How to Reduce Bias in Your Hiring Process

The thing about unconscious bias is precisely that—it’s unconscious. Even the most well-intentioned recruiters and hiring managers weed out qualified candidates because of bias, even when those stereotypes run counter to their stated value systems. Of course, hiring bias ultimately means missing out on excellent talent and on building a trusting and inclusive team. So once you’ve found and nurtured underrepresented talent, getting them into your pipeline (we know diversity isn’t a “pipeline problem”), how do you ensure they move through equitably, at the same rates majority talent passes through those same funnel stages? How do you know if—or where—there may be hiring bias anywhere in your process that’s causing female-identified talent or URGs to drop out, or to be overlooked or wrongfully rejected?

It’s a critical question to ask; and it entails asking a series of other questions about your process, from reachout to offer-out. Here’s what to consider: 

Is there bias in your sourcing or screening process?

Data from as recently as 2017 revealed that when recruiters source candidates on LinkedIn—regardless of the gender of the recruiter—they’re more likely to click into male profiles. LinkedIn’s 2018 Gender Insights Report bolstered this with its own data point: recruiters are 13% less likely to click on a woman’s profile when she shows up in a search. The takeaway is that unconscious bias exists even when URMs are doing the sourcing. (Indeed, it exists for all of us.) 

We’ve noticed it in our own customer data at Gem: more than 2x as many male candidates enter process as female candidates. Yet once they’re in process, female-identified candidates outperform their male peers, passing through to onsite stages at higher rates. The numbers suggest that recruiting teams have a more equitable interview process than they think. Diversity is more a top-of-funnel problem. 

So it’s worth considering the elements that make up your sourcing and screening processes. Have your recruiters put themselves through unconscious bias trainings—whether on Facebook, Harvard, Catalyst, Paradigm, or elsewhere? Have you considered technologies that remove identifiers from inbound applications to exclude personal information that’s been shown to trigger hiring bias? Even if you do the latter, we highly recommend the former. Unconscious bias training is humbling work; and while it won’t eliminate hiring biases, it’ll surface them to recruiters’ consciousness so they can be aware of them as they uncover and encounter talent, and respond accordingly. 

Are your interviews structured?

Are interviewers asking the same set of questions in the same order to every candidate? Are your rubrics standardized and is your hiring team calibrated on what weak, good, and excellent answers look like? Do you formalize hiring criteria before those interviews (indeed, before you even look at resumes) to keep interviewers from making subjective decisions based on “gut feeling” and amorphous criteria? “Structured interviews” mean no shortcuts, additional steps, or different bars for different candidates. Candidates get compared equally, apples-to-apples, based on the criteria that matter.

Are your hiring teams diverse? 

Are the folks included in your decision-making process reflective of the candidate pool you want? When underrepresented candidates are met with homogenous interview teams, it only draws their attention to the fact that yours is a culture they may not feel belonging in. What’s more, diverse hiring teams get to check each other’s unconscious biases, ensuring interviewers back up their assessments about a candidate with facts from the interview—based on a candidate’s answers—rather than “intuition.” The more balanced your team is, the fairer the decision will ultimately be.

Does your hiring process include a work sample or a performance task related to the role? 

Like structured interviews, skills assessments and project-based interview stages put all candidates on an even playing field, ensuring your hiring team is really prioritizing skills, key competencies, and proficiencies. Have you asked whether your traditional assessment methods work for all candidates? Three years ago, Microsoft replaced job interviews—which focus heavily on social skills—with a weeks-long vetting process and team-building exercises for candidates with autism, allowing them to better showcase their skills. Are there demographics for whom you might need to do something similar?

Are you still using the phrase “culture fit” in assessing candidates? 

It will be worth it to stop using that phrase and think about “values fit” and “culture add” instead. (“Culture add” gives prospects and candidates room to offer something unique and fresh to your current workforce, rather than reinforcing the group’s homogeneity.) Over time, “culture fit” has taken on a tribal meaning, facilitating hiring bias by giving recruiters, hiring managers, and companies permission to hire people that walk, talk, look, and think like them. Stop asking if a candidate will “mesh” with your team. Ask how well they’d perform the work the role requires. Ask if they’d bring the team something it doesn’t already have. 

Have you adopted the Rooney-Rule-Plus-One?

The Rooney Rule is an NFL regulation requiring one candidate for every head coaching and senior ops job to be an ethnic minority. Many talent teams have adopted it for their own purposes, so that one female-identified and one URM candidate must be considered in the final slate for each open position. But research has found that when the candidate pool only has a single minority candidate, that person has zero statistical chance of getting hired. When there are at least two female candidates in the final pool, however, the odds of hiring a female are 79x greater; and when there are at least two minority candidates in the final pool, the odds of hiring a minority candidate are 194x greater. In other words, your hiring team has to do better than the Rooney Rule if it wants to choose a minority candidate from your finalists at the end of the day.

Are you sending out candidate experience surveys?

This is critical to optimizing your hiring process across the board; but pay special attention to the experiences of minority candidates. CX surveys should go out regularly during the hiring process; and they should go out to accepted candidates, rejected candidates, and candidates who drop out of process alike. Be candid with your questions. What was the candidate’s perception of your culture and your team? Did they get the sense that they would thrive with you—and if not, why not? Where you see patterns in responses to these surveys, it’s time to take a closer look. 

Do you have recruiting tech that tracks diversity through the hiring funnel? 

At Gem, we offer top-of-funnel metrics to help you understand the pipeline your team is building. Sourcers, recruiters, and talent leaders can report on the gender and race/ethnicity breakdowns of their pipelines to discover how their efforts are impacting diversity. We also offer custom fields for tracking other segments of candidates that talent teams can tailor more specifically to their diversity initiatives. This data can not only shed light on whether teams are reaching out to a diverse talent pool, but also:

• reveal where systemic hiring biases might show up—by role, recruiter, or hiring manager—as some candidate segments get stuck at certain stages of the funnel. Are certain groups disproportionately dropping out of the funnel at specific points in the process? 

• detect whether email outreach is inadvertently alienating underrepresented groups so sourcers can rework their messaging to be more inclusive 

• inform hiring managers about diversity breakdowns in historical data 

• show via top-of-funnel metrics why the team did or did not hit its diversity hiring goals in a given quarter or year 

• view the difference in pipeline between focused efforts on diversity vs sourcing through unfocused methods

With both outreach and funnel analytics, you have all the intelligence you need to see not only how you’re trending toward your diversity goals, but also where you can optimize—and, where necessary, debias—your process, ultimately architecting a hiring process that’s thoughtful and equitable at every stage and every touchpoint. Data is the best way to inform yourself about the particulars of your process—including where hiring bias might lie. But if you don’t have the data to inform those pipeline decisions, the best questions to ask yourself are above.