1. Introduction

In an age where grief is live-streamed, private moments are transformed into monetized content, and casual selfies are algorithmically curated for public consumption, the line between frontstage and backstage has become increasingly porous. The late 20th century sociologist Erving Goffman [1] provided one of the most enduring metaphors in social science: life as a stage, where individuals perform roles before an audience in various social contexts. His dramaturgical model distinguished between the “frontstage”—the public arena where impression management occurs—and the “backstage”—a private space for preparation, relaxation, and the suspension of performance.

Goffman’s theory was developed in an era when audiences were physically co-present, when the timing and scope of self-presentation were largely under an individual’s control, and when personal performances could be withdrawn from public view with relative ease. The digital age, however, presents a new set of challenges. Social media platforms encourage constant sharing, instant visibility, and persistent archiving of self-presentations. The curated images, captions, and videos that populate Instagram, TikTok, and Rednote profiles are often perceived as spontaneous, but they are highly mediated by both the user’s choices and the platform’s algorithms.

This paper explores a central question: Does the backstage still exist in the age of social media? Or, has it been subsumed into an endless frontstage where the performance never truly ends? The discussion is grounded in Goffman’s dramaturgical theory, recent scholarship on digital self-presentation, and Foucault’s ideas on disciplinary power. This study argues that while Goffman’s framework remains insightful, it requires significant adaptation to account for the algorithmic governance, invisible audiences, and enduring data traces that define contemporary online life.

2. Literature review

2.1. Goffman’s dramaturgical model

In The Presentation of Self in Everyday Life, Goffman [1] likens everyday interaction to theatrical performance. Individuals occupy a frontstage, where they actively manage impressions to influence others’ perceptions, and a backstage, where they can step out of character. Central to his model is impression management, the conscious process of shaping how one is perceived. Importantly, Goffman rejects the notion of a single “true self”, instead viewing identity as fluid, situational, and co-constructed in social interaction.

2.2. Oversharing and the qualified self

Ben Agger [2] introduces the concept of “oversharing” to describe the culture of pervasive self-disclosure on the internet. In this context, even experiences of grief, illness, and personal hardship become elements of curated online identities. Lee Humphreys [3] extends this analysis with the notion of the “qualified self”, a form of selfhood that is not only internally motivated but increasingly shaped by external metrics such as likes, shares, and follower counts. These metrics act as feedback loops that influence future performances.

2.3. Disciplinary power in the digital context

Michel Foucault’s [4] theory of disciplinary power offers another lens. His work in Discipline and Punish examines how surveillance—both visible and invisible—produces self-regulation. In the digital environment, surveillance extends beyond human observers to include automated systems that track, store, and analyze user behavior. This pervasive observation challenges Goffman’s assumption that backstage spaces are private and protected.

2.4. Algorithmic mediation

Recent studies [5,6] emphasize that social media platforms do not passively host content; they actively curate visibility through algorithmic processes. These algorithms function as unseen directors, determining what is amplified and what remains hidden. As a result, users perform not only for their human audiences but also for platform logics that reward certain forms of engagement over others.

3. Theoretical analysis

3.1. Revisiting the frontstage/backstage distinction

Goffman’s [1] metaphor of everyday life as theater offers a compelling starting point for understanding self-presentation in the digital age. In traditional settings, the frontstage was the realm of public interaction, characterized by deliberate impression management, while the backstage was a space where performers could drop their public roles. In face-to-face interactions, this division was supported by physical boundaries—closed doors, private rooms, and intimate circles of trusted confidants.

In the social media environment, however, these boundaries are blurred. A post intended for a small circle of friends can be shared, screenshotted, or algorithmically promoted to strangers. Even “private” online spaces, such as direct messages or private groups, can be compromised by leaks or data breaches [7]. As a result, the backstage is not a clearly demarcated zone but a fragile, permeable space that requires constant negotiation.

On Rednote, many users cultivate a “refined persona” by sharing skincare routines, travel experiences, or café check-ins. Though this persona appears to showcase authentic daily life, it is in fact highly curated and embellished [8]. This persistent performance on the “front stage” often obscures users' “backstage” realities. When authentic life clashes with their online personas, it triggers public incidents of “persona collapse”, revealing the fragility of these carefully constructed personas.

3.2. The collapse of temporal boundaries

Goffman’s original framework also assumes a temporal rhythm to performances: one can step into the frontstage during a social interaction and retreat to the backstage afterward. Social media disrupts this rhythm by encouraging asynchronous, perpetual performance. Posts, stories, and live streams can be accessed long after their creation, creating what Mayer-Schönberger calls the “persistence of memory” in digital archives [9]. This temporal collapse means that even when the performer is “offstage” in real life, their digital persona remains active, accessible, and subject to interpretation by audiences unknown to them.

3.3. Algorithmic audiences as new directors

A striking development in the digital age is the emergence of algorithms as backstage managers. Unlike human stage managers who assist performers, algorithms govern visibility without negotiation or transparency [5]. These computational processes decide what appears in newsfeeds, which posts trend, and how audiences encounter content. The performer’s role is thus reframed: rather than crafting a performance solely for human perception, they must optimize for algorithmic logics such as engagement rates, click-throughs, and watch times.

This optimization often shapes not only content but also selfhood. As Bishop notes in her study of YouTube influencers, creators routinely alter their personal narratives, speech patterns, and even moral stances to align with platform metrics [10]. In effect, algorithms do not merely mediate performances; they co-produce them.

TikTok’s recommendation mechanism exemplifies the algorithm's role as a “director”. Ordinary users often achieve rapid fame by participating in dance challenges or mimicking viral memes, thereby entering the public eye [11]. Such sudden prominence relies not only on the performance itself but crucially on the algorithm’s distribution logic. Unlike Goffman's hypothetical face-to-face audience, today’s “audience” encompasses both the platform and the algorithm, compelling users to tailor their performances to the algorithm's preferences.

3.4. The expansion of the audience

In Goffman’s model, audiences are identifiable and bounded: a theatergoer is physically present in a specific space. Online, the audience is potentially infinite, invisible, and unpredictable. Content can be consumed by strangers, employers, or automated surveillance systems. This “context collapse” forces individuals to perform for multiple, often conflicting, audiences simultaneously [12].

On Instagram, the “business” mode of celebrities and influencers is particularly pronounced. They maintain their public personas through meticulously curated posts, stories, and short videos [13]. Simultaneously, many young users create “Finsta” (fake Instagram accounts) to attempt rebuilding a backstage space, sharing more casual content exclusively with close friends. However, this backstage remains fragile and still carries the risk of exposure.

The uncertainty of the digital audience amplifies self-monitoring behaviors, leading to performances that are both hyper-curated and sanitized. Yet paradoxically, authenticity has become a desirable performance trait online, prompting users to stage “casual” or “imperfect” moments that are, in fact, meticulously constructed [14]. This blurring of authenticity and performance challenges Goffman’s notion of a backstage where the “real” self resides.

3.5. Persistent data traces and the end of erasure

One of the most profound changes to the dramaturgical metaphor in the digital age is the persistence of data. In traditional theater, a performance ends when the curtain falls, leaving only memories. Online, performances leave durable digital traces: images, captions, metadata, and interaction logs that may outlast the performer’s intention to withdraw. As Mayer-Schönberger argues, the inability to forget can be socially and psychologically constraining, as past performances are perpetually available for re-interpretation [9].

For impression management, this means the “repertoire” of a performer is never fully under their control. Historical posts can resurface unpredictably, affecting reputations, employment opportunities, and personal relationships. Thus, the digital backstage is not only spatially compromised but also temporally eroded.

4. Critical reflection

4.1. The limits of Goffman’s individual agency model

Goffman’s dramaturgical model is deeply rooted in the concept of agency—the idea that individuals actively and consciously manage their performances to achieve desired impressions. This presupposes a high degree of autonomy: the performer decides when, how, and for whom to perform. In the digital age, however, this autonomy is significantly constrained.

Social media platforms embed their own scripts into the infrastructure of interaction. Interface design choices—such as the placement of “like” buttons, algorithmically sorted feeds, and ephemeral story formats—guide user behavior toward certain modes of self-presentation [6]. The result is that impression management is no longer entirely self-directed; it is shaped by platform affordances and commercial imperatives. As van Dijck argues, platforms are not neutral; they are socio-technical systems with their own agendas, often oriented toward maximizing engagement for profit [15].

4.2. From managing impressions to being managed

The central question that emerges is whether users today are still managing impressions, or whether they are increasingly being managed. Goffman conceived of impression management as a strategic interaction between performer and audience, mediated by social norms. Today, however, the interaction is triangulated by platforms that operate as invisible third parties. Algorithms determine not only what is visible but also what is worth performing in the first place.

On platforms such as TikTok and Bilibili, so-called “daily vlogs” or “healing lifestyle” videos, though touted as authentic, often strictly adhere to algorithmic recommendation logic in their visual language, subject selection, and tagging [16]. Creators’ self-presentation has gradually shifted from active impression management to passively conforming to platform norms and data feedback.

This structural shift aligns more closely with Foucault’s concept of disciplinary power, wherein observation—whether real or potential—conditions behavior without the need for direct coercion [4]. The omnipresence of digital surveillance, from data analytics to facial recognition, encourages self-regulation in ways that are often unconscious. Even when users are “alone” with their devices, they are performing for an imagined audience, including algorithmic observers.

4.3. The problem of context collapse and authenticity

Another limitation of Goffman’s framework is its assumption of audience segmentation. In face-to-face settings, performers can tailor performances to specific groups. Online, context collapse collapses multiple audiences—friends, family, employers, strangers—into a single undifferentiated mass [12]. This complicates impression management, as the norms and expectations of different audiences may conflict.

Instagram's celebrity “behind-the-scenes glimpses” or Xiaohongshu bloggers' “makeup-free selfies” are often perceived as displays of “authenticity”. However, research indicates that this very “sincerity” is frequently a carefully calculated strategy employed to balance the expectations of diverse audiences [17]. This “strategic authenticity” reveals how users navigate complex relationships with multiple audiences when contextual boundaries break down.

One common response to context collapse is the cultivation of strategic authenticity. Abidin notes that influencers often stage “behind-the-scenes” moments to signal transparency, yet these moments are themselves calculated performances designed to meet audience expectations of relatability [14]. Such practices undermine the backstage as a private, authentic space, further challenging Goffman’s binary distinction.

4.4. The erosion of forgetting and its psychological consequences

In Goffman’s world, failed performances could be contained and forgotten. In the digital age, the persistence of data makes forgetting nearly impossible [9]. The constant possibility that past performances may resurface creates an environment of permanent self-surveillance, in which individuals curate not only the present self but also attempt to retroactively manage the past.

This constant vigilance has psychological consequences. Research by Chae shows that prolonged exposure to curated images of others can exacerbate self-comparison and anxiety [18]. The inability to retreat to a true backstage for emotional and cognitive respite can lead to burnout, especially among content creators and individuals whose livelihoods depend on maintaining an online persona.

4.5. Toward a post-Goffman framework

While Goffman’s dramaturgical model remains a powerful analytical tool, its emphasis on face-to-face interaction and human audiences limits its applicability in the platform-mediated digital age. A post-Goffman framework might integrate three critical dimensions: Algorithmic mediation—recognizing platforms as co-directors of performances; Surveillance assemblages—incorporating Foucault’s disciplinary power to account for the normalization of self-regulation under constant observation; Temporal persistence—acknowledging the erosion of temporal boundaries in self-presentation, where past and present performances coexist in digital archives.

This integrated approach can better account for the structural forces that shape online performances, the invisible actors influencing visibility, and the long-term consequences of perpetual publicness.

5. Conclusion

Goffman’s dramaturgical model continues to offer a compelling metaphor for understanding the performative nature of self-presentation. His insights into the frontstage, backstage, and impression management illuminate how individuals navigate social expectations through strategic role-playing. However, the digital age has transformed the stage in ways that Goffman could not have anticipated.

The rise of social media has blurred, if not collapsed, the distinction between frontstage and backstage. Platforms like Instagram, TikTok, and Rednote encourage constant visibility, reward engagement through algorithmic amplification, and preserve performances in persistent digital archives. In this environment, impression management is no longer solely a matter of individual agency; it is increasingly shaped by platform design, algorithmic logics, and pervasive surveillance.

Three major departures from Goffman’s original framework are highlighted.

First, Algorithmic Co-Production of Performances. Users now perform for both human audiences and computational systems, which invisibly govern the terms of visibility and relevance.

Second,Erosion of Privacy and Audience Boundaries. Context collapse forces individuals to cater to multiple, often conflicting, audiences simultaneously, undermining the sanctity of the backstage.

Third, Persistence of Digital Traces. The impossibility of erasure means that past performances remain perpetually available, complicating efforts to manage one’s long-term identity.

These developments have prompted a reconsideration of what it means to “perform” in the digital public sphere. If, in Goffman’s time, performers could step offstage and reclaim a private self, today’s digital performers inhabit a world where the curtain never falls. In such a world, the question “Are we managing impressions or being managed?” becomes not just rhetorical but deeply political, touching on issues of autonomy, surveillance, and digital rights.

Future research should explore emerging “backstage” spaces that resist algorithmic capture, such as encrypted messaging platforms, ephemeral content formats, and decentralized social networks. Additionally, interdisciplinary approaches combining sociology, media studies, and computer science are needed to fully grasp how algorithmic governance intersects with human self-presentation.

As social beings in a platform-mediated world, we are always performing—but increasingly, the performance is co-authored by systems we do not control. Goffman opened the door to thinking about selfhood as a series of performances; The task now is to step through that door and confront the new actors, scripts, and constraints that define the digital stage.