What Works

Bohnet's starting point is that unconscious bias is everywhere and very hard to train out of people. Diversity programs show limited results, and individual effort often provokes backlash. Her alternative is behavioral design: change the procedures through which hiring, evaluations and promotions happen so that biased instincts have less room to operate. The evidence comes from companies, universities and governments in many countries, often from randomized trials.

Bohnet is a professor at Harvard Kennedy School, where she co-directs the Women and Public Policy Program and works on behavioral insights. The book distills her own research and the wider literature, from structured interviews to the blind auditions that changed orchestra hiring.

What stands out is the care. Carol Tavris in The Wall Street Journal praised Bohnet for not overgeneralizing, for warning against extrapolating from one group to another and for admitting where the data run out. The Financial Times and the New Statesman singled out its practical, business-case language for managers and boards. The caveat follows from that caution. Many interventions rest on specific studies whose effects may vary across firms, so readers should treat the book as a menu to test, not a guaranteed program.

It suits people leaders, recruiters, managing directors and board members, especially in finance, where hiring and promotion decisions still rest heavily on judgment.

1. Fix the process, not the person. Bias training has weak evidence behind it. Structured procedures change outcomes more reliably than trying to change attitudes.
2. Compare candidates side by side. Evaluating people jointly against the same criteria, rather than one at a time, reduces reliance on stereotypes in hiring and promotion.
3. Structure the interview. Asking every candidate the same questions in the same order and scoring answers separately predicts performance better than free-form conversation.
4. Measure what you manage. Tracking hiring, pay and promotion outcomes by gender and running trials on new policies turns diversity from a slogan into an operating metric.
5. Mind the environment. Formats such as guessing penalties, open-ended negotiations and lone token hires can disadvantage women regardless of ability. Redesigning them is cheap and fast.

The book opens with the promise of behavioral design and is organized in four parts that move from diagnosis to specific fixes, ending with a chapter on designing change.

1. The problem. Four chapters show that unconscious bias is everywhere, that de-biasing minds is hard, that individuals trying to fix bias on their own take risks such as backlash for negotiating, and that outside help like training or mentoring only goes so far.
2. Designing talent management. Bohnet covers applying data to people decisions, smarter evaluation procedures such as structured interviews and comparative evaluation, and attracting the right candidates through job ads and recruiting choices.
3. Designing school and work. Chapters on adjusting risk, including test formats that penalize guessing, and leveling the playing field, which deals with competitive settings that disadvantage women for reasons unrelated to ability.
4. Designing diversity. She examines role models, how to build groups and the effect of critical mass, how norms shape behavior, and how transparency and accountability can change results.
5. Designing change. A closing chapter pulls the tools together and argues for testing interventions with data rather than relying on good intentions.

Gender equality is a moral and a business imperative. But unconscious bias holds us back and de-biasing minds has proven to be difficult and expensive. Behavioral design offers a new solution. Iris Bohnet shows that by de-biasing organizations instead of individuals, we can make smart changes that have big impacts—often at low cost and high speed.