Intentional recruitment strategies designed to attract, evaluate, and hire candidates from underrepresented groups to build a more diverse workforce.
Key Takeaways
Diversity hiring is the practice of intentionally building recruitment processes that attract, fairly evaluate, and select candidates from a broad range of backgrounds, including different genders, ethnicities, ages, disabilities, sexual orientations, socioeconomic backgrounds, and cognitive styles. It doesn't mean lowering standards or filling quotas. It means removing barriers that prevent qualified candidates from underrepresented groups from entering and advancing through the hiring process. The business case is well-established. McKinsey's 2023 "Diversity Matters Even More" report, analyzing over 1,200 companies across 23 countries, found that companies in the top quartile for gender diversity on executive teams were 39% more likely to achieve above-average profitability. For ethnic diversity, the figure was 39% as well. Boston Consulting Group's 2023 research found that companies with above-average diversity produce 2.5x higher cash flow per employee.
Diversity hiring is not hiring underqualified candidates to meet demographic targets. It's not reverse discrimination. And it's not tokenism (hiring one person from an underrepresented group and calling the job done). What it is: removing bias from every stage of the hiring process so that the best candidates, including those from underrepresented groups, have an equal chance of being identified, evaluated, and selected. Most diversity hiring failures happen because organizations focus on the final hiring decision rather than fixing the upstream pipeline. If your job postings, sourcing channels, and screening criteria systematically exclude diverse candidates before they even reach the interview stage, no amount of good intentions at the offer stage will fix the problem.
In the US, Title VII of the Civil Rights Act prohibits employment discrimination based on race, color, religion, sex, and national origin. Diversity hiring practices must comply with this law. Quotas are generally illegal. However, affirmative action (proactive steps to increase diversity) is legal when properly structured. In the UK, the Equality Act 2010 allows "positive action": if two candidates are equally qualified, the employer can prefer the candidate from an underrepresented group. In the EU, the Employment Equality Directive provides similar protections. Canada, Australia, India, and other countries have their own frameworks. The common principle: you can widen the pipeline and remove bias, but you can't make protected characteristics the sole deciding factor.
The biggest challenge in diversity hiring isn't the final selection. It's getting diverse candidates into the pipeline in the first place.
Research by Textio (2024) found that job descriptions with gendered language attract 42% fewer applicants from the underrepresented gender. Words like "aggressive," "dominant," and "competitive" attract more male applicants. Words like "collaborative," "supportive," and "nurturing" attract more female applicants. Use tools like Textio, Gender Decoder, or Datapeople to audit your job descriptions for biased language. Also review requirements: does the role truly need a bachelor's degree, or is that a proxy that excludes candidates with equivalent experience but different educational paths?
If you only source from the same universities, job boards, and professional networks, you'll keep getting the same demographic profiles. Add channels that reach underrepresented communities: HBCUs and HSIs (for ethnic diversity in the US), disability-focused job boards (AbilityJobs, Disability:IN), veteran hiring platforms (Hire Heroes USA, RecruitMilitary), organizations for women in tech (Women Who Code, AnitaB.org), and LGBTQ+ professional networks (Out & Equal, myGwork). LinkedIn's 2024 data shows that sourcing from 5+ channels increases pipeline diversity by 28%.
Blind resume screening removes identifying information (name, photo, address, university name) from applications before they're reviewed. A landmark study by the National Bureau of Economic Research found that resumes with "white-sounding" names received 50% more callbacks than identical resumes with "Black-sounding" names. Removing names eliminates this specific bias. Several ATS platforms (Applied, GapJumpers, Toggl Hire) offer built-in anonymization features.
Employee referral programs are effective, but they tend to reproduce the existing workforce's demographic profile. People refer people who look like them. Counter this by tracking referral demographics, incentivizing referrals from underrepresented groups, and partnering with Employee Resource Groups (ERGs) to extend referral networks into diverse communities. Some companies offer double referral bonuses for candidates who increase team diversity.
Even with a diverse pipeline, bias in the evaluation process can filter out diverse candidates before they reach the offer stage.
Structured interviews (same questions, same order, same scoring rubric for every candidate) reduce interviewer bias by 40% compared to unstructured interviews (Campion et al., Journal of Applied Psychology). They also improve predictive validity, meaning they're better at identifying who will actually succeed in the role. Every candidate gets a fair, consistent evaluation.
Interview panels that include people from different backgrounds reduce groupthink and affinity bias. Research from the Kellogg School of Management found that diverse panels are 35% more likely to make hiring decisions that outperform those made by homogeneous panels. At minimum, ensure that no interview panel is entirely composed of one demographic group. Some companies have a policy that panels must include at least one member from an underrepresented group.
Define exactly what "qualified" means before reviewing any candidates. Create a scoring matrix tied to job-relevant skills, competencies, and experience. Share this matrix with all evaluators before interviews begin. When criteria are vague ("strong leader" or "good communicator"), interviewers fill in their own definitions, which are often influenced by unconscious bias. Specific criteria ("has led a team of 5+ people on a project with a deadline under 6 months") are harder to evaluate with bias.
Supplementing interviews with job-related work samples reduces bias because evaluators focus on the work, not the person. Google found that structured work samples are the best predictor of job performance, outperforming interviews, resumes, and GPA. For technical roles, use coding challenges or case studies. For creative roles, use portfolio reviews with anonymized submissions. The key is that the work sample directly mirrors what the person will do in the role.
What gets measured gets managed. These are the essential metrics for evaluating your diversity hiring efforts.
| Metric | What It Measures | How to Calculate | Benchmark |
|---|---|---|---|
| Pipeline diversity ratio | Demographic composition of applicant pool | % of applicants from underrepresented groups at each funnel stage | Should reflect or exceed local labor market demographics |
| Pass-through rate by demographic | Whether diverse candidates advance at the same rate as others | % of each demographic group advancing from one stage to the next | Rates should be comparable across groups (within 5%) |
| Interview-to-offer ratio by demographic | Whether bias exists in the final selection stage | Offers made / interviews conducted, by demographic group | Ratios should be consistent across groups |
| Diversity of hire rate | Actual diversity of new hires | % of new hires from underrepresented groups | Varies by industry and region |
| Retention by demographic | Whether diverse hires stay at the same rate as others | 12-month retention rate by demographic group | Should be within 5% of overall retention rate |
| Source effectiveness by diversity | Which sourcing channels produce the most diverse candidates | % of diverse hires by channel (job boards, referrals, direct sourcing) | Use to allocate sourcing budget |
Your job description is the first filter in the hiring process. Biased language filters out diverse candidates before they even apply.
A widely cited Hewlett-Packard internal study found that women apply for jobs when they meet 100% of the listed requirements, while men apply when they meet about 60%. LinkedIn's 2024 data confirms a similar pattern across genders. If your job description lists 15 "requirements" when only 5 are truly essential, you're inadvertently discouraging applications from women, minorities, and career changers who self-select out because they don't check every box. Solution: clearly separate true requirements from nice-to-haves, and add language like "We encourage candidates who meet some but not all requirements to apply."
| Problem | Example of Biased Language | Inclusive Alternative |
|---|---|---|
| Gendered language | "He will be responsible for..." or "Looking for a rockstar" | "You will be responsible for..." or "Looking for a skilled professional" |
| Unnecessary requirements | "Must have a 4-year degree from a top university" | "Bachelor's degree or equivalent experience" |
| Ableist language | "Must be able to stand for 8 hours" | "This role involves extended periods of standing. Accommodations available." |
| Age-coded language | "Digital native" or "5-7 years of experience maximum" | "Proficient with digital tools" or "5+ years of experience" |
| Exclusionary culture signals | "Work hard, play hard" or "Beer Fridays" | "We value dedication and team connection" or describe specific team activities inclusively |
| Excessive requirements | 15+ bullet points of "must-have" qualifications | Separate "required" (3-5 items) from "preferred" (3-5 items) |
These specific strategies have data supporting their effectiveness in increasing hiring diversity.
The Rooney Rule, originally adopted by the NFL in 2003, requires that at least one candidate from an underrepresented group is interviewed for every open position. Many corporations have adopted variations. Microsoft, Amazon, and Meta require diverse interview slates for all management and above roles. Research shows that when the slate includes at least two diverse candidates, the odds of a diverse hire increase by 79.1x (Harvard Business Review, 2016). A single diverse candidate on the slate has virtually no statistical effect because they're treated as the "token" outlier.
76% of job seekers evaluate company diversity when considering employers (Glassdoor, 2023). Your careers page, social media, and recruitment marketing need to reflect diversity authentically. Show real employees from diverse backgrounds, not stock photos. Publish diversity data and progress reports. Highlight Employee Resource Groups, mentorship programs, and accommodations for people with disabilities. Candidates from underrepresented groups research these signals more thoroughly than majority candidates.
Returnship programs help professionals re-enter the workforce after career breaks (often for caregiving, which disproportionately affects women). Companies like Goldman Sachs, IBM, and PayPal run structured returnship programs that provide 12 to 16 weeks of paid project work, training, and mentoring. Conversion rates to full-time roles average 75-85%. Returnships are one of the most effective strategies for increasing gender diversity in mid and senior-level roles.
These pitfalls undermine diversity hiring efforts and can actually harm the people they're intended to help.
Hiring diverse talent is only valuable if those employees stay and thrive. Retention is where most diversity efforts fail.
A 2024 McKinsey report found that women and people of color are 1.5 to 2x more likely to leave their employer within 2 years, even when they report similar job satisfaction levels at hire. The gap isn't about the job itself. It's about belonging: being included in informal networks, having visible career paths, receiving equitable sponsorship (not just mentorship), and seeing leaders who look like them. Without inclusive practices, diverse hiring becomes a revolving door.
Employee Resource Groups (ERGs) increase belonging and retention for underrepresented employees. Formal sponsorship programs (where senior leaders advocate for diverse talent) accelerate career progression. Equitable promotion criteria (structured, competency-based, not "who do leaders notice") ensure diverse employees aren't overlooked for advancement. Regular stay interviews (not just exit interviews) identify and address retention risks before employees leave.