Short answer: Women who become visible face a penalty men don't: competence and likeability move in opposite directions. The research is consistent across politics, media and venture capital. Women who self-promote are rated more competent and less likeable. Investors ask them about risk while asking men about potential. And the newsrooms that frame their stories have their own documented problem. The cost lands as burnout, because managing a bind nobody admits exists is unpaid labour on top of the job.

This article is for women founders who suspect the problem isn't their delivery.

Key takeaways

• The double bind is documented, not anecdotal — competence and warmth trade off for women in ways they don't for men

• Australian papers framed Gillard's misogyny speech as emotional, strategic or hypocritical; international coverage was more positive

• 41% of women journalists have experienced sexual harassment at work, overwhelmingly from colleagues and managers

• VCs ask men about gains and women about losses — and that question type predicts funding

• Ardern's memoir argues the traits treated as disqualifying can be leadership assets

Contact www.geode.agency for your marketing needs email hello@geode.agency

What the double bind actually is

Laurie Rudman's backlash research established the mechanism: women who self-promote are judged more competent and less likeable, with the likeability penalty damaging real outcomes. The two ratings move in opposite directions.

A man who is direct is decisive. A woman who is direct is difficult. Both are described accurately — the behaviour is the same — but only one description carries a cost. The bind is that there is no setting that satisfies both demands at once. Warm enough to be liked reads as insufficiently serious. Serious enough to be credible reads as cold.

This is not a confidence problem. It is a scoring problem.

What the Gillard coverage demonstrates

Julia Gillard's 2012 speech on sexism and misogyny is now globally famous. What happened in the Australian press at the time is the more instructive part.

Katharine Wright and James Holland analysed coverage across five Australian newspapers for the Australian Journal of Political Science. They found the speech was framed three ways: as a strategic attack for political gain, as an uncontrolled emotional outpouring, and as hypocrisy.

The language was specific. One columnist wrote of her "head glowing red" and "big guns blazing" — martial imagery the authors note would rarely be applied to a male politician. The Sydney Morning Herald reported that "her voice trembled as she exorcised two years' worth of rage." Paul Sheehan contrasted "Robotic Julia" with "Furious Julia" and called her tone "vicious." The Courier Mail said she "almost quivered with rage." The "handbag hit squad" label did its own work.

Note what is being described: not the argument, but the body and the voice delivering it. A speech becomes a temperature reading.

Wright and Holland argue the press acted as a "gendered mediator," reinforcing the double bind and casting Gillard as an outsider rather than an ordinary leader. The framings narrowed the speech's impact and undermined its message.

The detail that settles it: the framing was consistent across all five Australian papers — and international coverage was more positive. The same speech, the same words. Only the proximity to the domestic political establishment changed. Alongside the coverage of her speeches ran years of commentary on her appearance, her accent, her marital status and her childlessness, of a kind her male counterparts never attracted.

Who writes the story

Here's the part that matters for founders seeking press, and it isn't where most people assume.

WAN-IFRA's Women in News, with City, University of London, surveyed over 2,000 media professionals across 20 countries and interviewed 85 senior executives. On average 41% of women journalists had experienced verbal or physical sexual harassment at work, against 12% of men — women and gender non-conforming people were roughly three and a half times as likely to experience it.

About 80% of cases went unreported, mainly from fear of consequences. Only 11% of respondents knew whether their organisation had a harassment policy. Where cases were reported, organisations acted about half the time, most commonly by issuing a warning.

The perpetrators were overwhelmingly internal: fellow employees 39%, direct supervisors 19%, higher management 18.9%. Roughly 38% held management positions.

That last figure reframes the problem. The pressure on women journalists isn't primarily coming from the powerful men they interview — it's coming from inside the newsroom, from the people who assign, edit and approve their stories. The gatekeeping is structural rather than transactional.

There's a perception gap to match: 43.5% of the executives interviewed said they had personally experienced harassment, yet only 27% thought it remained an industry issue.

For a founder, the implication is uncomfortable but useful. The coverage you receive is shaped by institutions that have their own unresolved version of the problem you're being covered through.

Fundraising: the question asymmetry

The clearest evidence of the bind operating on money comes from Dana Kanze, Laura Huang, Mark Conley and Tory Higgins, who analysed investor Q&A at TechCrunch Disrupt in New York.

Investors asked men promotion questions — about potential, upside, how big this could get. They asked women prevention questions — about risk, downside, what happens if it fails. Same competition, same stage, different interrogation.

Female entrepreneurs in the study raised five times less funding, and the questioning difference accounts for much of the gap. A follow-up experiment found founders asked promotion questions received roughly twice the funding of those asked prevention questions, which points to cause rather than correlation.

The practical move: answer the question they should have asked. Asked "how will you prevent churn?", respond with the retention figure and then pivot to the growth it enables. Acknowledge the risk briefly, then reframe to upside. You cannot control the question. You can control which frame your answer leaves in the room.

Why this shows up as burnout

The bind is expensive because it is constant and invisible.

Every public moment requires a calculation nobody else in the room is running: how direct can I be here, and what will it cost. That calculation doesn't appear in any workload audit. It has no deliverable. It simply runs underneath the work, all day, for years.

Burnout research distinguishes exhaustion from workload. This is the other kind — the fatigue of sustained self-monitoring, with no version of the performance that is safely correct.

Ardern's counter-argument

Jacinda Ardern's memoir A Different Kind of Power (Penguin, June 2025) is the most useful recent text on this, and not because it is triumphant.

The book runs chronologically from her childhood and schooling through to her resignation, covering the Christchurch mosque attacks, the COVID-19 pandemic and the 2022 Wellington protest. It discloses that she was pregnant during coalition negotiations, and that she first seriously considered resigning in late 2022 after a doctor found a breast lump initially feared to be cancerous — it was not.

Its central argument is the relevant one here. Ardern writes openly about imposter syndrome and the confidence gap, and contends that these feelings can benefit the leaders who experience them — that self-doubt and sensitivity are not disqualifications from power but a different basis for exercising it. Her stated purpose in writing was to encourage more people into public service.

It was shortlisted for the General Non-Fiction Award and won Best First Book at the 2026 Ockham New Zealand Book Awards. She narrated the audiobook herself.

The value for founders isn't inspiration. It's the existence proof: someone who led through genuine crisis, named the self-doubt publicly rather than performing its absence, and argues the trait was load-bearing.

What to actually do

Separate the signal from the scoring. When feedback arrives about tone, manner or likeability rather than substance, that is data about the scoring system, not about your work.

Reframe prevention questions in real time. Brief acknowledgement, then upside. Practise it until it's automatic.

Build the archive before you need it. Founders who weather hostile coverage are almost always those with an existing public record. An empty archive means someone else writes the first draft.

Find the rooms where the bind is weaker. It isn't uniform. Some investors, some journalists, some boards are measurably better. Spend your visibility where it compounds.

Frequently asked questions

What is the double bind in leadership? The pattern where women are penalised for being too warm to be credible or too assertive to be likeable, with no setting that satisfies both. Men's competence and likeability ratings don't trade off the same way.

Is the funding gap really about bias rather than business quality? The Kanze research isolates the mechanism — identical competition, different questions, and an experiment showing question type causally affects funding amounts.

What is Jacinda Ardern's book about? A chronological memoir of her life and premiership, arguing that imposter syndrome and sensitivity can strengthen rather than undermine leadership. Published June 2025; won Best First Book at the 2026 Ockham Awards.

Does media training solve this? It helps with delivery and it's worth doing. It doesn't change how a room scores you — which is why treating the problem as entirely personal is both exhausting and inaccurate.

You can't out-perform a scoring system. You can stop grading yourself with it.

Sources: Wright & Holland, Australian Journal of Political Science, WAN-IFRA Women in News sexual harassment study, Kanze et al. on VC questioning, A Different Kind of Power