Resume screening is the most discussed and most misunderstood function of an ATS. Learn how modern screening actually works, why the old keyword-stuffing advice is outdated, and how to build a fairer, faster screening process.

Modern ATS platforms parse resumes into structured profiles rather than scanning for keywords. Skills, job titles, and years of experience are extracted and normalized for accurate comparison.
Define must-have and nice-to-have qualifications per role. The system highlights which requirements each candidate meets, speeding up the initial review without hiding anyone from view.
Machine learning models rank applicants by relevance to the job description, considering the full profile rather than isolated keywords. Recruiters see the most promising candidates first.
Structured screening criteria and consistent evaluation frameworks ensure every resume is assessed on the same basis, reducing the risk of unconscious bias in the initial review.
Review, tag, and advance or archive large batches of candidates with bulk actions. Handle high-volume roles without sacrificing thoroughness.
A persistent myth suggests that ATS software automatically rejects the majority of applicants before a human ever sees them. In reality, most modern platforms, including Draft, do not auto-reject anyone. They organize, parse, and present candidates so recruiters can make informed decisions faster.
The real problem is not the technology. It is process design. When a job posting attracts hundreds of applicants and the team lacks a clear screening framework, candidates sit unreviewed. The solution is structured criteria, not fewer applications.
When a resume arrives in Draft, the automated CV parser extracts data into a rich candidate profile. Work history is broken into positions with dates, skills are categorized, and education is standardized. This structured data powers the AI search engine, which lets recruiters query the entire candidate database using natural language.
Recruiters can set up screening questionnaires that candidates fill out during the application process. Responses appear alongside the parsed profile, giving reviewers a complete picture without opening a single attachment.
For high-volume roles, bulk actions let recruiters tag, advance, or archive groups of candidates at once. Combined with customizable pipeline stages, screening becomes a streamlined workflow rather than a bottleneck.
Most modern ATS platforms, including Draft, do not auto-reject candidates. They parse and organize applications so recruiters can review them efficiently. Auto-rejection only happens if your team explicitly configures knockout criteria.
Use a clean layout with standard section headers like Experience, Education, and Skills. Avoid tables, text boxes, and excessive graphics. Most importantly, focus on clearly describing your experience rather than stuffing in keywords.
Yes. Draft's parser handles PDF, Word, and other common formats with high accuracy, including multi-column layouts and creative designs.
Keyword filtering looks for exact matches and misses synonyms and context. AI screening understands that 'full-stack engineer' relates to both frontend and backend skills, providing more nuanced and accurate results.
Any tool reflects the criteria you set. Draft encourages structured, competency-based screening criteria that reduce bias. Standardized questionnaires and consistent evaluation rubrics further level the playing field.
The full technical breakdown of ATS mechanics.
Read moreSee how Draft organizes candidate profiles and communications.
Read moreBest practices for reducing bias throughout your hiring process.
Read moreCreate structured screening questionnaires for consistent evaluation.
Read moreDraft's AI-powered parsing and search give your team a faster, fairer way to screen candidates. Stop losing great people in the pile.
Get started freeNo credit card required