On this page11 sections
Key takeaways
- Greenhouse parses resumes under 2.5 MB into profile fields; bigger, image-based, or heavily formatted files just sit as attachments (Greenhouse Support).
- Greenhouse is the most-used ATS overall at 19.0% of 13,130 companies in Jobscan's 2026 report, though Workday leads the Fortune 500.
- Per Greenhouse's Fall 2026 benchmark, every 1,000 US applications produced about 75 screens, 32 first interviews, and 5 hires.
- The only published automatic rejection comes from employer-configured auto-reject rules on Yes/No and multiple-choice screener questions, not resume keyword scoring.
- You see application statuses in MyGreenhouse only when the employer opts to share them.
The moment you hit submit, the Greenhouse ATS tries to parse your resume into structured fields on a candidate profile, stores the original file as an attachment, and places you in an Application Review queue. From there, an employer-configured auto-reject rule can bounce you out instantly, or a recruiter (sometimes nudged by an assistive AI match score) decides whether you move on to the interview plan. If your file is over 2.5 MB, image-based, or full of tables and graphics, the parse fails and a recruiter sees an attachment with nothing auto-filled.
Why should you care about the plumbing? Because Greenhouse is the most-used ATS in Jobscan’s 2026 analysis of 13,130 companies, detected at 19.0% of them, ahead of Workday (17.5%) and Lever (15.6%). Workday dominates the Fortune 500, but for startups, scale-ups, and mid-market US employers, applying through Greenhouse is often the default way in.
The whole pipeline at a glance
Your application passes through eight steps, from job board to offer or rejection, and each one is documented in Greenhouse’s own support library.
- You click Apply on the employer’s careers page or a Greenhouse-hosted job board at boards.greenhouse.io.
- A source-tracking token records where you came from.
- You fill out the application form: contact details, resume upload, employer questions.
- Greenhouse parses your resume (if the file cooperates) and creates your candidate profile.
- Auto-reject rules, AI match scores, and fraud checks run before any human looks.
- A recruiter works through Application Review and clicks Advance or Reject.
- Advanced candidates hit employer-defined interview stages, each with its own scorecard.
- The process ends at an Offer stage, or a rejection email, if the employer configured one.
Where you’re actually applying: the job board and the application form
Most “Apply” buttons open a Greenhouse application form, either hosted at boards.greenhouse.io or embedded directly in the employer’s careers site. Employers increasingly pick the embedded option: an iFrame that keeps you on the company’s site while you browse jobs and apply, so the URL in your browser may never say “greenhouse” at all.
The form is configurable per job. Employers decide which questions and fields you see when they style their job board, and they can add custom questions, the Yes/No and multiple-choice screeners about work authorization, location, or salary expectations. Many also ask for extras like a cover letter or portfolio; anything over 100 MB belongs on a third-party link instead, per Greenhouse’s upload guidance.
One detail most applicants miss: Greenhouse tracks where you came from with its own token, not analytics tags. A gh_src= parameter in the URL identifies the source (job board, referral link, or careers page) and it’s added to your profile automatically when you apply through a tracking link. Recruiters see your source on your profile and report on which channels produce hires.
What the Greenhouse ATS does with your file the moment you submit
Resume parsing turns your file into structured profile fields (name, contact details, employers, job titles) so recruiters can search and filter you, not just open you. It only works under specific conditions.
Greenhouse’s parser handles files up to 2.5 MB. Uploads can reach 100 MB and the accepted formats are .doc, .docx, .pdf, .rtf, and .txt, but anything over the 2.5 MB parsing threshold uploads “successfully” and then quietly goes unparsed. A graphic-heavy PDF with logos and photos can blow past that limit without you noticing.
The documented failure causes are worth reading once, slowly:
- Contact info stuck in a header, footer, or text box
- Tables and multi-column layouts
- Photos, graphics, or word art, and image-only “scanned” resumes, which parse nothing
- Abbreviated job titles; Greenhouse’s own example is “Sr. Account Exec” instead of Senior Account Executive
- Company names without legal identifiers, like “Inc.” or “LLC”
- Placeholder text such as “Company 1,” which the parser treats as fake data
When parsing fails, the file “has only been attached to the candidate,” and a human has to type your details in by hand. That’s not always fatal (recruiters can open your attachments directly during review) but an unparsed profile can’t be searched or matched properly, so you’ve handed the system a worse version of yourself. Greenhouse’s portal guidance recommends a plain, text-based PDF for the best parse. (For the general mechanics, see what an applicant tracking system actually reads.)
How the Greenhouse ATS screens your application before a human reads anything
Three automated gates can act before a recruiter opens your file: auto-reject rules, AI match scores, and fraud checks. None of them works the way keyword-filtering myths suggest.
The only automatic rejection Greenhouse publishes is employer-built auto-reject rules tied to application questions. Per Greenhouse’s documentation, employers add a rule to a job post, choose a Yes/No, single-select, or multi-select question, and define the answer that triggers automatic rejection, with a rejection reason and an optional automated email. That’s why some candidates get rejected minutes after applying. Recruiters aren’t even notified about auto-rejected applicants, and the feature exists only on Plus and Pro subscription tiers, so plenty of employers never use it.
Talent Matching is the AI layer, and it’s more conservative than the headlines imply. It generates match scores from recruiter-defined calibrations (usually four to six weighted skills) and sorts candidates into five bands, from Strong Match down to Needs Manual Review. Greenhouse’s product documentation, updated October 5, 2026, states plainly: “Talent Matching is assistive AI, not automated-decision-making. It does not automatically advance or reject candidates.” Only candidates sitting in the Application Review stage get scored, and where employers enable it, you can opt out and request a manual, non-AI review.
Then there’s fraud. Greenhouse’s Real Talent suite includes identity verification powered by CLEAR, a government-ID and selfie check running “over 60 security signals”, on Plus and Pro tiers. Verification results land on your profile to inform recruiter decisions, and verified candidates can be filtered and prioritized, but the check doesn’t reject you by itself. In May 2026, Greenhouse also announced it would acquire Ezra AI Labs, a voice-AI interviewing platform, which tells you where structured first-round screening is headed.
Your first human gate: the Application Review stage
Every Greenhouse interview plan automatically starts with an Application Review stage, and that’s where a recruiter first seriously looks at you. Recruiters get an “Applications to Review” dashboard and work through candidates one by one.
What they see: your parsed profile and original attachments, your source, the date you applied, your answers to application questions, and your history on other jobs at the same company. If another profile matches your name, email, phone, or LinkedIn URL, the system flags you as a possible duplicate so the records can be merged. Recruiters can also set review priority, the documentation specifically mentions prioritizing referrals, internal applicants, agency submissions, and “Dream Job” candidates flagged from the portal.
The recruiter’s actions are Advance or Reject. Rejecting requires picking a reason and optionally sending an email now or scheduling it for later. They can also mark you as spam, which auto-rejects future applications from your email address. There’s no elaborate scoring at this stage; it’s a person making a fast judgment call with your parsed profile on screen and your actual resume a click away.
If you advance: stages, self-scheduling, and scorecards
Advancing drops you into the stages the employer built, typically a recruiter phone screen, then interviews with the team, ending at an Offer stage. The middle is entirely customizable: Greenhouse’s interview plan docs confirm each plan “starts with the Application Review stage,” “ends with an Offer stage,” and everything between is up to the employer, including optional take-home tests and hiring manager review stages.
Scheduling is more self-service than it used to be. Recruiters can send a self-schedule request that lets you pick a single time slot from interviewer availability; once you choose, the interview books itself and video links from Zoom, Microsoft Teams, or Google Meet attach automatically.
Each interview gets a scorecard, and by default interviewers can’t skip them. Interviewers rate you on the role’s focus attributes using a five-point scale (from “Definitely Not” to “Strong Yes”) then give an overall recommendation answering one question: did the candidate pass the interview? This is the structured-hiring pitch: everyone gets assessed against the same preset criteria rather than vibes. Private notes stay hidden from the hiring team unless admins grant access, and scorecard visibility between interviewers can even be switched off to reduce bias.
What you see while all this happens
You see very little, and that’s by design: the MyGreenhouse portal shows application statuses only if the employer shares them. If a company doesn’t share updates, you’ll literally see “Not sharing updates on an application.” Applications show as Active or Inactive, and the portal “doesn’t show every internal hiring step.”
The portal isn’t just for checking your status, though. You can store a profile to autofill future applications with one click, set job alerts, and use your monthly Dream Job flag, one application per month marked as your top choice, which surfaces it to that employer’s hiring team. One flag, no undo, so spend it deliberately. Sign-in takes a single-use email code or a Google, Apple, or LinkedIn account.
Rejection emails are not automatic by default. Employers decide whether sending one is required, optional, or skipped entirely, and whether it goes out instantly or on a delay. Meanwhile, your data has an expiration clock: Greenhouse’s data retention rules start when you’re hired or finally rejected, after which the system can automatically anonymize your personal data, a privacy-compliance feature applied retroactively. The pattern is similar to what candidates face in Lever’s application flow: visibility is the employer’s choice, not yours.
Your actual odds, according to Greenhouse’s own data
Greenhouse publishes funnel benchmarks from its own platform, and its Fall 2026 edition (161.8 million applications across 9,473 US companies, year-to-date through August 2026) puts applications per hire at 194, with time to fill at 62 days.
Of every 1,000 applications, only 32 ever reach a first interview, and about 5 end in a hire. Source: Greenhouse Recruiting Benchmarks, Fall 2026 edition.
Company size changes your odds a lot. Enterprise openings averaged 171 applications per hire in the same report, while SMB openings averaged 474, more than twice the pile for the same single opening. The trend is at least bending your way: hires rose 27.7% from January to August 2026 while applications stayed flat, though one in three openings still closed with no hire at all.
Independent data corroborates the scale. Ashby’s analysis, covered by HR Dive in May 2026, found open roles now average more than 300 applications, roughly triple the 2021 level. Greenhouse’s own March 2026 benchmark report, cited by HeroHunt.ai, counted 244 applications per job posting in 2025 (up 111% since 2022) and found just 69.8% of jobs closed with a hire. Different datasets, same conclusion: the volume is real, and waiting out the process is normal.
The flow, stage by stage, and where you still have leverage
The table below maps every stage to what the system does, what a recruiter actually sees, and what you can do about it.
| Stage | What the Greenhouse ATS does | What a recruiter sees | What you can do |
|---|---|---|---|
| Submit | Attaches your file, attempts parsing, records your source token | A new-application notification (unless auto-rejected first) | Use a text PDF under 2.5 MB; apply through links that credit the right source |
| Screening | Runs auto-reject rules on question answers; optional AI match scores; fraud and ID checks | Score bands, verified badges, flagged duplicates | Answer screener questions accurately, they’re the actual automated gate |
| Application Review | Queues your profile; flags possible duplicates; applies review priority | Parsed fields, attachments, source, answers, your other applications | Use your monthly Dream Job flag; make sure the parsed fields are accurate |
| Interviews | Sends self-schedule requests; requires scorecards per interview by default | Attribute ratings and overall recommendations on each scorecard | Pick a slot quickly; watch for portal tasks if the employer enables them |
| Decision | Moves you to Offer or rejected; starts the retention clock; can anonymize you later | Rejection reasons, offer approvals, source reports | Expect the configured email if any; don’t expect portal detail |
A few takeaways worth repeating. First, the fastest way to fail is technical, not competitive: an oversized, image-based, or table-heavy resume that never parses. Second, screener questions carry more automated weight than any formatting trick, because they’re the only input the published automation reads.
Third, speed and completeness still count, since recruiters working through a queue of hundreds reward early, parseable applications. And if you’re applying through Workday instead, the mechanics differ more than you’d expect, see applying through Workday: the candidate experience.
FAQ
Does the Greenhouse ATS automatically reject resumes? No, the published automation only auto-rejects based on answers to Yes/No and multiple-choice application questions, which employers configure themselves. There’s no resume keyword-scoring gate, and Greenhouse states its Talent Matching AI “does not automatically advance or reject candidates.”
What file format works best for Greenhouse applications? A text-based PDF under 2.5 MB, in one column, without tables, headers, footers, or images. Greenhouse accepts .doc, .docx, .pdf, .rtf, and .txt files up to 100 MB for upload, but the parser stops working above 2.5 MB.
Why did I get rejected minutes after applying through Greenhouse? An auto-reject rule fired on one of your screener answers, and the employer enabled an automatic rejection email. These rules run before any recruiter sees your application, and recruiters receive no notification about auto-rejected candidates.
Does Greenhouse use AI to screen candidates? Only assistively. Talent Matching generates match scores from recruiter-defined skill calibrations and surfaces candidates in Application Review, but recruiters make every decision, and you can opt out where the employer enables that control.
Will I know if I was rejected in Greenhouse? Only if the employer sends a rejection email or shares statuses in MyGreenhouse, neither is guaranteed. Many candidates see an application stay Active indefinitely or flip to Inactive with no explanation.
Can I reapply to the same company through Greenhouse? Yes. Greenhouse flags possible duplicates matching on name, email, phone, or LinkedIn URL so records can be merged, meaning reapplying updates your history rather than creating a stranger. Note that retention rules may have already anonymized data from older rejections.
How many people apply through Greenhouse per job? Far more than get a careful read: Greenhouse’s Fall 2026 benchmark counted 194 applications per hire in the US, ranging from 171 per hire at enterprises to 474 at small businesses.
Sources
- Greenhouse Support: Unsuccessful resume parse
- Greenhouse Support: Supported formats for resumes, cover letters and other candidate uploads
- Greenhouse Support: Embed a Greenhouse job board on your career site
- Greenhouse Support: Tracking a candidate’s source
- Greenhouse Support: Auto-reject
- Greenhouse Support: Talent Matching
- Greenhouse Support: Greenhouse identity verification overview
- Greenhouse Support: Application review stage
- Greenhouse Support: Interview plan overview
- Greenhouse Support: Scorecards FAQ
- Greenhouse Support: Candidate self-scheduling overview
- Greenhouse Support: MyGreenhouse FAQ for Candidates
- Greenhouse Support: Configure a data retention rule
- Greenhouse: Recruiting Benchmarks, The Hire Standard, Fall 2026 edition
- Greenhouse Newsroom: Greenhouse has entered into a definitive agreement to acquire Ezra AI Labs (May 5, 2026)
- Jobscan: 2026 ATS Usage Report / Fortune 500 ATS usage (published August 20, 2026)
- HR Dive: Recruiters see job applications triple to more than 300 per role (May 13, 2026)
- HeroHunt.ai: Recruiting funnel conversion rates 2026, citing Greenhouse’s March 2026 Benchmark Report
Written by ResumeFlex. Questions or corrections? Contact ResumeFlex. Read our source and correction standards.
Filed under
ATS & Parsing