Diversity and Inclusion in Recruitment: Current Trends and Best Practices

The language around diversity has changed a lot in the past two years. Some companies renamed their programs. Some scaled them back. Some doubled down.

For recruiters, the useful question hasn’t changed: are we finding the best person for the job from the widest possible pool, and judging everyone on the same job-relevant things?

That isn’t politics. It’s good hiring. Here are the practices that hold up.

1. Widen the top of the funnel

You can’t hire people you never meet. If every intern comes from the same two or three campuses, every intern class will look roughly the same, with the same programs, networks and blind spots.

• Recruit across more schools, including colleges, polytechnics and universities outside your usual list.
• Recruit by program and skill, not school ranking. The best data science or mechatronics students might be at a school your team has never visited.
• Use co-op and internship programs as a pipeline. They let you see people do real work before a full-time offer.

Multi-campus sourcing used to need a big travel budget. It doesn’t anymore. We go deeper on this in Diversity Starts With Where You Recruit.

2. Write job posts for skills, not pedigree

Many job posts filter out great candidates before anyone reads an application.

• Remove requirements the job doesn’t need. “Two years of experience” on an entry-level role screens out the exact people you’re trying to reach.
• List the three to five skills that predict success, and say how you’ll assess them.
• Show the pay range. It’s now required in places like Ontario and British Columbia, and it signals fairness everywhere.
• Cut the jargon and the “rock star” language.

Then go further than the posting. When the Burning Glass Institute and Harvard Business School tracked companies that dropped degree requirements, most barely changed who they hired. Their conclusion: real change takes more than editing the job ad (HR Dive). What matters is how people actually evaluate candidates.

3. Structure your interviews

Unstructured interviews feel natural, and they’re one of the weaker ways to predict performance. A major 2022 review of decades of hiring research ranked structured interviews as the single best predictor of job performance among common selection methods (SIOP).

A structured interview means:

• The same core questions for every candidate for the same role
• Questions tied to the skills the job needs
• A scoring guide that defines a strong, average and weak answer
• Interviewers scoring on their own before they discuss

It’s fairer and it’s more accurate. You don’t have to choose.

4. Take bias out of screening

Bias in screening isn’t theoretical. In a well-known field experiment, researchers answered real job ads in Boston and Chicago with fictional resumes and randomly gave them white- or Black-sounding names. The resumes with white-sounding names got 50% more callbacks (J-PAL).

Practical fixes:

• Hide names, photos and addresses during the first screen where you can.
• Write down your skills criteria before you read the first application, then screen against them.
• If you use AI screening, ask the vendor how they test for bias and check your own pass-through rates. AI can repeat old patterns at scale. Amazon scrapped an experimental recruiting tool after it learned to downgrade resumes that included the word “women’s” (Reuters).

5. Make the process accessible

Small barriers quietly knock out good candidates.

• Offer accommodations up front, not only on request.
• Give virtual options for early rounds so students don’t pay for travel or miss class.
• Share the timeline and interview format in advance.
• Keep take-home tasks short. A three-hour assignment during exam week filters for free time, not talent.

6. Measure your funnel

You can’t improve what you don’t measure. For each stage, track:

• How many candidates come from each source and each campus
• Pass-through rates from application to screen, screen to interview, interview to offer and offer to acceptance
• Where strong candidates drop out, and why

If one source produces great hires and another produces none, you’ll see it. If one stage loses a whole group of candidates, you’ll see that too. Collect demographic data only where it’s legal and voluntary, and keep it separate from hiring decisions.

7. Be open about AI

Candidates are wary of AI deciding their fate. In a Pew Research Center survey, 66% of U.S. adults said they wouldn’t want to apply to an employer that uses AI to help make hiring decisions (Pew Research Center). The rules are tightening too. Since January 2026, Ontario requires employers with 25 or more employees to say in public job postings whether AI is used to screen, assess or select applicants (blogTO).

Tell candidates where AI is used. Keep a human accountable for every decision.

The bottom line

Inclusive recruiting isn’t a side program. It’s what good recruiting looks like: a wide search, clear criteria, a consistent process and honest data.