The group chat has become one of the primary sites of collective friendship maintenance in contemporary life, and almost no one talks about the work it takes to keep one alive. This is partly because the work doesn't look like work. You're just texting. You're just reacting to someone's meme with a laughing emoji. You're just dropping a link that seemed relevant. But behind these acts lies a continuous low-grade labor: reading the room, calibrating tone, tracking who hasn't spoken in a while, deciding whether to escalate a thread or let it die, managing the person who always derails, and holding in mind the relational history that makes any given exchange legible. This is relational labor, and in most group chats, it is distributed unequally.
The concept of emotional labor, developed by Arlie Hochschild in her study of flight attendants, describes the effort required to manage one's feelings in accordance with social expectations. Relational labor is adjacent but distinct: it is the work of managing relationships on behalf of a collective, maintaining the conditions under which connection can occur. In workplaces, this labor is done disproportionately by women, often unacknowledged in performance evaluations and rarely compensated. In group chats among friends, the same dynamic appears with remarkable consistency: the person who keeps the thread alive, who follows up when someone goes quiet, who holds the emotional tenor of the group, is usually the same person—and that person is usually not thanked for it because the labor is invisible.
The invisibility operates on multiple levels. It is invisible to the group because it looks like participation rather than maintenance. It is invisible to the person performing it because it has been internalized as disposition—caring, attentive, social—rather than recognized as effort. And it is invisible to the culture because digital friendship labor rarely appears in economic analysis, feminist theory, or public health discourse. The domestic labor analogy is exact: just as the person who cooks and cleans is doing work the household depends on but is not typically named as work, the person who sustains the group chat is doing work the friendship depends on but is not named as such.
What happens when the person doing this labor burns out? They stop. They go quiet, reply less, eventually mute the chat or let it drift. And the group, which has no other maintenance structure, begins to atrophy. Other members, who were never positioned as caretakers, find themselves unsure how to restart the momentum. The chat that once seemed self-sustaining reveals itself as having been sustained by one person's quiet effort. This is the classic dynamic of invisible labor withdrawal: the moment someone stops doing work that nobody acknowledged, the work's necessity becomes apparent through its absence.
There is a generational dimension here. Group chats as friendship infrastructure are disproportionately central to the relational lives of people who came of age with smartphones—roughly millennials and Gen Z—but the labor patterns they generate are not age-specific. They reproduce the same gendered, unacknowledged care work that has characterized friendship maintenance across all periods. What is distinctive about the group chat era is the scale and visibility of the imbalance: the read-receipt function, the seen-by count, the emoji reaction as minimal participation—these features make it possible to trace exactly who is doing the relational work and who is free-riding on it, in real time, in ways that older friendship maintenance forms did not permit.
The group chat also introduces a particular kind of audience problem. In a one-on-one friendship, you know who you are talking to and can calibrate accordingly. In a group chat, every message is simultaneously addressed to all members, and the relational labor of maintaining that group requires you to hold in mind multiple people's contexts, moods, and histories at once. This is cognitively demanding in a way that is rarely acknowledged. Writing a message that will land well for the person having a hard week, not alienate the person who is touchy about a topic that came up last month, and also land as funny for the three people who are just here for jokes is a genuinely complex social task. The people who do it well are performing a kind of distributed relational intelligence that the rest of the group benefits from without naming.
Naming this labor is not a call to reduce it. It is a precondition for distributing it more equitably, for thanking the people who do it, and for building collective awareness that friendship does not maintain itself—it is maintained, by specific people, through specific effort, at a real cost.