We redesigned a financial institution's job postings to increase women's applications, without affecting men's.
The institution faced an underrepresentation of women in leadership positions and wanted to increase their applications for executive roles.
A critical factor was the ambiguity of the job descriptions, which often discourages potential candidates. On top of that were perceptions and implicit barriers limiting gender diversity at senior management levels. The goal: create a more inclusive and equitable environment that would promote greater participation of women in leadership roles.
We designed four versions of a job posting based on behavioral science and measured the real effect of each one on intent to apply and the emotions it generated when read.
Ambiguous, incomplete information, as a comparison point.
More specific, concrete information about duties, requirements, and benefits (uncertainty-reduction principle).
The same principle as above, plus a message spelling out the gender equality opportunities the organization offers.
Uncertainty reduction plus the video of a female leader, to reinforce women's role as role models. This was the winning strategy.
You can debate in a meeting which posting will work best, or launch it and cross your fingers. A controlled experiment does something different: it compares several versions with real users and measures which one actually moves the needle, and by how much.
That way you scale the version that proved it works, with the numbers behind it.
Debating which version wins is quick and cheap. The experiment is what gives you the answer with evidence.
We randomly assigned participants to one of four versions of the posting (a controlled trial). Each version added one element on top of the previous one: clearer information, then an equality message, and finally the video of a female leader. By comparing the groups, we measured the real effect of each element on intent to apply and on the emotions it generated. That's how we knew exactly which combination to take to production.
Figure 1 · Willingness to apply, by posting
Figure 2 · Emotions generated by each posting
The most effective strategy combined uncertainty reduction with the video of a female leader. Giving clearer information already increased intent to apply for both men and women; adding the equality message raised it further and reduced negative emotions; and the video was what generated the most applications and identification.
Clarity in job postings, combined with equality messaging and female role models, increases women's applications for leadership roles without affecting men's applications.
A key finding for behavioral design: audiovisual media can deliver more information without triggering cognitive overload. These results gave the organization a concrete basis for designing more effective job postings and promoting gender diversity in its leadership.