A persistent claim in recruitment circles states that up to 75% of submitted resumes are automatically rejected by Applicant Tracking Systems (ATS) before human review. To evaluate the validity of this assertion, Enhancv conducted structured interviews with 25 U.S.-based talent acquisition professionals between September and October 2025. The full methodology and dataset are available in the complete ATS rejection study.
A very different operational reality is shown by the study: 92% of participants attested that their applicant tracking systems do not automatically reject applicants based on their format or content. Rather, the majority of rejection decisions are still made by hand, with very little automation outside of predefined eligibility criteria.
ATS Configuration and Rejection Mechanisms
Core Functionality
The primary functions of contemporary ATS solutions, like Workday, iCIMS, Greenhouse, Bullhorn, BambooHR, SuccessFactors, Teamtailor, and Phenom, are data aggregation and workflow management. To make applicant records searchable, parsing algorithms extract structured fields (contact information, work history, and abilities). Non-standard formatting-based parsing errors just lower the quality of the data for later human evaluation, they do not result in rejection.
Automated Rejection Prevalence
- Content/Design Auto-Rejection: 8% (2/25) of systems are configured to reject based on match thresholds (e.g., <75% keyword alignment or missing required skills). Both instances involve Bullhorn and BambooHR in high-volume environments.
- Eligibility-Based Auto-Rejection: 100% of respondents utilize knockout questions for binary criteria (work authorization, licensure, location, minimum education). Failure routes candidates to rejection or a low-priority queue, depending on configuration.
AI-Generated Fit Scores
AI-driven match scoring is available in 44% of reported platforms (primarily Lever, Greenhouse, Teamtailor, Phenom). Usage patterns:
- 36% treat scores as prioritization signals with mandatory manual verification.
- 8% apply hard thresholds for rejection.
- 56% either disable the feature or ignore it entirely.
Volume as the Primary Screening Bottleneck
The most important aspect in resume visibility is not algorithmic filtering, but rather application volume. Across 25 interviews, recruiters reported that entry-level and administrative positions often receive 400 to 600 applications within three to five days of posting. Remote customer-service positions frequently receive over 1,000 applications in the first week, whereas software-engineering openings, particularly hybrid or entirely remote, can receive 1,500 to 2,000 before screening begins. Senior or highly specialized positions are the exception, rarely exceeding 200 candidates and frequently accumulating over one to two weeks.
Faced with these numbers, talent acquisition teams routinely pause postings once 300 to 500 applications arrive, allowing them to focus on the initial wave. Interviews frequently start with this early cohort, meaning later submissions, regardless of qualification, may never receive review. 52%t of respondents explicitly stated that submission timing influences shortlist composition, with early applicants gaining a structural advantage in high-volume scenarios.
Recruiter Priorities in Resume Evaluation
When asked to identify the attributes that most influence their decision to advance a candidate, recruiters consistently emphasized structural clarity and scannability, cited by 92% as a primary concern. Prior experience relevance came in second at 88%, and 76% of respondents praised the organic incorporation of job-specific keywords. While 68% of respondents voted for consistent, professional layout free of unnecessary visuals, 72% preferred concise, bullet-point terminology over complex paragraphs. They also had clear preferences when it comes to length constraints – 64% expect one to two pages maximum, and 52% prioritize quantified achievements that demonstrate impact.
No recruiter mentioned font selection, decorative elements, or purported “ATS-optimization” templates as meaningful evaluation criteria. The consensus is unambiguous – resumes must be engineered for rapid human comprehension, not for speculative parsing compliance.
Origins and Persistence of the Rejection Myth
68% of recruiters first encountered the 75% rejection statistic via candidate anecdotes on LinkedIn or TikTok. 20% attributed that trend to career-coaching materials. The myth persists because of confirmation bias: silent rejections are more commonly attributed to faulty algorithms than to human capacity restrictions.
Implications for AI-Driven Recruitment Tools
The findings carry direct relevance for developers of AI recruitment solutions:
- Over-automation risk – Hard rejection thresholds, while efficient, exclude viable candidates when match algorithms misinterpret context (e.g., transferable skills).
- Human-in-the-loop design – Systems that surface AI scores as sortable metadata—rather than binary gates—align with current recruiter workflows.
- Volume-management features – Dynamic posting pauses, tiered review queues, and early-applicant flagging address the root cause of unseen applications.
Conclusion
Applicant Tracking Systems function as organizational infrastructure, not autonomous gatekeepers. Rejection is primarily due to human selection under volume pressure, which is supported by precisely defined eligibility filters. Job seekers looking to improve their chances should prioritize submission timing, structural clarity, and verifiable agreement with listed requirements over speculative formatting changes.
For companies developing or refining AI recruitment technologies, the statistics highlight the ongoing importance of human judgment and the limited appetite for totally automated disposition.
Enhancv provides AI-assisted resume and cover-letter tools designed for both ATS compatibility and human readability. The platform’s Job Tailoring module maps user experience to specific postings using recruiter-validated criteria.
