Lean In

Sandberg, then Facebook's COO, argues that the leadership gap persists partly through internal mechanisms: women underestimate themselves, leave before they leave by planning around a family they do not yet have, and negotiate less because the social penalty for doing so is real.

Her recommendation is to sit at the table, pursue the next role, and choose a partner who genuinely shares the domestic load — which she identifies as the single most consequential career decision.

Her sharpest and least contested point is that the partner you choose matters more to your career than any employer.

Devmark’s reading — not a quotation from the book

The research on the likeability penalty — that success and likeability correlate positively for men and negatively for women — is well supported and the book made it widely known.

Its treatment of how women's ambition is read differently, and of the specific double bind in negotiation, holds up.

The book was substantially criticised, most influentially by bell hooks and by Michelle Obama's later comment on the phrase itself, for locating too much of the problem in individual behaviour and for generalising from an unusually privileged position.

Sandberg has since acknowledged that the book underestimated how much harder this is without money, partner support or flexible work — a point she made directly after her husband's death.

Its author later conceded the central limitation: the advice assumes resources many readers do not have.

Devmark’s reading — not a quotation from the book

The research finding the book rests on most solidly is that competence and likeability correlate positively for men and negatively for women, so that the same assertive behaviour produces opposite social consequences. It is well replicated and it reframes a great deal of advice about self-presentation as a genuine double bind rather than a confidence problem.

The same behaviour makes him respected and her disliked. That is a finding, not an impression.

Devmark’s reading — not a quotation from the book

Sandberg's most-quoted argument is that women begin scaling back ambitions years ahead of any actual constraint — declining a stretch role in anticipation of a family that does not yet exist — and thereby arrive at the decision point with a weaker position than they needed to have. As individual advice it is sound; as an explanation of the aggregate gap it is doing far more work than it can carry.

The most substantively supported chapter concerns domestic labour: the argument that the single largest career variable for a woman in a heterosexual partnership is how the unpaid work is divided, and that this is negotiable and usually never negotiated. The supporting evidence on the division of housework and childcare is strong.

A cluster of observations about meeting behaviour — women sitting at the edges of rooms, understating contributions, attributing success to luck and help — are presented as correctable habits. They are real and documented; the difficulty is that correcting them runs directly into the likeability penalty the book itself establishes.

The most influential objections came from feminist writers who argued the book addresses a narrow population — women with education, capital, childcare and partners — and converts a structural problem into a personal deficiency, so that the remedy asks the disadvantaged party to adapt to conditions it leaves unchallenged. That critique is substantive and has largely held.

The book's reputation has shifted considerably since 2013, affected both by the structural critique and by later scrutiny of the author and the company she led. The research it cites remains valid; the framing has aged into an example of a genre — individual optimisation offered in place of structural change — that its own evidence partly refutes.

Take the likeability research and the chapter on domestic division, both of which are well supported and useful. Treat the central injunction as advice that works for a specific and fortunate population, and read one of the major critical responses alongside it rather than after.