Greenhouse powers hiring at Airbnb, Stripe, Figma, and 7,500 other companies. A single formatting mistake can cost you an interview β here is exactly what to fix.
Greenhouse is the top-rated applicant tracking system on G2 in 2025, used by over 7,500 companies ranging from fast-growing startups to established mid-market firms. Unlike legacy enterprise systems, Greenhouse was built for structured hiring workflows, with collaborative scorecards, pipeline stages, and detailed reporting baked in. Its parser is more capable than older ATS systems, but it still has specific formatting requirements that trip up even experienced candidates.
The most important thing to understand about Greenhouse is that it does not just store your resume β it actively extracts data from it to populate candidate profiles. If your resume uses a multi-column layout, contains text inside headers or footers, or uses non-standard date formats, the extracted profile will be incomplete or wrong. Recruiters search and filter these parsed profiles, not the original document, so formatting errors directly reduce your visibility.
DOCX consistently produces cleaner extracted data than PDF in Greenhouse. Text-based PDFs work acceptably for simple layouts, but multi-column PDFs β even simple two-column designs β cause column content to merge into a single garbled stream.
Used by 7,500+ companies worldwide
Specific formatting decisions that cause extraction failures in Greenhouse
Greenhouse's PDF extractor reads text left-to-right in a single horizontal pass. A two-column layout causes text from both columns to interleave on the same line β your job title from column one merges with a date from column two mid-sentence. The parser then fails to identify any coherent sections. This affects even simple side-by-side layouts, not just complex designs.
Many candidates put their name, phone, and email in the page header for a clean visual design. Greenhouse ignores this content entirely β it processes the document body only. If your contact details live in a Word header or PDF header element, the extracted profile will have no contact information at all. Move all critical contact details into the main body text.
Greenhouse's date parser expects formats like 'January 2021' or 'Jan 2021'. Formats like '01/2021', '2021-01', or 'Q1 2021' are frequently misread or ignored, causing employment gaps to appear in your profile timeline. Recruiters filtering by experience duration will see inaccurate numbers. Abbreviations like 'Jan', 'Feb' work fine β the key is including the month, not just the year.
When you hyperlink the word 'LinkedIn' or 'View Profile' in Word, Greenhouse extracts only the display text, discarding the actual URL behind it. The candidate profile shows the word 'LinkedIn' with no clickable link. Paste the full URL as visible text: linkedin.com/in/yourname β this displays correctly and gets extracted as a proper link.
Actionable fixes that improve parsing accuracy and recruiter visibility
Create your resume in Microsoft Word or Google Docs using a single-column layout. Avoid text boxes, tables used for layout purposes, and any design that places content side by side. DOCX preserves semantic structure that Greenhouse uses to identify sections.
Greenhouse explicitly looks for a section labeled 'Skills' to populate the skills field in candidate profiles. A heading like 'Technical Competencies' or 'Core Strengths' may not be recognized. List your skills in this section as a comma-separated line or a simple bulleted list β no columns, no rating graphics.
Greenhouse's keyword search is case-insensitive but requires exact string matches. If the job posting says 'Salesforce CRM', write 'Salesforce CRM' β not just 'Salesforce' and not 'SFDC'. Spell out acronyms alongside abbreviations: 'SQL (Structured Query Language)' covers both search variants.
Greenhouse surfaces candidate profiles to hiring teams with parsed work history. Quantified bullets β 'reduced deployment time by 35%', 'managed $2M annual budget' β stand out immediately when recruiters scan extracted profiles. Vague bullets like 'improved processes' carry no weight.
Put your name, email, phone number, city, and LinkedIn URL in the first few lines of the document body, not in a Word header or footer. Format them as plain text separated by pipes or line breaks. This ensures Greenhouse extracts and links them correctly to your candidate record.
Yes, Greenhouse accepts text-based PDFs, but DOCX produces more reliable parsing. The critical issue is layout: a single-column PDF typically parses well, while any multi-column design causes the parser to merge content from different columns into garbled output. If you use PDF, test it with a plain single-column template and verify there is no content in page headers or footers.
Greenhouse does not use an automated scoring algorithm like some older ATS systems. Recruiters and hiring managers search the candidate database using keyword filters, then manually review profiles. This means your parsed profile data quality matters enormously. Incomplete extracted profiles β missing skills, garbled work history, no contact info β simply do not appear in searches for qualified candidates.
A single-column reverse-chronological DOCX file is the most reliable choice. Use standard section headings: Summary, Experience, Skills, Education. Put contact details in the document body, not the page header. Use 'Month YYYY' date formats throughout. Avoid tables, text boxes, columns, and any graphic elements. This structure gives Greenhouse's parser the cleanest possible input for extracting your profile data.