You will not eliminate all the bias at your company with a single effort or in a single year — and that should not be the goal. The goal should be to take small steps consistently. If you build evidence-based tweaks into your fundamental business systems, the organizational change you effect will be more resilient and long-lasting than a CEO-driven, conversation-based culture change.
The first step is improving how you recruit and hire candidates.
For example, in 2015, only 10% of new data scientists at Airbnb were women. By implementing the kind of best practices described below, the company increased female new hires to 47% in just one year, doubling its overall number of female data scientists from 15% to 30%.
The first step was to use data analysis to find out what percent of data scientist applicants were women.
Second, the company increased conversion (the proportion of applicants hired) by taking some steps to interrupt bias in résumé review, such as providing a detailed description of what to look for and making ratings clearer and more objective.
Third, Airbnb changed the interviewing process to avoid the situation where women candidates face unending rooms of male-only interviewers, requiring instead that women make up at least half of interview panels for female candidates.
The company also offered all candidates — men as well as women — an informal coffee chat with a member of the interview panel so they would not face a room full of unfamiliar faces. (Notably, this is an example of how improving business processes helps every group.)
Also, in assessing interviews, Airbnb focused the discussions about the candidates on the objective traits of the presentation rather than the subjective interpretations of the interviewers.
Your company can use such bias interrupters to improve diversity in hiring. While the metrics for your company may differ from Airbnb’s, here are some suggestions for interrupting bias in recruiting and hiring:
1. Upgrade résumé reviews. Before a résumé review begins, hold a meeting to go over how to identify and interrupt bias in hiring. Research shows that it is important to review this information orally, rather than just give people a worksheet.
2. Commit to what’s important — and require accountability. Put in writing which qualifications are important, both in entry-level and in lateral hiring. When qualifications are waived for a specific candidate, require an explanation of why — and keep track of which candidates receive these waivers.
3. Rate independently, use a consistent rating scale, and discount outliers. Establish clear grading rubrics and ensure that everyone grades on the same scale. When a group of people are evaluating candidates, the best procedure is for each evaluator to give their rating independently, without consultation. Consider having each résumé reviewed by two or more managers and averaging the score. If possible, discount outliers (reviewers who disproportionately loved or loathed a candidate).
4. Ask questions about résumé gaps. Don’t count gaps in a résumé as an automatic negative. Instead, give the candidate an opportunity to explain their work history and anything that might seem unusual about it. There are many, many reasons people might have a gap between jobs. Judge them by the work they’ve done, not the time they’ve missed. This is fairer to people with family caregiving responsibilities, as well as anyone who has had to take time off from work for serious health issues. If someone has taken time away from paid work to look after children or aging parents, don’t infer that they’ll be less committed to the job they’re applying for now.
5. Provide all candidates with a handout detailing expectations. This can level the playing field for first-gen professionals, Asian Americans, women, and introverts; these are all groups more likely to feel pressure to be modest or self-effacing. Setting expectations clearly allows them to make their best case for themselves. One handout we’ve used includes advice like, “Present yourself with quiet confidence. Bluster doesn’t work well here, nor does excessive modesty.” We also think it’s a good idea to remind candidates of the qualities and skills you’re looking for and share the metrics you’ll use to interviewees after the conversation.
6. Use structured interviews and ratings. Ask every candidate the same list of questions. Ask questions directly relevant to the job. Don’t ask about hobbies and the like unless your goal is to replicate your existing workforce. After the interview is over, interviewers should rate the candidate using a consistent rating scale. After collecting these ratings, discount the outliers just as you did with the résumé screening process.
7. Rethink culture fit. If culture fit is used in hiring, first ask the hiring manager to articulate in writing what “culture fit” means. Then keep track of who is seen as a poor fit and look for demographic patterns. This isn’t only about people of color and white women; keep track of first-gen candidates as well. If the culture fit criterion is having an exclusionary effect, the company needs to define it in terms of work-relevant skills and dispositions, not in terms of preferred leisure activities or the lunch test (would I like to have lunch with this person?). Another problem with the concept of culture fit is that, once “fit” is felt, disqualifying factors and attributes may well be minimized or ignored, and objective requirements that have been applied to others may be set aside.
8. Ask performance-based questions and/or use work sample screenings. Questions like, “Tell me about a time you had too many things to do and had to prioritize” provide concrete information about job-relevant skills. If applicable, ask candidates to take a skills-based assessment. For example, if part of the job is analyzing data sets and making recommendations, ask the candidate to do that. If part of the job involves writing, ask candidates to submit writing samples.
9. Track what happens after the interview. If women, people of color, and first-gen white men are getting called in for interviews but not being offered jobs at the end of the process, conduct an audit of your interview process. Are these candidates rated differently by interviewers? Are interviewers treating them differently than in-group candidates? A study of economists found that, when presenting research to peers, women get more questions than men, including ones that are patronizing, disruptive, demeaning, or hostile. The same study showed that women were interrupted more frequently. If your company is attracting a diverse pool of applicants but those people are not making it through the interview process, it’s fair to ask if dynamics like these could be at play.
Reprinted by permission of Harvard Business Review Press. Adapted from Bias Interrupted: Creating Inclusion For Real and For Good by Joan C. Williams. Copyright 2021 Joan Williams. All rights reserved.