A popular role can draw more applications than anyone can read closely. Nobody reads them all with equal care, so the order a recruiter reads them in, and the rules used to drop people, decide who gets a chance. Resume screening software makes those rules explicit and applies them to everyone.
That is its value and its risk. A good rule applied at scale saves days. A bad rule applied at scale quietly rejects your best applicants and can land you in breach of a hiring law. This guide explains the five main approaches, how to compare vendors and what to check on bias and compliance before you buy.
Disclosure: Cohesyve publishes this guide and sells one of the approaches described. Other vendors are described from their own public pages as of October 2026; check their sites for prices.
What is resume screening software?
Resume screening software reviews job applications and decides, or recommends, which applicants move forward. It ranges from simple keyword filters inside an applicant tracking system (ATS) to AI models that grade each CV against the job's requirements.
Most tools sit at the top of the hiring funnel, between "applied" and "first interview". Some only sort the list for a recruiter to read. Others reject applicants automatically, which raises the stakes on accuracy and on the law.
What are the main approaches to applicant screening?
There are five, and most hiring teams combine two or three of them.
| Approach | How it works | What it checks | Main weakness |
|---|---|---|---|
| Keyword parsing | The ATS reads each CV and matches words and phrases against a list | That the CV mentions the right terms | Rewards keyword-stuffing; misses people who describe the same skill differently |
| AI ranking | A model compares each CV with the job requirements and grades or orders applicants | How closely the CV matches the job | Can learn bias from past data; hard to explain without a rationale per grade |
| Knock-out questions | Yes/no application questions reject anyone who fails a hard requirement | Eligibility: work rights, licences, shift availability, location | Only as accurate as the candidate's answer; easy to over-use |
| Skills tests | Candidates complete a timed test of a skill or aptitude | A sample of ability on one skill | Generic tests may not reflect the real job |
| Job simulations | Candidates do a realistic task from the role, such as handling a customer chat or analysing data | How someone performs the actual work | Asks more of the candidate than a CV upload |
The first three judge the application. The last two judge the person. That distinction matters more than any feature list, because a CV is a self-description and a test is evidence.
Practise this before it counts
Run an AI job simulation of the real work and get a scored report on where you stand. 3 free tests, no card.
How does AI resume screening work?
AI resume screening reads each CV, extracts skills, roles and experience, and compares them with the job's requirements to produce a grade, a score or a ranked list. Newer tools also message applicants to fill gaps the CV leaves open.
Vendors describe it in their own terms. Workday says HiredScore "compares applicants to requirements, with rationale behind each grade" and that demographic data is never used in scoring. Workable describes an AI agent that evaluates each inbound applicant against up to 14 job criteria, then messages them to verify interest and fill gaps, with the recruiter able to override any call. Manatal's AI recommendations suggest the best candidates for a job and enrich profiles with information beyond the resume.
The speed is real. The open question for any buyer is what the model is actually rewarding. In 2018 Reuters reported that Amazon had dropped an experimental CV-screening tool. It had learned from CVs submitted mostly by men and marked down women applying for technical roles. Any model that learns from past hiring decisions can inherit their patterns.
Our guide to AI recruitment tools covers the wider category, including sourcing and scheduling.
What is the best applicant screening software?
The best applicant screening software is the one that matches your hiring volume and your roles, explains every decision it makes, and lets a person overrule it. No single tool is best for every team.
| Vendor | Public positioning (Oct 2026) | Screening approach | Pricing |
|---|---|---|---|
| Workday (HiredScore) | AI recruiting inside the Workday HR suite | AI grades applicants against requirements, with a rationale for each grade | Check their site |
| Workable | ATS and HR software with an AI recruiting agent | AI screens inbound applicants against job criteria and messages them; assessments and scorecards | Check their site |
| Greenhouse | ATS, sourcing and voice AI interviewing | AI-powered ATS; Real Talent flags fraud and prioritises best-fit applicants | Check their site |
| Manatal | ATS and recruitment CRM for companies and agencies | AI recommendations and profile enrichment | Check their site |
| HireVue | Interviews and assessments at scale | AI Interviewer gives every applicant a conversation; tests and simulations | Check their site |
| Paradox | Conversational hiring for frontline and hourly roles | Chat-based apply, screening questions and scheduling | Check their site |
| Harver | Volume hiring for contact centres, retail and BPOs | Job-relevant screening questions, predictive and game-based assessments | Check their site |
| Cohesyve (our product) | From a job description to a ranked shortlist | Resume pre-vetting on must-haves, then AI job simulations of the real work | From $199/month billed annually; 10 free candidates |
How to choose. If you already run Workday, Workable or Greenhouse, start with the screening built into your ATS, because it sees every applicant without extra integration. If you hire hourly staff at volume, a conversational tool such as Paradox, or an assessment-led one such as Harver or HireVue, will move people faster than CV review. If CVs are a poor guide to ability in your roles, put a skills test or job simulation early and use CV screening only for hard requirements. For alternatives to two of these, see our pages on HireVue alternatives and Harver alternatives.
Is CV screening software accurate?
CV screening is accurate at checking facts a CV states clearly, such as a licence, a degree or a number of years in a role. It is much weaker at predicting how well someone will do the job, because a CV shows what a candidate chose to write.
Three problems recur:
- Inflated or AI-written CVs. AI writing tools make it easy to tailor a CV to any job description, so keyword match tells you less than it used to.
- False negatives. Strong candidates with unusual career paths, gaps or plain writing get filtered out, and you never see them.
- Unclear reasons. If a tool cannot show why it rejected someone, you cannot spot a bad rule or explain the decision.
The practical answer is to narrow what you ask a CV to do. Use it to confirm must-haves, then judge ability with evidence. Our job simulations guide explains how to test the real work, and our screening checklists list the must-haves for common roles.
What bias and compliance checks should you run before buying?
Before you switch on automated screening, confirm that the tool's decisions are job-related, explainable and audited for adverse impact, and check which AI hiring laws cover the places you hire. This section is a plain-English summary, not legal advice.
Adverse impact. Under the US Uniform Guidelines (29 CFR 1607.4), a selection rate for any race, sex or ethnic group that is below four-fifths (80%) of the rate for the highest group is generally treated as evidence of adverse impact. Smaller gaps can still count if they are statistically and practically significant. Ask every vendor how often they test for this and whether you can run the numbers on your own applicants.
New York City. Local Law 144 covers automated employment decision tools used on job candidates and employees who live in the city. Enforcement began on 5 July 2023. An employer or employment agency may only rely on a bias audit for one year, must publish a summary of the results, and must notify candidates 10 business days before using the tool, with instructions for requesting a reasonable accommodation. These duties fall on the employer or employment agency; a vendor may commission an audit of its own tool. Our NYC Local Law 144 guide covers the detail.
Illinois. HB 3773, effective 1 January 2026, amends the Illinois Human Rights Act. It bars using AI that has a discriminatory effect in recruitment and hiring, bars using zip codes as a proxy for protected characteristics, and requires employers to give notice when AI is used for these decisions.
Colorado and the EU. Colorado's 2024 AI Act has been replaced by SB 26-189, effective 1 January 2027. The EU AI Act treats recruitment AI as high-risk, and those obligations apply from 2 December 2027.
Our explainer on automated employment decision tools covers how to evaluate and govern a tool, and the AI hiring laws guide sets out the obligations law by law.
Questions to ask every vendor:
- Does the tool reject applicants automatically, or only rank them for a person to review?
- Can it show the reason for each decision, down to the line of the CV?
- Has it had an independent bias audit, and can we see the summary? (Eightfold, for example, publishes third-party bias audit results for its matching model.)
- Which data does it use beyond the CV, and can we switch that off?
- Can a recruiter restore a rejected applicant in one step?
- What notice and accommodation options does it give candidates?
How do you set up resume screening that works?
Set it up so that software removes only clear mismatches and every remaining applicant is judged on the same evidence.
- Write three to five must-haves. Requirements a person must meet on day one, such as work rights, a licence or shift availability. Keep preferences out of this list.
- Screen only on those. Use knock-out questions or a resume check for must-haves, and do not reject anyone for a missing keyword.
- Test everyone who passes. Send a short skills test or job simulation to every eligible applicant, so ranking rests on what they can do.
- Keep a person in the loop. Review the shortlist, spot-check the rejections each week and restore anyone screened out by mistake.
- Measure it. Track selection rates by stage and compare the tool's top-ranked applicants with who succeeded in interview.
For high-volume roles, our high-volume recruiting strategies post covers the rest of the funnel.
Where does Cohesyve fit?
Cohesyve takes a job description, sources candidates, does the outreach, puts them through AI job simulations of the real work and returns a ranked shortlist. Its resume pre-vetting checks each CV against your must-haves, knocks out only clear misses and shows the line from the CV that decided each requirement, so a recruiter can restore anyone with one click. The pre-vetting walkthrough shows the steps. Pricing is a subscription from $199 a month billed annually, and a new account includes 10 free candidates.
Common questions
Is resume screening software the same as an ATS? No. An ATS manages the whole hiring pipeline, and many include screening features. Standalone screening tools plug into an ATS or replace part of it.
Is AI resume screening legal? In most places, yes, but some laws add duties. New York City requires a bias audit and candidate notice for automated tools, Illinois bars discriminatory AI use and requires notice, and Colorado and the EU have rules taking effect in 2027. Existing anti-discrimination law applies whether a person or software makes the decision.
Should software reject applicants automatically? Only on clear, objective must-haves that the candidate has stated, such as work authorisation. For anything that needs judgement, rank applicants and let a person decide.
What is the difference between CV screening and skills testing? CV screening checks what a candidate says about themselves. Skills testing and job simulations check what they can do. Many teams use both, in that order.
How do I compare screening tools fairly? Run the same live role through each tool on trial, then compare the shortlists with your own judgement and with who does well at interview. Our comparison of pre-employment assessment tools covers the testing side.