The Variance Penalty: Bodies Aren't Content Calendars

This essay started with a sentence I believed. During the luteal phase, the week or two before a period, focus and output drop, and a feed that rewards posting every day quietly punishes women for half of every month. Men, the sentence went on, run on a 24-hour hormonal clock, and the content calendar was drawn for that clock.

I fact-check this series by habit, and lately I have made a point of checking the claims that flatter me hardest of all. This one flattered something I care about, so it went through the same mill. Half of it came out the other side in pieces. The half that survived is sharper than what I walked in with, and it covers far more people.

The surviving version fits in one line, and the rest of this essay defends it. Visibility systems reward uninterrupted output, capacity varies from week to week for a great many people, and the cost of that mismatch is paid in private, through working while ill, burning out, and quietly leaving. I call that the variance penalty. Women carry a large and badly measured share of it. The claim they think less well for part of the month is where it starts to go wrong.

The sentence that didn't survive

Start with cognition, because that is where a critic would start. In 2017, Brigitte Leeners and colleagues followed 68 women across two full cycles, testing working memory, attention, and cognitive bias against measured hormone levels. Some effects appeared in the first cycle. None of them replicated in the second, and the authors concluded there was no consistent association between hormones and performance. Then, in 2025, Jang, Zhang, and Elfenbein published the largest review to date in PLOS ONE: 102 articles, 3,943 participants, 730 comparisons across attention, creativity, memory, spatial and verbal ability. They found no systematic, dependable evidence that performance shifts across the cycle.

So the luteal-phase version of my thesis hands any opponent a meta-analysis to swing. Worse, it recycles an old argument that has done real damage: that women's hormones make them less reliable, an argument historically used to keep them out of rooms. I don't want to build on it, and the evidence says I don't have to.

The 24-hour man fares a little better, then also breaks. The framing comes from the cycle-syncing literature, popularized by Alisa Vitti's In the FLO (2020), which contrasts a daily male rhythm with a monthly female one. There is a kernel. In men aged 30 to 40, testosterone runs 20 to 25 percent lower at four in the afternoon than at eight in the morning, a gap that narrows to about 10 percent by seventy. Men have a pronounced daily rhythm. Everyone has circadian rhythms. People who menstruate add a roughly monthly one on top. The neat 28-day month is shakier still: in 612,613 cycles logged by 124,648 app users, only 13 percent were exactly 28 days long, and the mean was 29.3. As for cycle syncing as a productivity method, its headline result, 92 percent fewer PMS symptoms after 60 days, was reported by FLO Living itself, the company that sells the method. I'll defend anyone's right to plan around their own body. I won't sell a schedule as science.

The corrected line reads better anyway. Work and platforms run on a clock that treats every day as interchangeable. Most bodies don't work that way, and menstruating bodies demonstrably don't.

What survived: the cost nobody sees

Average cognition doesn't move across the cycle. Experienced capacity does, for many people, and badly for some. Pain, fatigue, broken sleep, and mood are where the cost lives, and the best data on it is about hiding.

In 2019, Schoep and colleagues surveyed 32,748 Dutch women aged 15 to 45 about menstruation-related symptoms and work. 13.8 percent had taken time off for them, a mean of 1.3 days a year. 80.7 percent had kept working while symptomatic, on an average of 23.2 days a year, at a third less productivity, which adds up to 8.9 lost days a year. Presenteeism cost nearly seven times what absence did. Of the women who did call in sick, only 20.1 percent told their employer or school the real reason, and two thirds wished for more flexibility in their tasks and hours during their periods. One caution travels with these numbers: respondents were recruited through social media, which likely tilts the sample toward women with symptoms. Read them as what respondents reported, and they still describe a cost almost entirely invisible to whoever is counting output.

Behind the averages sit conditions with names. Premenstrual dysphoric disorder affects 3.2 percent with confirmed diagnosis, pooled across 44 studies and 50,659 participants. The WHO puts endometriosis at an estimated 10 percent of reproductive-age women, about 190 million people, and polycystic ovary syndrome at 10 to 13 percent, with up to 70 percent of those affected unaware they have it. A 2019 meta-analysis of 32 papers by Jang and Elfenbein found the severe end clusters too: the risk of suicide death was 26 percent higher at menstruation.

Then the cycle ends and a new variable arrives. In the Fawcett Society's 2022 survey of about 4,000 UK women aged 45 to 55, 10 percent of those employed through menopause had left a job because of symptoms, 14 percent had cut their hours, and 8 percent had not applied for a promotion. The LSE's policy review has warned that "one in ten" has become a sticky statistic, repeated without its sample, so here is the sample. The direction holds either way. Capacity that varies, carried quietly, priced as lower commitment.

What the platforms actually reward

Now the feed. The popular story is that algorithms reward consistency. The documented story is plainer, and harder on the people this essay is about.

YouTube's own help center contradicts the folklore in writing. Asked whether creators need to upload daily or weekly, the FAQ answers that growth in views "is not correlated with time between uploads," and it says a study of thousands of channels that took breaks found no correlation between break length and changes in views. It adds a fair caveat: an audience can take time to "warm up" again after a pause. Instagram's guidance has been vaguer, reported secondhand from its 2021 Creator Week as two feed posts a week and two stories a day. TikTok has said that neither follower count nor past hits are direct factors in its recommendations.

Third-party data points the other way. In Buffer's analysis of Instagram accounts, reach per post rose about 12 percent at three to five posts a week, 18 percent at six to nine, and 24 percent at ten or more, compared with posting once or twice, and weekly follower growth climbed to 0.66 percent at the top. vidIQ's July 2026 study of 10,210,278 YouTube channels found median monthly view growth of 0.53 percent for channels uploading less than once a month and 2.18 percent for those uploading twelve or more times. Both are correlational, and vidIQ says so itself: larger, healthier channels may simply have more resources to publish often. Both companies also sell tools for posting more.

The two pictures reconcile once you stop reading consistency as the mechanism. Ranking systems score each post on how its audience responds. More posts mean more chances to land. Volume is a lottery, and people with more capacity buy more tickets: more time, more staff, more health, more childcare, more good weeks banked against bad ones.

The pressure doesn't need the code to be on its side. Sophie Bishop's research on YouTube beauty vloggers named the mechanism algorithmic gossip: community theories about the algorithm that push creators toward uploading frequently, whatever the system actually does. Kelley Cotter described Instagram influencers playing a visibility game against rules they can only infer. Brooke Erin Duffy and colleagues, interviewing 30 creators across five platforms, found that anticipating algorithmic change had become part of the job itself, one layer of what they call nested precarity. The surveys that exist put burnout high: 52 percent of 1,000 US and UK creators in a 2025 Billion Dollar Boy survey, with financial instability ranked the top cause, and 62 percent of 542 North American creators in a Creators 4 Mental Health study. Both are commissioned, non-random samples, and neither summary I found broke the results down by gender. That gap is part of my argument. Nobody is measuring how cycles, menopause, or caregiving meet the content calendar.

I am inside this system too, and I chose my side of it deliberately. I just queued the next forty-two editions of my LinkedIn newsletter to go out one a month, a pace I can hold in a bad month as easily as a good one. That choice is available to me because the editions were written ahead, which is its own form of surplus capacity. Batching helps. It presumes the good weeks exist and that someone can afford to spend them twice.

The worker the calendar was drawn for

None of this is new to the sociology of work, which is why the essay can stop guessing. In 1990, Joan Acker argued in Gender & Society that the abstract job in organizational theory assumes a disembodied, universal worker, and that this worker, looked at closely, is a man: someone whose life centers on full-time work while someone else handles his body's needs, his household, and his children. Acker tied that figure to the control strategies of industrial capitalism, so this was never a story about intent. Joan C. Williams later named the same figure the ideal worker, in Unbending Gender (2000): always available, no time out for childbearing or care.

A content calendar is an abstract job in Acker's sense. Three posts a week, every week, assumes a worker whose capacity in March equals her capacity in April. Elizabeth Freeman gave the discipline a name in Time Binds (2010): chrononormativity, the use of time to organize bodies toward maximum productivity. Alison Kafer offered the counterweight in Feminist, Queer, Crip (2013): crip time, time bent to fit the body instead of the reverse. Kafer wrote it for disabled and chronically ill people, and it is the frame that keeps this essay honest, because the variance penalty falls on them, on carers, on anyone in a hard season, on many men. Women are the largest under-measured case of a general design failure. That keeps the argument clear of biology-as-destiny and points it at the thing that can be changed, the design.

The structure also answers the objection I hear first: the algorithm is gender-blind, it only optimizes engagement. Granted, fully. Acker's point was always that neutral-looking rules carry a body inside them. The bias lives in what gets measured, steady output and fast response, and in who can supply those on demand.

Why the targeted fix keeps failing

The obvious remedy is a period-shaped policy, and the record on that is sobering. Spain made paid leave for incapacitating menstruation law on June 1, 2023, with a medical diagnosis required to claim it. In its first eleven months it was used 1,559 times. By the Spanish ministry's own figures, 2,668 women used it between June 2023 and February 2025, against an estimate of about 60,000 workers who might qualify. Researchers writing in The Conversation point to stigma and uncertainty; critics quoted by People Management add that the diagnosis requirement is itself the barrier, given how hard one is to obtain.

The design lesson matches Schoep's survey, where only one in five women told the real reason for a sick day. Policies that demand disclosure, diagnosis, and a gendered label will be underused by the people they were built for. Universal flexibility works without anyone having to explain their body: asynchronous work, judging output over hours, banked low-capacity days available to everyone.

What I'd design instead

I design systems for a living, so here is the brief I would write for a platform that wanted to stop charging the variance penalty. None of these exist as described. They are proposals.

A pause mode that freezes a creator's ranking baseline, so two quiet weeks don't read as decline. Evaluation per post rather than per week, so one strong piece counts as fully as five adequate ones. YouTube's own break data, the finding that pauses don't correlate with view loss, put inside the creator dashboard where the anxiety actually lives, instead of buried in a help page. Scheduling and queue tools framed and priced as accessibility features. Ranking weighted toward saves, shares, and watch time over raw volume, which rewards the work over the worker's stamina.

The same brief applies inside studios and brands that hire creators. Contract for a body of work across a quarter, and leave the weekly output count out of it. I have written before about attention that runs in deep channels rather than a steady trickle. Calendars built for the trickle waste the people who work the other way.

Where this could be wrong

Three tiers, as this series runs.

Documented, re-verified against live pages and abstracts on October 15, 2026: both cognition reviews and their numbers; the Schoep survey figures with their recruitment caveat; the PMDD, endometriosis, and PCOS prevalence; the menstrual-phase mental-health meta-analysis; the Fawcett menopause figures and the LSE warning about them; YouTube's FAQ, quoted verbatim; the Buffer and vidIQ figures with their own caveats; TikTok's recommendation statement; the Instagram guidance, flagged as secondhand; the testosterone rhythm; the cycle-length data; the Spanish leave figures; the two creator burnout surveys, flagged as commissioned. Left out on purpose because I could not verify them to the standard of the rest: a widely shared Mosseri quote about sustainable posting, the creator pay-gap figures from uncontrolled industry data, and a figure on Japan's menstrual-leave uptake that survives only in a search snippet.

Inference, and mine: the variance penalty itself. Each link in the chain is sourced, but nobody has measured the whole chain at once. I know of no study that follows menstrual or menopausal symptoms through to creator output, ranking, and income. The causal story here is assembled from parts, and a careful reader should hold it as a well-supported hypothesis.

Watching, with named thresholds. One: a gender-disaggregated study of creator burnout and output, which would confirm or shrink the share of the penalty this essay assigns to women. Two: a major platform shipping a pause mode or putting its break data in the dashboard; if creators who pause stop losing ground in practice, the penalty is a design choice being corrected, and I'll say so. Three: a peer-reviewed, controlled trial of cycle-based scheduling. If one shows real gains, the planning half of the cycle-syncing story deserves a fairer hearing than I gave it here.

References

• Acker, J. (1990). Hierarchies, jobs, bodies: A theory of gendered organizations. Gender & Society, 4(2), 139–158. https://doi.org/10.1177/089124390004002002
• Bishop, S. (2019). Managing visibility on YouTube through algorithmic gossip. New Media & Society, 21(11–12), 2589–2606. https://doi.org/10.1177/1461444819854731
• Brambilla, D. J., Matsumoto, A. M., Araujo, A. B., & McKinlay, J. B. (2009). The effect of diurnal variation on clinical measurement of serum testosterone and other sex hormone levels in men. Journal of Clinical Endocrinology & Metabolism, 94(3), 907–913. https://doi.org/10.1210/jc.2008-1902
• Buffer. (2026). How often to post on social media: A data-backed guide. https://buffer.com/resources/social-media-frequency-guide/
• Bull, J. R., Rowland, S. P., Scherwitzl, E. B., Scherwitzl, R., et al. (2019). Real-world menstrual cycle characteristics of more than 600,000 menstrual cycles. npj Digital Medicine, 2, 83. https://doi.org/10.1038/s41746-019-0152-7
• Cotter, K. (2019). Playing the visibility game: How digital influencers and algorithms negotiate influence on Instagram. New Media & Society, 21(4), 895–913. https://doi.org/10.1177/1461444818815684
• Duffy, B. E. (2020). Algorithmic precarity in cultural work. Communication and the Public, 5(3–4), 103–107. https://doi.org/10.1177/2057047320959855
• Duffy, B. E., Pinch, A., Sannon, S., & Sawey, M. (2021). The nested precarities of creative labor on social media. Social Media + Society, 7(2). https://doi.org/10.1177/20563051211021368
• Fawcett Society & Channel 4. (2022). Menopause and the Workplace. Reported in People Management: https://www.peoplemanagement.co.uk/article/1754967/one-10-women-quit-job-menopause-symptoms-survey-reveals
• Fitch, J. (2026). Cycle Syncing Method demonstrates reductions in premenstrual syndrome symptoms. Contemporary OB/GYN. https://www.contemporaryobgyn.net/view/cycle-syncing-method-demonstrates-reductions-in-premenstrual-syndrome-symptoms
• Freeman, E. (2010). Time Binds: Queer temporalities, queer histories. Duke University Press.
• Jang, D., & Elfenbein, H. A. (2019). Menstrual cycle effects on mental health outcomes: A meta-analysis. Archives of Suicide Research, 23(2), 312–332. https://doi.org/10.1080/13811118.2018.1430638
• Jang, D., Zhang, H., & Elfenbein, H. A. (2025). Menstrual cycle effects on cognitive performance: A meta-analysis. PLOS ONE, 20(3), e0318576. https://doi.org/10.1371/journal.pone.0318576
• Kafer, A. (2013). Feminist, Queer, Crip. Indiana University Press.
• Leeners, B., et al. (2017). Lack of associations between female hormone levels and visuospatial working memory, divided attention and cognitive bias across two consecutive menstrual cycles. Frontiers in Behavioral Neuroscience, 11, 120. https://www.frontiersin.org/journals/behavioral-neuroscience/articles/10.3389/fnbeh.2017.00120/full
• LSE Public Policy Review. Women who quit: Media and policy discourse about gender and work. https://ppr.lse.ac.uk/articles/10.31389/lseppr.121
• People Management. (2024). A year on from Spain's menstrual leave law. https://www.peoplemanagement.co.uk/article/1876316/year-spains-menstrual-leave-law-%E2%80%93-learned
• Reilly, T. J., et al. (2024). The prevalence of premenstrual dysphoric disorder: Systematic review and meta-analysis. Journal of Affective Disorders, 349, 534–540. https://doi.org/10.1016/j.jad.2024.01.066
• Schoep, M. E., Adang, E. M. M., Maas, J. W. M., et al. (2019). Productivity loss due to menstruation-related symptoms: A nationwide cross-sectional survey among 32 748 women. BMJ Open, 9, e026186. https://doi.org/10.1136/bmjopen-2018-026186
• Social Media Today / Hutchinson, A. (2021, June 10). Instagram chief Adam Mosseri offers insights on how to grow reach. https://www.socialmediatoday.com/news/instagram-chief-adam-mosseri-offers-insights-on-how-to-grow-reach-algorith/601667/
• The Conversation. (2025). Menstrual leave: Stigma and uncertainty mean these progressive measures are being underused around the world. https://theconversation.com/menstrual-leave-stigma-and-uncertainty-mean-these-progressive-measures-are-being-underused-around-the-world-271449
• TikTok Newsroom. (2020, June 18). How TikTok recommends videos #ForYou. https://newsroom.tiktok.com/en-us/how-tiktok-recommends-videos-for-you
• vidIQ. (2026, July). YouTube upload frequency study. https://vidiq.com/research/youtube-upload-frequency-study/
• Vitti, A. (2020). In the FLO. HarperOne.
• Williams, J. C. (2000). Unbending Gender: Why family and work conflict and what to do about it. Oxford University Press.
• World Health Organization. Endometriosis [fact sheet]. https://www.who.int/news-room/fact-sheets/detail/endometriosis
• World Health Organization. Polycystic ovary syndrome [fact sheet]. https://www.who.int/news-room/fact-sheets/detail/polycystic-ovary-syndrome
• YouTube Help. YouTube performance FAQ & troubleshooting. https://support.google.com/youtube/answer/141805?hl=en

Source note: every URL and DOI above was checked live on 2026-10-15, journal figures against the published abstracts. The creator burnout figures come from commissioned surveys, cited through Tubefilter (Creators 4 Mental Health, November 2025) and Net Influencer (Billion Dollar Boy, July 2025), and are carried as indicative. Industry posting-frequency studies are correlational and come from companies that sell posting tools; both facts are stated where the numbers appear.