership toward internet memes. In contrast, shared psychological ownership is a more specific construct that describes a feeling of joint ownership toward a meme marketing message, which is a combination of meme and marketing messages.
Shared psychological ownership not only fosters a personal connection to what is owned but also influences interpersonal relationships (Jami, Kouchaki, and Gino 2021) because it enhances interactions and relations between different participatory parties (Storz et al. 2022). In the context of meme marketing, we propose that shared psychological ownership shapes interactions and affects how netizens perceive and engage with brands. In other words, if consumers experience a sense of shared psychological ownership over a brand's marketing messages, this perception could contribute to a stronger self–brand connection. This connection is the link between a consumer's self-concept and perceptions of the brand concept (Moore and Homer 2008) or the extent to which consumers integrate brands into their self-concepts (Escalas 2004). Consequently, the interactions and relations between netizens and brands fostered by shared psychological ownership of a meme marketing post strengthen the extent to which netizens feel connected with the meme marketing brand.
Previous literature has established that consumers react more positively to brands that resonate with their self-concept than to less meaningful brands (Escalas 2004). Therefore, consumers who hold stronger self–brand connections with a particular brand exhibit more aligned values and behaviors, resulting in more favorable brand attitudes (Moore and Homer 2008). Accordingly, we anticipate that meme marketing can increase consumers’ perceptions of shared psychological ownership toward the marketing message, further enhancing self–brand connection and ultimately increasing the effectiveness of the brand's digital ads.
The Moderators
To test our theorizing and explore its managerial relevance, we examine two potential boundary conditions of the meme marketing effect: meme characteristics (meme life cycle stage) and product characteristics (niche vs. mainstream product). We selected these moderators based on two guiding considerations. First, the emergence of shared psychological ownership toward meme marketing messages relies on the sense of psychological ownership over the meme in the first place. Memes in some stages of their life cycle gain less investment, control, and intimacy from netizens, thereby constraining the development of psychological ownership. Second, certain products may not be compatible with the mainstream culture of meme marketing, rendering the shared psychological ownership toward meme marketing messages incompatible with the self–brand connection and positive brand responses. We elaborate on each of these moderators in the following sections.
Meme life cycle stage
As introduced, the memetic process contains key components of creation, replication, variation, and selection (Shifman 2014). Aligned with the memetic process and referring to the life cycle framework (Levitt 1965), memes can be categorized according to their life cycle stages: introduction, growth, maturity, and decline. Memes begin as spreadable content in the introduction stage, gain visibility and adaptations during growth stage as emergent memes, reach peak recognition and cultural relevancy in the maturity stage, and finally lose novelty and fade into decline stage as outdated or overused, replaced by newer trends. Figure 2 illustrates how memes in the four life cycle stages differ in terms of creator, format, content, and diffusion (see also Web Appendix E).
The Four Life Cycle Stages of a Meme.
We theorize that meme marketing is most effective when memes are mature, as netizens are more likely to develop psychological ownership for a mature meme. Mature memes—characterized by not only peak popularity but also extensive cocreation, standard and familiar templates, cultural relevance, and widespread recognition—foster a sense of personal investment, perceived control, and intimate knowledge. These factors contribute to psychological ownership toward the memes (Kirk, Peck, and Swain 2018; Peck et al. 2021) and enhance shared psychological ownership toward the meme marketing message. In contrast, memes in the introduction or growth stages lack widespread sharing and remixing and have limited participatory engagement and investment opportunities. Memes in the decline stage often suffer from overexposure, leading to cognitive fatigue and emotional disengagement. In all three cases, when consumers do not feel a sense of psychological ownership of a meme, they are less likely to develop shared psychological ownership of the associated marketing message, limiting its impact. Thus, we hypothesize:
Niche versus mainstream product
We conjecture that the positive effect of meme marketing only appears when the promoted product type is compatible with the nature of the meme. As a societal phenomenon, internet memes gain popularity among a vast number of netizens and serve as shared cultural symbols for the mass population (Shifman 2014). Certain products may not align with the mainstream culture of meme marketing, for example niche products. Niche products target a small group of consumers with special needs (Chang and Chen 2012), unique tastes (Schaefers 2014), or requirements for personalization or subcultural connection (Hoskins et al. 2021). These products often symbolize a small community, allowing individuals to deviate from the mainstream crowd (Schaefers 2014). When the mass appeal of memes conflicts with the small-scale nature of niche products, shared psychological ownership of meme marketing messages does not necessarily lead to self–brand connection and positive brand responses. Therefore, for niche products, meme marketing might be less successful at generating positive brand responses than for mainstream products.
Overview of Studies
We tested our hypotheses using a multimethod approach, including surveys, secondary data analysis, and field and laboratory experiments (see Web Appendix F for a summary and Web Appendix G for measures). Study 1 analyzed a large secondary dataset of 2,981 brands and 900,139 posts for 13 months and provides robust evidence for the effectiveness of meme marketing, supporting H1. Studies 2a–2c further tested this effect by manipulating meme marketing in more controlled experimental settings. Specifically, Study 2a demonstrated the effect in a field experiment, collaborating with a leading grocery e-commerce company in China; Study 2b replicated the effect by simulating a natural social media setting in a controlled experiment; and Study 2c established the sequential mediation effect of shared psychological ownership and self–brand connection, supporting H2. Studies 3 and 4 examined two theoretically and managerially relevant moderators: meme life cycle stage (H3) and product type (H4). Additionally, three supplementary studies in the Web Appendix further substantiated the findings: a secondary data analysis focusing on a single meme (Web Appendix H), an experiment validating the effectiveness of meme marketing by comparing it with other digital practices (Web Appendix I), and an investigation of the sense of internet community as a consumer-related moderator (Web Appendix J). Together, these studies reveal and explain the distinctive role of meme marketing in enhancing digital advertising effectiveness and provide actionable insights for marketers.
Study 1: Secondary Data Analysis
To gain an initial understanding of meme marketing, we examined its impact on social media platforms, where meme marketing is prevalent and native. Specifically, we investigated whether integrating memes into firms’ posts impacts the number of likes the posts receive. As an exploratory study, we focused on a single meme on Weibo, a leading Chinese social media platform. We found that when brands incorporated this specific meme into their posts, the number of consumers’ likes of the brand's posts increased significantly, reflecting the effectiveness of meme marketing (see Supplementary Study A in Web Appendix H). To broaden our results beyond the potentially idiosyncratic characteristics of this single meme, we aim to replicate the findings with a holistic dataset that includes multiple memes.
Method
To systematically collect the post data on Weibo, we first employed a snowball sampling technique to collect a list of 2,981 brands (Ellard-Gray et al. 2015) (see Web Appendix K1). 3 Specifically, we first selected 190 initial seed brands from the “hotly discussed brand list” of the platform. Then, we incorporated the brands that these seed brands follow into our database. Next, we collected all original posts from these followed brand accounts for the past 13 months, from October 2021 through October 2022. In total, our sample contained 900,139 posts. The textual content for each post was used to count the number of words in the post. We also observed whether the post contained video, the number of pictures and hashtags included, and the number of likes. We used the number of likes as an indicator of meme marketing effectiveness. To control for potential confounding characteristics at the brand level, we included factors describing the firm that owns the brand, including its history, ownership, and size, collected from tianyancha.com, which is a third-party database that provides firm profiles and legal information. We measured the history by the time duration in years since its establishment, ownership by whether the firm was publicly listed, and size by the number of employees.
We defined posts containing memes as meme marketing posts and the others as nonmeme marketing posts. 4 We identified 879 posts as meme marketing posts; a total of 13 memes were used in these meme marketing posts. 5 The most frequently used memes were: “yyds” (55.40%), “be like” (18.20%), and “The five types of things in the world” (16.61%) (for more details about these memes, see Web Appendix K2). Summary statistics for posts in the final sample are presented in Web Appendix K3.
Results
We use two modeling approaches. First, we employ a linear regression model with a log-transformed dependent variable to correct for skewness and include brand fixed effects to account for unobserved brand-level heterogeneity. Second, given that likes are count data with a substantial number of zeros, we estimate a zero-inflated negative binomial (ZINB) model. Because ZINB relies on maximum likelihood estimation, fixed effects cannot be included, due to the incidental parameters problem (Lancaster 2000). Instead, we include the three key brand-level controls: history, ownership, and size.
We report the results based on different model specifications in Table 2. Model I is a linear regression controlling for post characteristics and time fixed effects. Model II extends this by including brand fixed effects. Model III presents estimates from the ZINB model (full results in Web Appendix K4). Across all specifications, we find a consistent positive effect of meme marketing (b1 = .68, p < .001; b2 = .10, p = .020; b3 = 1.31, p < .001) on likes, indicating that meme marketing leads to at least 10% increase in social media likes.
The Impact of Meme Marketing on Social Media Engagement.
Notes: DV = dependent variable; N.A. = not applicable; AIC = Akaike information criterion. Standard errors are in parentheses.
To enhance the validity of the results, we perform a battery of robustness checks (see Web Appendix K5 for details). First, we use two additional outcomes, the number of comments and the number of reposts, to test the robustness of our findings. We find that meme marketing can boost not only posts’ likes but also their comments (b = 1.22, p < .001) and reposts (b = 1.67, p < .001), indicating the overall effectiveness of meme marketing in enhancing consumer engagement. Second, we address the selection/endogeneity issue due to observable factors. Meme posts and nonmeme posts may systematically differ in terms of observable factors, resulting in estimation bias. We implement propensity score matching to construct a sample of meme posts and nonmeme posts within the same brand that are comparable on observable dimensions, such as having a similar number of pictures and words (Abadie and Imbens 2016). We then estimate the impact of meme marketing using this matched sample and find consistent results (b = .12, p = .021). Third, we implement the Heckman correction to resolve the selection/endogeneity issue due to unobservable factors (Heckman 1979). For instance, for those posts with a meme, brands may spend more effort editing the post (unobserved in our data), leading to higher content quality of the post. To address this issue, we use an instrument variable that should be correlated with whether the post has a meme but uncorrelated with the number of likes the post receives. We use the monthly ratio of meme posts in the same industry as the instrumental variable for the focal brand. Specifically, for a given month, we count the number of meme posts and the number of all posts in an industry to calculate the meme ratio, while excluding all posts from the focal brand. In the two-stage estimation based on the instrumental variable, we find consistent positive effects of meme marketing (b = .10, p = .020). Finally, to account for the textual information of post content, we use machine learning techniques to match meme posts and nonmeme posts with comparable semantic meaning. Our study uses the Universal Sentence Encoder developed by Google, which is a type of Word2Vec model that converts unstructured text (e.g., a Weibo post) into 512-dimensional vectors (Yang et al. 2020). Based on these vectors, we then calculate cosine similarity scores between posts with memes and posts without memes. The analysis based on this matched sample demonstrates similar effects of meme marketing (b = .27, p = .018).
Discussion
Using a large, multimeme dataset, this study demonstrates that integrating memes into brand content can increase consumers’ positive engagement with brand posts. The results enhance the external validity of our findings and demonstrate the real-world value of meme marketing for firms. However, due to the secondary nature of the dataset, our main analysis may suffer from endogeneity concerns. To corroborate our findings, we conduct a series of robustness checks and use different empirical approaches to control both observable and unobservable confounding factors. Despite these efforts, we cannot perfectly identify the causal impact of meme marketing on digital advertising effectiveness. In our subsequent analysis, we use both lab and field experiments to strengthen the causal identification of the meme marketing impact and shed light on the psychological mechanism that drives such impact.
Study 2: Main Effect and Underlying Mechanism Evidence
While Study 1 provided initial observational evidence that meme marketing is associated with higher consumer engagement in the marketplace, its nonmanipulative design leaves important open questions about causality and underlying mechanisms. In Study 2, we conducted three controlled experiments that served different purposes but had two main common objectives: (1) verify the effectiveness of meme marketing across diverse product categories and experimental settings, and (2) identify the psychological processes through which meme marketing influences consumer responses.
Study 2a began with a field experiment embedded in a live digital advertising campaign that used a real meme. This experiment provided ecologically valid evidence of meme marketing's effectiveness on key performance metrics such as click-through rate (CTR), conversion rate (CR), and return on advertising spend (ROAS). However, as with other online experiments conducted directly on established social media sites, this experiment had several limitations, including a lack of mutual exclusivity and the nonrandomized assignment (Braun et al. 2024). To address these limitations, Study 2b and all later studies employed a controlled experimental design with full randomization. Specifically, in Study 2b, we created a simulated social media feed to mimic the naturalistic context in which consumers encounter a meme. This approach represented an externally valid manipulation of whether content was perceived as a meme. Study 2c transitioned to a tightly controlled laboratory experiment to test the mediating role of shared psychological ownership and self–brand connection. Together, this set of studies aimed to provide rigorous and converging evidence for both the effectiveness of meme marketing and the psychological processes that underlie its impact.
Study 2a Method
Design
We conducted a field experiment in collaboration with a leading fresh grocery e-commerce company in China, Dingdong (Cayman) Limited, leveraging the popular meme at the time, “Tell Laomo, I want to eat fish” (see Web Appendix L1–1). An A/B test (meme marketing: meme vs. nonmeme) was implemented on WeChat Moments (see Web Appendix L1–2), a social media platform integrated within the WeChat ecosystem that enables users to share status updates, alongside sponsored advertisements (Manners 2021). A pretest showed that consumers were familiar with both the brand and the meme (see Web Appendix L1–3).
Procedure
Two versions of freshwater fish ads were developed: one using the meme's ad tagline (i.e., “Laomo, I want to eat fish tonight”) and the other using a regular ad tagline (i.e., “High-quality fresh freshwater fish”) (see Web Appendix L1–2). All other information presented in these two ad materials were identical. Another pretest revealed that consumers could recognize meme marketing, indicating a successful manipulation (see Web Appendix L1–3). If consumers clicked the ads, they were redirected to the product purchase page.
We allocated a budget of RMB 5,000 to each version of the two ads. We paid for WeChat Moments advertising based on cost per click (CPC), a commonly used charging method in digital advertising (Paharia 2020). CPC was budgeted for a certain number of clicks; the number of impressions the advertisement receives to get the targeted number of clicks varies based on its attractiveness. The ad that requires fewer impressions to generate clicks is regarded as more effective under the CPC charging structure.
For dependent variables, we chose three metrics commonly used to gauge the effectiveness of digital marketing campaigns (Scupola 2021). The first was advertising CTRs, which indicate the percentage of users who clicked on the ad (Kupor and Laurin 2020). The second metric was advertising CRs, which measure the percentage of users who place an online order to purchase the company's products. The third metric was ROAS, which measures the revenue generated for every dollar spent on an advertising campaign (Amazon Ads 2023).
Study 2a Results
After the budgets for both ads were completely spent, we recorded the number of impressions, clicks, and orders, as well as the total revenue for each ad. The two ads generated 423,565 impressions, 3,803 of which were clicked on (.90% CTR), 6 and 268 of which resulted in placed online orders (7.05% CR), resulting in a total revenue of RMB 13,322. We ran a logistic regression of whether users clicked on the ad (coded as 0 = did not click on ad, 1 = clicked on ad) on meme marketing (coded as 0 = nonmeme marketing, 1 = meme marketing). The results indicated that meme marketing significantly increased CTRs (1.40%) compared with nonmeme marketing (.65%; Wald χ2 = 559.17, SE = .03, p < .001, Exp(B) = 2.16). We ran another logistic regression in which we regressed whether users placed an online order after clicking the ad (coded as 0 = did not place an order, 1 = placed an online order) on meme marketing (coded as 0 = nonmeme marketing, 1 = meme marketing). Similarly, meme marketing significantly increased CRs (8.27%) compared with nonmeme marketing (5.75%; Wald χ2 = 81.63, SE = .13, p < .001, Exp(B) = 1.48). The ROAS for the meme marketing campaign was 1.52, higher than that for the nonmeme marketing campaign (1.14). Table 3 summarizes the results.
Performance Metrics for WeChat Moments Ad Campaigns.
Study 2b Method
Participants and design
After preregistering the study (https://aspredicted.org/6gcw-twx6.pdf), we recruited 1,000 participants from Credamo (58.6% female, 41.1% male, .3% preferred not to say; Mage = 28.82 years). This study used a two-cell (meme marketing: meme vs. nonmeme) between-subjects design.
Procedure and pretest
To enhance ecological validity and minimize demand effects, we developed a social media feed that emulates immersive browsing experiences. Similar to other real platforms (e.g., Facebook, X), our feed allows users to seamlessly scroll through vertically oriented content that mixes regular and meme-related posts, click the hyperlink to redirect to the meme homepage, engage with posts by liking and sharing, and create their own posts. 7 Specifically, participants were first introduced and instructed to evaluate a new social media site called Fushengyiji (which means “Notes from Everyday Life”). Next, participants were directed to one of two versions of the social media corresponding to their assigned conditions and instructed to view the posts on the website that were formed into a scrollable feed. While the contents of posts were kept identical across the two conditions, one-third of the posts (i.e., 34%, a similar percentage to some real-life memes in their peak popularity) in the meme marketing condition were rephrased in a meme style by following a consistent template (i.e., “Regarding…, my body refused to…, but chose to…”) (see Web Appendix L2–1).
Next, participants were informed that Fushengyiji might include a sponsored advertisement within its feed. They were shown an in-feed ad for a fictitious brand, Baoyaobao, promoting back support products (see Web Appendix L2–2). The ad was identical across both conditions. After viewing the ad, participants reported their likelihood of engaging with the ad (i.e., “How likely is it that you would hit the ‘like’ button for this post?”; 1 = “very unlikely/highly impossible,” and 9 = “very likely/highly possible”; r = .93) and their purchase intentions toward the promoted products (i.e., “How interested are you in trying Baoyaobao's back support product?” and “How likely are you to use Baoyaobao's product?”; 1 = “very unlikely,” and 9 = “very likely”; r = .89). A pretest (N = 200) revealed no significant differences across conditions in participants’ perceptions of the browsing experience and the quality of the posts, or brand familiarity. In addition, 98.0% of participants who viewed the meme-embedded social media site (i.e., the meme marketing condition) correctly identified the meme without being explicitly informed that it was a meme. This is important because memes will not be explicitly introduced in the current study, and consumers need to be aware of the content's meme nature for its effects to emerge (see Web Appendix L2–2 for details).
In brief, we manipulated meme marketing by creating a meme (i.e., incorporating a meme template into some of the social media posts), which was then used in the marketing post. As a manipulation check, participants were then provided with a brief explanation of memes and meme marketing and were asked to indicate the extent to which they believed the social media posts represented meme marketing (1 = “not at all,” and 9 = “very much so”). At the end of the survey, participants reported their mood (α = .93), gender, and age.
Study 2b Results
Manipulation check
A one-way ANOVA showed that the post in the meme (vs. nonmeme) marketing condition was perceived to fit the meme marketing definition more (Mmeme = 6.62, SD = 2.08 vs. Mnonmeme = 5.05, SD = 2.35; t(998) = 11.16, p < .001, d = .71).
Likelihood of liking the brand post
Participants in the meme marketing condition reported significantly higher likelihood of liking the brand post (M = 5.40, SD = 2.47) than participants in the nonmeme marketing condition (M = 4.60, SD = 2.55; t(998) = 5.01, p < .001, d = .32).
Purchase intention
Participants in the meme (vs. nonmeme) marketing condition indicated higher purchase intentions than did those in the nonmeme marketing condition (Mmeme = 6.08, SD = 2.07 vs. Mnonmeme = 5.60, SD = 2.11; t(998) = 3.58, p < .001, d = .23).
Additional analysis
Mood did not differ across conditions (Mmeme = 6.69, SD = 1.68 vs. Mnonmeme = 6.53, SD = 1.79; t(998) = 1.48, p = .140, d = .09). Adding mood as a covariate did not change the effect of meme marketing on participants’ likelihood of liking the post (F(1, 997) = 23.14, p < .001,
Study 2c Method
Design and participants
Six hundred U.S. participants from Prolific (62.7% female, 35.0% male, 1.8% nonbinary, .5% prefer not to say; Mage = 40.94 years) participated in this study, which used a two-cell (meme marketing: meme vs. nonmeme) between-subjects design.
Procedure
Participants first read a brief introduction of a movie about friendship, which ended, “There are five types of smiles: the shy grin, the joyful beam, the mischievous smirk, the comforting smile, and the one that says, ‘Stay with me awhile.’” Next, they were randomly assigned to read either a social network screenshot of netizens leaving their own versions of “There are five types of…” (meme marketing condition) or a screenshot of netizens leaving their opinions about the plot of the movie in general (nonmeme marketing condition; see Web Appendix L3). Then, on the next page, we presented a post from a fictitious athletic apparel brand “Fittros,” which stated: “There are five types of fitness enthusiasts in the world: cardio kings, yoga sages, weightlifting warriors, marathon maestros, and those who seamlessly blend style and comfort, behold Fittros, a fitness acclaim.” Similar to Study 2b, although the content of the marketing posts in the two conditions was identical, we expected that participants who saw netizens discussing and creating their own variations of the line would perceive the brand post as an instance of meme marketing, whereas the participants in the other condition would not.
Next, we measured participants’ purchase intentions (i.e., “How much would you like to try the service?” “How possible would you like to use the service?”; 1 = “very unlikely/highly impossible,” and 9 = “very likely/highly possible”; r = .89), shared psychological ownership (e.g., “Our netizens are one of the owners or creators of this social media post”; 1 = “strongly disagree,” and 9 = “strongly agree”; α = .97; Web Appendix G; adapted from Peck et al. [2021]), and self–brand connection (e.g., “I feel a personal connection to Fittros”; 1 = “not at all,” and 9 = “very much so”; α = .95; Web Appendix G; adapted from Escalas and Bettman [2003] and Moore and Homer [2008]). Last, as a manipulation check, we explained to participants what meme marketing is and asked them to what extent they thought the social media post was an instance of meme marketing (1 = “not at all,” and 9 = “very much so”).
Study 2c Results
Manipulation check
A one-way ANOVA showed that the post in the meme (vs. nonmeme) marketing condition was perceived to fit the meme marketing definition more (Mmeme = 7.18, SD = 2.14 vs. Mnonmeme = 5.18, SD = 2.53; t(598) = 10.46, p < .001, d = .85).
Purchase intention
Participants in the meme marketing condition indicated higher purchase intention than did those in the nonmeme marketing condition (Mmeme = 4.79, SD = 2.24 vs. Mnonmeme = 4.20, SD = 1.87; t(598) = 3.54, p < .001, d = .29).
Shared psychological ownership
Participants in the meme (vs. nonmeme) marketing condition indicated a stronger sense of shared psychological ownership toward the post (Mmeme = 4.98, SD = 2.21 vs. Mnonmeme = 3.97, SD = 2.12; t(598) = 5.75, p < .001, d = .47).
Self–brand connection
Participants in the meme marketing condition also indicated a stronger self–brand connection with the fictitious brand (Mmeme = 3.45, SD = 2.10 vs. Mnonmeme = 2.52, SD = 1.68; t(598) = 5.99, p < .001, d = .49).
Sequential mediation
Results from PROCESS Model 6 indicated significant effects of meme marketing on shared psychological ownership (b = 1.02, SE = .18, t(598) = 5.75, p < .001), of shared psychological ownership on self–brand connection (b = .43, SE = .03, t(597) = 13.75, p < .001), of self–brand connection on purchase intention (b = .74, SE = .03, t(596) = 22.44, p < .001), and, most importantly, significant sequential mediation effects of meme marketing on purchase intention through shared psychological ownership and self–brand connection (effect = .32, SE = .07, 95% CI: [.1998, .4552]).
Discussion
The experimental studies in Study 2 collectively provide robust causal evidence for the main effect (H1) and its underlying mechanism (H2). Study 2a tested meme marketing in a real-world digital advertising context using actual ads that leveraged a widely recognized meme and behavioral data from real products purchases. The results showed that meme marketing outperformed nonmeme marketing on key business outcomes. Interestingly, despite having fewer impressions, meme marketing generated comparable or higher clicks, suggesting that memes capture consumer attention more efficiently under a CPC model (Paharia 2020).
To address the possible limitations in Study 2a (e.g., lack of exclusivity, nonrandomized assignment on established social media platform; Braun et al. 2024; Roggenkamp, Boegershausen, and Hildebrand 2025), Study 2b introduced a custom-built, simulated social media platform that allowed for full randomization of participants and controlled exposure to ads. This design ensured that participants could recognize a fictitious meme embedded in a naturalistic scrolling feed without being explicitly informed that it was a meme. Moreover, by using identical content framed either as a meme or as a standard brand post, we isolated the effect of meme presentation from differences in ad content. Results again demonstrated the effectiveness of meme marketing.
Building on these effects, Study 2c moved into a tightly controlled laboratory environment to examine the psychological process underlying the impact of meme marketing. This study provided direct evidence for the proposed mechanism by showing that meme marketing enhances purchase intention through increased perceptions of shared psychological ownership of the marketing message, which in turn strengthens consumers’ self–brand connection. Finally, as additional empirical evidence, Supplementary Study B in Web Appendix I confirmed the robustness of the proposed mechanisms and further evaluated meme marketing's uniqueness by comparing it with other widely used digital marketing strategies.
Taken together, these studies move from externally valid, real-world evidence to more controlled and internally valid settings, progressively addressing potential alternative explanations and deepening our theoretical understanding of meme marketing.
Study 3: Moderation of Meme Life Cycle Stage
Study 3 explores the moderating role of the meme life cycle (H3). As discussed previously, the life cycle of an internet meme can be divided into four stages (i.e., introduction, growth, maturity, and decline; see Web Appendix E for a detailed analysis). Our theory suggests that meme marketing would be most effective when memes are in their maturity stage, compared with the other stages. For the memes in the maturity stage—in terms of peak popularity, widespread use, understanding, and remixing—their features along four core aspects create a foundation for consumers to perceive the meme as a collectively “owned” cultural artifact. When brands leverage such memes, consumers’ preexisting psychological ownership toward the meme transfers to the marketing message, fostering a sense of shared ownership and strengthening self–brand connections. In contrast, when brands leverage memes from other stages, consumers are less likely to feel a strong sense of ownership toward these memes. The meme marketing message thus cannot generate a shared psychological ownership among consumers, thereby attenuating the proposed effect. Study 3 also rules out alternative explanations of narrative transportation (Razzaq, Shao, and Quach 2024) and emotional attitudes toward the meme (Shen, Lee, and Lin 2024).
Method
Participants and design
After we removed 34 participants who did not answer the attention check question correctly, 1,561 U.S. participants from Prolific (58.6% female, 39.7% male, 1.5% nonbinary, .3% prefer not to say; Mage = 40.59 years) were included in the final sample of this study. This study adopted a 2 (meme marketing: meme vs. nonmeme) × 4 (meme life cycle stage: introduction vs. growth vs. maturity vs. decline) between-subjects design.
Procedure and pretest
Participants first read a brief introduction to a fictitious TV series that sparked fictitious spreadable media content, “Three Things,” and then a paragraph and some pictures that showed the discussions on the internet around this topic (see Web Appendix M1). The discussions were varied to manipulate different meme life cycles: In the introduction stage, participants read, “There hasn’t been anyone developing or creating social media posts based on the contents in the movie (e.g., ‘only three things’)”; in the growth stage, participants read “Some viewers of the TV series are sharing this line with others, and a few are starting to add their own twist. These adaptations are still limited but signal growing engagement.” In the maturity stage, participants read, “This line has evolved into a popular meme, known as ‘Only Three Things.’ A significant number of accounts you follow have embraced it, using the same sentence structure to create their own versions. This meme has become widely understood, reflecting shared experiences across different contexts.” In the decline stage, observed participants were told that the meme has lost its appeal, and few users engage with it anymore.
Participants then saw and evaluated one of two different versions of a marketing post from a fictitious bedding brand, CozyDreams; one version of the post remixed the product information with “Only Three Things” (i.e., “I come to our beddings for only three things—cozy, cozy, and cozy.”), mimicking brands’ meme marketing efforts in the real world; the other version of the post had a line that was irrelevant to the movie or “Only Three Things” (i.e., “Sleep tight, feel light, with bedding just right.”), reflecting a general nonmeme marketing post. Results from a pretest indicated that these manipulations were successful (see Web Appendix M2).
After reading the stimuli, we measured purchase intention (r = .87), shared psychological ownership toward the marketing post (α = .97), and self–brand connection (α = .95). We also measured alternative explanations of narrative transportation (e.g., “While thinking about this social media post, I could easily imagine myself in the cozy environment it described”; 1 = “not at all,” and 9 = “very much so”; α = .89; adapted from Razzaq, Shao, and Quach 2024) and emotional attitude toward the meme (i.e., “This social media post (meme) is pleasant/good”; 1 = “not at all,” and 9 = “very much so”; r = .87; adapted from Shen, Lee, and Lin [2024]).
Results
Purchase intention
A 2 × 4 ANOVA revealed a significant interaction (F(3, 1,553) = 2.93, p = .032,
Self–brand connection
A 2 × 4 ANOVA revealed a significant interaction (F(3, 1,553) = 2.83, p = .038,
Shared psychological ownership
A 2 × 4 ANOVA revealed a significant interaction (F(3, 1,553) = 2.67, p = .046,
Discussion
This study provides empirical evidence for H3, demonstrating that meme marketing elicits its proposed effect only when memes are deployed in their maturity stage, with no significant effects observed in the introduction, growth, or decline stages. These findings underscore the critical role of timing in meme adoption, suggesting that brands must select a meme in the right life cycle stage to maximize the meme marketing effect. Nonetheless, we acknowledge that the stimuli employed to manipulate different meme life cycle stages in this study are somewhat artificial. In real-world contexts, internet users usually assess a meme's life cycle stage through multiple browsing experiences over time and across diverse sources. Our experimental manipulation served as a proxy for such assessments. Future research could explore the meme life cycle in more naturalistic settings to enhance ecological validity.
Study 4: Moderation of Product Type
Study 4 explores the moderating role of product type (H4) on meme marketing. We speculate that meme (vs. nonmeme) marketing leads to higher purchase intention when the promoted products are mainstream products (e.g., a popular movie); however, this effect dissipates when the promoted products are niche products (e.g., an artistic movie). To test the generalizability of the meme marketing effect across different modalities, we used a video-based meme instead of the text- or image-based memes used in previous studies. To increase realism, we adopted a real meme (“Two cats conversation”) that originated in China but was still in its introduction stage in the United States when this study was conducted (March 2024). This allowed us to use the same manipulation method as in previous studies, controlling whether the video is perceived as a meme while keeping the marketing message consistent.
Method
Participants and design
After preregistration (https://aspredicted.org/Y43_KTF) and removal of three participants who failed the attention check, 797 U.S. participants from Connect (52.9% female, 45.3% male, 1.4% nonbinary, .4% prefer not to say; Mage = 40.42 years) completed this study for a nominal payment. The study adopted a 2 (meme marketing: meme vs. nonmeme) × 2 (product type: mainstream vs. niche) between-subjects design.
Procedure
To manipulate meme marketing, participants were first presented with four short videos. In the meme marketing condition, each video featured two cats having conversations on various topics, all of which were variations of the “Two cats conversation” meme. In contrast, in the nonmeme marketing condition, the four short videos simply featured daily activities of different cats (see Web Appendix N). Next, participants were presented with an introduction to a fictitious movie, “Echoes,” which was described as an action-adventure movie (i.e., “It's supposed to be an awesome movie for everyone”) in the mainstream product condition or an artistic movie (i.e., “It's supposed to be something really different”) in the niche product condition. Next, all the participants saw the same video-based social media post, which combined the “Two cats conversation” meme and the information about the movie “Echoes.” The participants were then asked to indicate their intention to watch the movie (i.e., “How likely are you to watch the movie ‘Echoes’?” and “How interested are you in watching the movie ‘Echoes’?”; 1 = “very unlikely/not interested,” and 9 = “very likely/very interested”; r = .93). Next, the participants were asked to respond to manipulation checks about meme marketing (using the same question as in previous studies) and the product type (e.g., “The movie ‘Echoes’ is designed to cater to a specialized segment of the movie market”; Zheng et al. 2023). Lastly, the participants were asked to indicate their previous exposure to the related videos of “Two cats conversation” (i.e., “Have you watched the video featuring the dialogue between two cats before participating in this survey?”; “No/Yes/Not Sure”) and their individual chronic preference for artistic movies (e.g., “I actively seek out and watch movies that are not mainstream”; 1 = “strongly disagree,” and 9 = “strongly agree”; α = .91; Schaefers 2014).
Results
Manipulation check
A 2 × 2 ANOVA on meme manipulation check revealed only a significant main effect of meme conditions (Mmeme = 7.24, SD = 2.01 vs. Mnonmeme = 6.82, SD = 2.03; F(1, 793) = 8.78, p = .003,
Purchase intentions
A 2 × 2 ANOVA on intention to watch the movie revealed significant main effects of meme conditions (F(1, 793) = 5.33, p = .021,
Previous meme exposure and chronic preference toward niche products
While 23.6% of participants indicated they previously had seen videos related to the “Two cats conversation” meme, such previous meme exposure did not moderate the interaction effect of meme marketing and product type (three-way interaction: F(2, 785) = 1.33, p = .266,
Discussion
Supporting H4, the results of this study replicated the meme marketing effect for the mainstream product, but the effect was dissipated for the niche product. The results of this study implied that brands should consider whether the type of promoted product is compatible with the nature of the meme, such that meme marketing might not be suitable for products that are tailored to the tastes of a narrow group of people. A point worth noting is that the manipulation of niche products in this study may evoke uniqueness-seeking tendencies among certain consumers. Although we focus on product type as a moderator from a product–meme incompatibility perspective in the current research, future studies could examine the incompatibility between uniqueness-seeking consumers and the mass-cultural nature of internet memes, a moderator from a consumer–meme incongruity perspective.
General Discussion
Memes have emerged as a powerful asset in digital marketing, offering a novel and engaging way to connect with consumers (Krishnamurthy and Abhi C 2025). Although marketers have embraced meme marketing, it has not received equivalent attention in the marketing literature. To address this gap, we explored how meme marketing influences digital advertising effectiveness. Using a multimethod approach—including secondary data (Study 1), a field experiment (Study 2a), and laboratory experiments (Studies 2b and 2c)—we demonstrated that meme marketing fosters a sense of shared psychological ownership toward the marketing message, which in turn strengthens self–brand connection and enhances the effectiveness of digital advertising. Consistent with this theory, our findings suggest that this effect is moderated by the meme life cycle stage (Study 3) and niche versus mainstream products (Study 4). Studies 2b, 2c, and 4 used identical content across conditions to rule out potential content-related confounds, manipulating meme marketing by simulating a real-world memetic process. Among them, Study 2b manipulated meme marketing without explicitly indicating the meme. A posttest confirmed that the stimuli varied in humor levels (see Web Appendix O), yet the meme marketing effect remained robust. These findings yield important theoretical insights and practical implications.
Theoretical Contributions
Despite the rapidly growing meme economy, research on memes in the business domain is still in its nascent stage (see Table 1 for a literature review). The present research thus adds to this emerging research stream by identifying the unique role of meme marketing in boosting digital marketing effectiveness. Specifically, we demonstrated that social media posts adopting memes can effectively drive likes, purchase intention, and sales indices compared with posts that do not use memes. More importantly, the current research conceptualizes and supports with empirical evidence that meme marketing leads to a sense of shared psychological ownership of meme marketing messages and a subsequent enhanced self–brand connection, which in turn boosts the effectiveness of digital advertising. This research also lays the groundwork for future investigations into meme marketing. Specifically, by situating meme marketing within a holistic communication framework (source, message, channel, receiver) and outlining potential moderating factors, our work offers a conceptual roadmap that future research can build on, as depicted in Figure 1.
By examining meme marketing through the theoretical lens of psychological ownership, the current research also contributes to related streams of research. While consumers have been found to generate psychological ownership of a wide range of targets, including purchased products (Gineikiene, Schlegelmilch, and Auruskeviciene 2017), public facilities (Kirk, Peck, and Swain 2018), and digitized goods and services (Morewedge et al. 2020), recent research has indicated that consumers can also glean psychological ownership over user-generated content on digital media (Zhang and Hu 2024). The current research identifies the core four aspects of memes, which involve collective creation by netizens, use of an easily recognizable and repeatable template, conveyance of cultural resonance contents, and wide diffusion through continuous remixing. These core aspects help internet users foster a sense of psychological ownership toward memes and later transfer as a shared psychological ownership toward meme marketing posts generated by brands. This associated psychological process distinguishes it from other commonly used digital marketing practices (e.g., influencer marketing; see Web Appendix D and Supplementary Study B in Web Appendix I for details). Similarly, traditional marketing tools—such as ads featuring testimonials—may involve individual consumer participation but lack collaborative content creation and are therefore less likely to evoke the shared psychological ownership that meme marketing evokes.
Furthermore, our study aligns with the broader research agenda of extending the concept of psychological ownership to collective levels (Hulland, Thompson, and Smith 2015), as we identify psychological ownership perceptions of the meme. Specifically, extending prior research showing that investing effort, controlling, and getting to know a target intimately can contribute to psychological ownership (Kirk, Peck, and Swain 2018), the current research concludes that the joint effort in creating memes, controlling their production with templates, and the collective knowledge among netizens build up psychological ownership of a meme. Finally, our findings add to the downstream effects associated with psychological ownership (Jussila et al. 2015), including consumer attitudes and behaviors, such as social media interactions (e.g., liking posts) and purchase intentions toward products featured in digital marketing content.
Concerning brand research, we identify meme marketing and shared psychological ownership as novel antecedents by which brands strengthen their connection to consumers. Existing research has acknowledged that self–brand connection is an important psychological manifestation of brand equity (Escalas and Bettman 2003) and has noted its positive effects on brand attitudes (Moore and Homer 2008) and purchase intentions (Escalas 2004). Identifying antecedents of such a valuable consumer–brand relationship proxy thus remains a meaningful goal for brand literature. Adding to prior findings showing that narrative processing (Escalas 2004), advertising (Trana et al. 2020), and celebrity endorsements (Ilicic and Brennan 2020) can increase self–brand connections, we show that meme marketing can boost consumers’ self–brand connection through a shared sense of psychological ownership toward the marketing message.
Practical Implications
The simplicity and affordability of memes make them an excellent vehicle for promoting marketing messages (Krishnamurthy and Abhi C 2025). As a relatively “soft” digital marketing tactic, meme marketing enables brands to engage customers without being too promotional (Chopra 2024), making it an accessible addition to the digital marketing toolkit. In addition, under the frameworks of owned media (i.e., brand-controlled channels for direct communication) and earned media (i.e., organic exposure generated through consumer engagement) (Colicev et al. 2018), meme marketing is distinguished from other digital media tactics by its ability to develop both. Brands may create and distribute meme marketing posts through their channels (owned media) while encouraging organic sharing and remixing of the meme marketing posts among netizens, driving organic reach and engagement (earned media).
Our field study also highlighted the economic significance of meme marketing in enhancing digital marketing effectiveness. Specifically, Study 2a used a real-world field setting and demonstrated that meme marketing can substantially boost both CTR in digital ads and, more importantly, the ROAS—a key ROI metric for advertising. Study 2a revealed that meme marketing increases the CTR by 115%, from .65% to 1.40%. Notably, the industry average CTR is .60%, indicating that the control condition in our study aligns closely with this benchmark (Statsig 2024). Moreover, ROAS improves from 1.14 to 1.52 with the adoption of meme marketing. Considering that firms are projected to collectively spend approximately US$558 billion on digital advertising in 2026 (Statista 2024), a simple calculation suggests that meme marketing could generate an additional US$212 billion in revenue. While the accuracy of this extrapolation depends on various factors (e.g., industry specifics and competitive dynamics), these findings underscore the substantial financial potential of meme marketing—especially given its small implementation costs.
Our research also uncovers the subtleties of meme marketing effectiveness and delineates boundary conditions that marketers must consider when implementing meme marketing strategies. First, regarding the meme life cycle, marketers should prioritize using mature memes over memes in the introduction, growth, and decline stages. This preference stems from the fact that memes in the other stages lack sufficient features for consumers to develop a sense of psychological ownership toward them. Notably, marketers should evaluate mature memes beyond popularity metrics by considering creator engagement, template consistency, and how familiar and recognizable the meme content is to the target audience. Second, mainstream products are more likely than niche products to derive greater benefits from meme marketing, due to the mismatch between the mass appeal of memes and the narrow focus of niche products.
Limitations and Future Research
As our research is among the very first to investigate meme marketing effectiveness empirically, we admit the following limitations that arise with this first attempt to examine this understudied digital marketing practice, which, however, may stimulate valuable future research.
First, given that marketing practitioners may utilize more than one type of these marketing practices simultaneously, it would be interesting to investigate the synergy of meme marketing and other broader corporate practices. For example, when an influencer/consumer promotes the company's products by leveraging a meme, we can treat them as a synergistic practice of meme marketing and influencer/eWOM marketing. The effectiveness of these synergistic digital marketing practices may involve the interplay between the meme and the specific characteristics of an influencer or the eWOM message. Additionally, companies could probably benefit more by combining meme marketing with other traditional marketing strategies according to brand characteristics, for example synergizing with celebrity-endorsed TV commercials for brands that hope to leverage the support from both mass consumers and celebrities. However, the nuances of the effectiveness and implementation of these combinations are worth extensive examination in future research.
Second, while our experiments used memes with different modalities (e.g., text, image, video) to demonstrate the robustness of the meme marketing effect, many other types of memes still deserve further investigation. For example, media communication scholars have argued that TikTok's design and logic encourage the memetic process of imitation and replication, thereby extending internet memes to the platform's infrastructure level (Zulli and Zulli 2022). When a brand participates in a popular TikTok challenge or leverages memes produced by the challenge, it can be regarded as an instance of meme marketing. However, it's important to note that TikTok challenges are platform-specific phenomena. Many operational nuances related to platform features may play complex roles, which could be a future research direction.
Third, promising research avenues exist for exploring nuances in meme marketing practices. In Figure 1, we synthesize insights from prior literature and introduce several promising psychological processes and contextual factors that may serve as moderators for future research. Special attention should also be given to various factors pertaining to the characteristics of source, message, channel, and receiver in the meme transmission process, which could moderate meme marketing effectiveness. Memes addressing sensitive or controversial topics, such as those with political content, may require careful consideration and nuanced investigation when leveraged by firms (message). Meme marketing effectiveness might be attenuated if a brand's identity or social media persona is not harmonious with the meme it leverages (source) or if a brand already has a good communal relationship with consumers online (receiver). Marketing-related factors such as the alignment of meme marketing with a brand's overall strategy and its synergistic effect with other digital strategies (message), as well as broader societal and internet factors, including social inclusiveness and internet censorship policies (channel), may significantly influence the effectiveness of meme marketing campaigns. For instance, high social inclusiveness may foster a stronger sense of connectedness to other netizens and thus enhance meme marketing effectiveness.
Supplemental Material
sj-docx-1-jmx-10.1177_00222429261417674 - Supplemental material for From Owning to Connecting: Understanding and Leveraging the Effect of Internet Meme Marketing
Supplemental material, sj-docx-1-jmx-10.1177_00222429261417674 for From Owning to Connecting: Understanding and Leveraging the Effect of Internet Meme Marketing by Lu Wang, Xueni (Shirley) Li, Qiyuan Wang and Lei Su in Journal of Marketing
Footnotes
Coeditor
Shrihari Sridhar
Associate Editor
Dhruv Grewal
Acknowledgments
The authors thank the review team, colleagues, conference and seminar participants for their constructive feedback, which greatly improved this work. They also thank Dingdong (Cayman) Limited for supporting the field experiment. All errors and omissions remain the authors’ responsibility. The corresponding authors of this publication are Lei Su and Xueni (Shirley) Li.
Author Contributions
The four authors contributed equally and share first authorship.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This research is supported by grants from the National Natural Science Foundation of China (72002222), the Alumni Fund of the School of Business Administration, Zhongnan University of Economics and Law, awarded to the first author, and the Hong Kong SAR Research Grants Council, awarded to the second (HKBU22500521 and HKBU12500723) and the last author (CityU12501722, CityU11509323, CityU11506124, and NSFC72372142).
Data Availability
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
