At 2 a.m., a red badge in Slack is one more reminder that our attention is always up for grabs. Most of this extraction is invisible: the unpaid care that keeps our homes running, the “always-on” pings that fracture our focus, the buried checkbox that pretends to be consent.
Over the past months I’ve written about many of these extractions:
- How default settings pull our attention to maximize engagement (Slow Reading Is a Form of Refusal).
- How emotional labor (checking in on a colleague after a tough week or smoothing over a family conflict) holds teams and households together even though it never appears on a spreadsheet (Quiet Is Not Peace, Flexible Reciprocity).
- How “consent” often hides behind a small “Cancel” link, which makes it performative unless we actively keep it alive (Consent Has to Be Kept, Not Just Given).
- How original contributions vanish until someone with authority repeats them, erasing attribution (Emergence Is an Excuse).
- How institutional neglect stays invisible until a counter-archive of “vanished requests” forces accountability (Nobody Is Too Expensive).
- And how these hidden costs accumulate most heavily for those already marginalized (Collapse Capitalism, Time Is a Commons).
None of these are minor glitches. They are structural choices that shape how we live and work. Calling for “more inclusion” or “better empathy” doesn’t reach the place where power hides.
The costs don’t fall evenly, and the unevenness runs by sex. In the Bureau of Labor Statistics’ American Time Use Survey for 2023, 86 percent of women and 71 percent of men did some household activities on an average day, and on the days they did, women spent 2.7 hours on them to men’s 2.1. Among adults living with children under six, women spent 1.2 hours a day on physical care such as bathing and feeding; men spent 34 minutes.1
At work, the same pattern attaches to tasks that keep things running and don’t get anyone promoted. When the economists Linda Babcock, Maria Recalde, Lise Vesterlund and Laurie Weingart examined a large public university’s annual email asking 3,271 faculty members to volunteer for a committee, 7.0 percent of the women volunteered and 2.6 percent of the men. In their lab experiments, women in mixed groups volunteered for a task everyone wanted someone else to do 50 percent more often than men did. In single-sex groups the gap disappeared, and when an outside requester had to pick someone, they asked women more, and women said yes more.2 That unevenness is why these are feminist questions.
So my feminism isn’t “soft.” It’s practical, precise, and systemic.
In practice, I work from these commitments.
- Budgeting care as a resource. Treat emotional labor and meeting facilitation as line items—no more assuming someone will “just take care of it.” When we plan a release or a community meeting, we explicitly account for who will handle the follow-ups, who will check in with people, and how we’ll share that work. Naming the work keeps it from sliding into “invisible labor” and burnout.
- Reversing default notifications. Instead of auto-subscribe or “opt-in analytics,” we start with “mute by default,” requiring each person to choose which updates matter. This treats attention as finite and pushes platforms to justify why they need our focus in the first place.
- Building in real opt-outs. Every interface—whether a household chore chart or a project management tool—should include a visible, shame-free way to step back. No more “Cancel” links hidden behind three layers of UI. If opting out is easy, consent stays real instead of becoming a checkbox we click and forget.
- Crediting original voices. Whenever an insight or suggestion is repeated, the originator gets named. Ideas can then circulate without losing their source, and expertise stops depending on job title or seniority.
- Turning private logs into public pressure. If a request, whether for maintenance, policy change, or emotional support, goes unanswered, it’s recorded in a shared space. That visibility makes it harder for organizations to pretend they didn’t know, shifting accountability from “I didn’t realize” to “Here’s what we missed.”
Say a team is planning a release. “Can someone take the follow-ups?” is the faculty email again, and the answer will come from whoever the room expects to say yes. Babcock and her colleagues concluded that the gap most likely came from that expectation, since it vanished when everyone in the group was the same sex. Budgeting care, the first commitment above, means the follow-ups go on the plan with a name and a rotation, the way the deploy does, so there’s no open call for the expectation to answer.
These strategies do more than “feel good.” They change the architecture of how decisions are made and whose well-being counts.
On behalf of everyone affected, they ask:
- Who benefits?
- Who pays the hidden cost?
Care and consent stop being side effects and become baseline metrics, next to performance targets, sales goals, and quarterly deliverables.
Feminism can’t stop at affirmations, or at practices one team adopts and the next one drops. The commitments above count only once they’re part of the structure under our policies, our interfaces, our norms: the rotation written into the plan, the mute set as the default, the opt-out built into the tool.
Every pause before clicking “Agree,” every “no” said without apology, every piece of unseen work made visible strengthens the systems that protect mutual respect and shared responsibility.
Feminism is anything but soft. It’s the structural insistence that care, attention, and consent are never afterthoughts. It’s the steel framework that holds up a different way of living and working, one where well-being matters as much as productivity, and where saying “no” carries as much weight as saying “yes.”
Notes
U.S. Bureau of Labor Statistics, “American Time Use Survey — 2023 Results,” news release, June 27, 2024. BLS
L. Babcock, M. P. Recalde, L. Vesterlund, and L. Weingart, “Gender Differences in Accepting and Receiving Requests for Tasks with Low Promotability,” American Economic Review 107, no. 3 (2017): 714–747. The faculty data cover the 2012–2013 academic year. doi:10.1257/aer.20141734