Shortlisting
The process of screening job applications to select the most qualified candidates for interviews, typically reducing a large applicant pool to 3-8 finalists.
What Is Shortlisting?
Key Takeaways
- ✓Shortlisting is the structured process of evaluating applications against predefined criteria to select a manageable number of candidates for interviews.
- ✓The average corporate job posting receives 250 applications, making shortlisting essential for efficient hiring (Glassdoor).
- ✓Recruiters spend an average of 7.4 seconds on initial resume review, which highlights the need for clear screening criteria (Ladders, 2023).
- ✓Effective shortlisting uses a scoring matrix based on job requirements, not gut feeling or superficial signals.
- ✓Poorly designed shortlisting processes introduce bias, eliminate strong candidates, and waste interview time on poor fits.
Shortlisting is the screening stage that sits between receiving applications and conducting interviews. It's the filter. Out of 250 applicants (the average for a corporate role, per Glassdoor), shortlisting narrows the pool down to 4-6 candidates who meet the role's essential requirements and are worth the time investment of formal interviews. The term "shortlisting" literally means creating a short list of candidates from a long list of applicants. It sounds simple. It isn't. Done well, shortlisting is a structured, criteria-based evaluation that gives every applicant a fair assessment against the same standards. Done poorly, it's a rushed, biased scan where recruiters make snap judgments based on university names, company logos, or formatting preferences. The quality of your shortlist directly determines the quality of your hire. If the shortlist is weak, even the best interview process can't save the outcome. If strong candidates get eliminated during shortlisting because of unclear criteria or unconscious bias, you'll never know what you missed.
Where shortlisting fits in the recruitment funnel
The typical recruitment funnel has five stages: sourcing and attraction (getting applications), shortlisting (screening and filtering), interviews (assessing shortlisted candidates), selection (choosing the finalist), and offer and onboarding. Shortlisting is the biggest volume reduction step. You might go from 250 applicants to 20-30 after an ATS keyword screen, then to 8-12 after a recruiter review, then to 4-6 after a hiring manager review. Each stage has different screening criteria and different people involved.
Shortlisting vs screening
The terms are often used interchangeably, but there's a useful distinction. Screening is the binary pass/fail check: does the candidate meet the minimum requirements? Right to work authorization, required certifications, willingness to relocate, minimum years of experience. Shortlisting is the comparative ranking: among the candidates who passed screening, which ones are the strongest matches? Screening eliminates the unqualified. Shortlisting ranks the qualified.
How to Set Shortlisting Criteria
The criteria you use to shortlist candidates should come directly from the job description and be agreed upon with the hiring manager before applications arrive. Setting criteria after reading applications invites bias.
Essential vs desirable criteria
Split your criteria into two categories. Essential criteria are non-negotiable requirements: the candidate must have them, or they're automatically screened out. Examples: a valid nursing license, 5+ years of financial modeling experience, proficiency in Python. Desirable criteria are nice-to-haves that differentiate candidates who all meet the essentials. Examples: experience with a specific tool, industry background, leadership experience. A candidate who meets all essential criteria but none of the desirable ones should still make the longlist. A candidate who has impressive desirable qualifications but lacks an essential requirement should not. This distinction prevents the common trap of ranking candidates by how impressive their resume looks rather than how well they meet the role's actual requirements.
Building a shortlisting scorecard
Create a matrix with criteria as rows and candidates as columns. Score each candidate on each criterion (a simple 0-1-2 scale works: 0 = doesn't meet, 1 = partially meets, 2 = fully meets). Weight essential criteria higher than desirable ones. Total the scores and rank candidates. This approach does three things: it forces consistent evaluation, it creates documentation for bias audits, and it gives the hiring manager a transparent rationale for why specific candidates were selected or rejected. Don't skip the scorecard. Without it, shortlisting decisions become "I just got a feeling about this one," which is bias wearing a polite hat.
Criteria to avoid
Don't use university prestige as a shortlisting criterion unless the role genuinely requires knowledge only available from a specific academic program. Don't use company brand names ("must have worked at a Big Four firm"). Don't use gaps in employment history as a negative signal without context. Don't screen based on address or commute distance unless the role is strictly on-site with no flexibility. Don't disqualify candidates for being "overqualified" without investigating their motivation. Each of these criteria disproportionately impacts specific demographic groups and introduces systemic bias.
The Shortlisting Process Step by Step
A structured shortlisting process reduces bias, improves speed, and produces better interview slates.
Step 1: ATS-assisted pre-screening
Most applicant tracking systems can automatically filter applications based on knockout questions ("Do you have a valid CPA license?") and keyword matching (required skills, certifications, tools). This automated pass removes clearly unqualified applicants and reduces the pool by 50-75%. Roughly 75% of resumes don't pass ATS filtering (Jobscan, 2024). While some of these are genuinely unqualified, ATS filters also reject candidates whose resumes aren't formatted for machine reading. This is a known limitation. Don't rely solely on ATS filtering for critical roles.
Step 2: Recruiter longlist review
The recruiter manually reviews the applications that passed ATS filtering. This typically takes 7-15 seconds per resume for an initial scan and 2-3 minutes for a closer read of promising candidates. The recruiter creates a longlist of 15-25 candidates who meet the essential criteria and show potential. During this stage, the recruiter should use the scorecard, not their intuition. It's tempting to speed through hundreds of resumes and pick the ones that "feel" right, but that feeling is usually pattern recognition biased toward candidates who look like people who were previously hired.
Step 3: Hiring manager shortlist review
The hiring manager reviews the longlist and selects the final shortlist (typically 4-6 candidates) for interviews. The hiring manager brings domain expertise the recruiter may lack: they can assess project relevance, evaluate technical depth from work samples, and identify cultural indicators that matter for the team. Present the longlist with scorecard data so the hiring manager understands why each candidate was included. If the hiring manager rejects candidates who scored well, ask for specific reasons. This calibration helps the recruiter refine future shortlisting for this role.
Step 4: Shortlist notification
Notify shortlisted candidates promptly (within 48 hours of the decision) and provide clear next steps: what the interview process looks like, how many rounds, timeline, and who they'll meet. Equally important: notify rejected candidates. Even a brief, polite rejection email is better than silence. Candidates who apply and hear nothing form negative impressions of your employer brand. The average time to shortlist notification is 6-12 business days after the application deadline (SHRM). Faster is better. Top candidates are interviewing elsewhere.
Reducing Bias in Shortlisting
Shortlisting is one of the most bias-prone stages in the hiring process because decisions are made quickly, often by a single person, based on incomplete information.
Name and demographic bias
A 2023 study by researchers at UC Berkeley and the University of Chicago found that resumes with traditionally white-sounding names received 9.5% more callbacks than identical resumes with Black-sounding names. This bias hasn't improved meaningfully since the landmark 2004 Bertrand and Mullainathan study. Anonymous or blind shortlisting (removing names, photos, addresses, and university names from applications before review) reduces this bias. Several ATS platforms now offer built-in anonymization features.
Affinity bias
Recruiters and hiring managers tend to favor candidates who share their background: same university, same previous employer, same hobbies listed on the resume. This isn't malicious. It's pattern matching. But it narrows the talent pool and reinforces homogeneity. Mitigation: use structured scorecards (not free-form evaluation), rotate shortlisting responsibilities among team members, and require multiple reviewers for the shortlist decision.
The halo and horn effects
One impressive credential (a prestigious employer, a famous university) creates a halo effect that inflates the recruiter's perception of everything else on the resume. Conversely, one perceived weakness (a gap year, a stint at a company that failed) creates a horn effect that colors the entire evaluation. The scorecard helps because it forces criterion-by-criterion evaluation rather than an all-or-nothing "thumbs up or thumbs down" judgment. Evaluate each criterion independently before calculating a total score.
Technology and AI in Shortlisting
AI-powered shortlisting tools are growing fast, but they're not without controversy.
How AI shortlisting works
AI shortlisting tools analyze applications using natural language processing and machine learning. They compare candidate profiles against the job requirements and rank applicants by predicted fit. Some tools go beyond keyword matching: they assess semantic similarity (understanding that "led a P&L" and "managed budget responsibility" mean similar things), predict candidate success based on patterns from previous successful hires, and score resumes on multiple dimensions simultaneously. Tools like HireVue, Pymetrics, and Eightfold.ai offer AI-driven shortlisting capabilities that can process hundreds of applications in minutes.
Risks of AI shortlisting
AI learns from historical data, which means it can learn historical biases. Amazon's famously scrapped internal AI recruiting tool because it penalized resumes that contained the word "women's" (as in "women's chess club" or "women's college"). The EEOC's 2023 guidance made clear that employers are liable for discriminatory outcomes from AI tools, even if the tool was built by a vendor. NYC's Local Law 144 (effective July 2023) requires annual bias audits for AI hiring tools used in New York City. The EU AI Act classifies recruitment AI as "high-risk," requiring conformity assessments and human oversight. AI shortlisting should augment human judgment, not replace it. Use it as one input, not the sole decision-maker.
Shortlisting Best Practices
Practical recommendations for building a shortlisting process that's fast, fair, and effective.
- Define all shortlisting criteria before the job is posted, not after applications arrive
- Use a scoring matrix for every role, even if it's a simple 3-point scale
- Require at least two people to independently review the shortlist for any role at manager level or above
- Set a maximum shortlist size (4-6 for most roles, 8-10 for high-volume positions) to prevent interview overload
- Communicate decisions to all candidates within 2 weeks of the application deadline
- Track shortlisting demographics quarterly to identify systemic bias patterns
- Calibrate with hiring managers after every search to improve criteria accuracy for future roles
- Keep detailed records of why each candidate was advanced or rejected for compliance and audit purposes
Frequently Asked Questions
How many candidates should be on a shortlist?
Who should do the shortlisting?
Should I tell rejected candidates why they weren't shortlisted?
How do I shortlist when 50+ candidates all look qualified?
Is blind shortlisting effective?
Written by Adithyan RK