CV and ATS glossary

How to use this glossary

Each term below has one definition of one or two sentences, written to stand on its own when quoted. The sections group the terms by who uses them; the full alphabetical list is under Terms, and every entry has its own anchor, so you can link to a single definition.

Why the words matter to a parser

Workday, Greenhouse, Lever, Taleo and SuccessFactors do not read a CV; they parse it. A parser looks for headings it knows, dates it can order and keywords it can match against the job description. A CV that says "Professional journey" where a parser expects "Experience" gives it nothing to hold on to.

Why the words matter to a recruiter

A recruiter reads what the parser passed through. "Results-driven" and "proven track record" are the filler this glossary names; a quantified bullet in XYZ or PAR form is the alternative. Knowing the names makes it easier to see which one a line is.

Parsers and applicant tracking systems

Five terms describe the software between you and the recruiter, in the order a CV meets them.

1. Applicant tracking system: the database and workflow a company hires through.
2. Parser: the component that turns your file into fields.
3. Keyword matching: how the parsed CV is scored against a job description.
4. Parser safety: a property of the CV, how reliably parsers read it.
5. Round-trip test: the check that proves parser safety for one file.

What a parser cannot do

No documented parser in Workday, Greenhouse or Lever detects AI-written text, ranks candidates on writing style, or reads text placed in images, headers or side-by-side columns. A parser will not penalise machine text, and it will not rescue a good line hidden in a sidebar.

Writing the line

Six terms describe the bullet itself. The examples are illustrations, not figures from any real CV.

Which formula to use

XYZ and PAR produce the same line in a different order. XYZ puts the result first and suits a CV, where the reader scans the start of each line. PAR tells a short story and suits an interview answer. Both need a real number; neither creates one.

Provenance and verification

These are the terms behind the cited CV. They come from three fields: archival science (provenance), academic writing (citation) and machine learning (grounding, hallucination, entailment). In a cited CV they form a pipeline.

1. A fact ledger holds every fact the candidate supplied, each with an id.
2. Grounding ties each written line to ids in the ledger.
3. A citation makes the tie visible on the line.
4. Verification checks that the line does not exceed its facts; a line that fails is dropped.
5. A human-likeness score, from a filler linter, is a separate check on style, not on truth.

Hallucination, defined narrowly

In language-model research, hallucination is fluent output that the input or the facts do not support; Huang and colleagues (arXiv 2311.05232, 2023) classify its kinds. On a CV the common kind is factual: an invented percentage, a promoted title, a merged date range. The word is used here in that narrow sense.

Layout, order and formats

These terms describe the shape of the document rather than its content.

• Single-column layout and standard section headings are the two properties that make a CV parser-safe.
• Reverse-chronological, functional and hybrid are the three orders a CV can take; Harvard, MIT and Prospects show reverse-chronological by default.
• Summary and objective are the two kinds of opening paragraph; hard and soft skills the two kinds of skills section.
• Europass is the EU's standard CV format, common in public-sector and EU-funded applications.

Regional differences

A US résumé and a UK CV omit a photo and a date of birth; Turkish and German CVs customarily carry a photo. The section headings and the order stay the same across regions; the personal details block changes.

Data protection

A CV is personal data by definition. Two laws govern how a CV builder and an employer may hold it.

• GDPR, Regulation (EU) 2016/679, applies to people in the EU and to services offered to them.
• KVKK, Law No. 6698, is Turkey's personal data protection law and applies to data processed in Turkey.
• Both give the data subject the right to access and erase their data, and both require a lawful basis for processing.

Credits in CitedCV

CitedCV charges by credit rather than by subscription. A credit is the unit of spend; a plan is how credits are bought. Both entries are specific to CitedCV.

Terms

Action verb

An action verb is the verb that opens a CV bullet and names what the candidate did: led, built, negotiated, reduced. It replaces "responsible for", which names a duty rather than an act.

AI slop

AI slop is fluent, generic text produced by a language model with nothing specific behind it, recognisable by phrases such as "leveraged synergies to drive impact". On a CV it is text a recruiter can neither check nor remember.

Applicant tracking system (ATS)

An applicant tracking system is the software a company receives, stores, parses and ranks job applications in; Workday, Greenhouse, Lever, Taleo and SuccessFactors are examples. It is a database and a workflow, not an AI detector.

Citation (on a CV)

A citation on a CV line is a visible reference, usually a superscript number, to the fact the line was written from: a form field, a sentence in an uploaded document or an interview answer. The sense is academic, not legal.

Credit (in CitedCV)

A credit is CitedCV's unit of spend. A full CV build costs 10 credits, a quick draft 6 and a single line edit 1; exports cost nothing, and a new account starts with 5 free credits.

Europass

Europass is the European Union's standard CV format and online CV editor, maintained by the European Commission. It is a fixed layout, widely accepted in EU public-sector and EU-funded applications.

Fact ledger

A fact ledger is the list of every fact a candidate has supplied to a CV builder, each with an id, a value and a source. It is the only material a cited CV may be written from.

Filler phrase

A filler phrase is a stock expression with no checkable content, such as "results-driven", "proven track record" or "dynamic team player". A filler linter counts them; a recruiter skips them.

Functional CV

A functional CV groups experience by skill rather than by employer and date. Prospects describes it as a skills-based CV for career changers; parsers read it poorly because the dates are detached from the roles.

GDPR

The GDPR, Regulation (EU) 2016/679, is the European Union's data protection law. It governs how any organisation serving people in the EU may collect, store and delete personal data, and a CV is personal data.

Grounding

Grounding ties every sentence a language model writes to specific input it was given, so the output can be checked against that input. A grounded CV line points to a fact; an ungrounded one to the model's general knowledge.

Hallucination (in language models)

Hallucination is language-model output that reads as fluent and confident but is not supported by the input or by fact. Huang and colleagues classify its kinds (arXiv 2311.05232, 2023); on a CV the common kind is an invented number or title.

Hard skills vs soft skills

Hard skills are teachable, testable abilities with names: SQL, IFRS, AutoCAD, German B2. Soft skills are working habits such as communication and leadership, and belong inside a bullet that shows them rather than a list that asserts them.

Human-likeness score

A human-likeness score is a number a filler linter gives a CV for how much it reads like a person wrote it, based on the density of filler phrases, repeated openings and stock adjectives. It measures style, not truth.

Hybrid CV

A hybrid CV, also called a combination CV, opens with a short skills summary and follows it with a reverse-chronological work history. It is the usual compromise for career changers.

Keyword matching

Keyword matching is how an applicant tracking system scores a parsed CV against a job description, by counting the terms from the posting that appear in the CV. It matches strings, not meaning, so "project management" and "managed projects" may not count as the same.

KVKK (Turkish personal data law)

KVKK, the Kişisel Verilerin Korunması Kanunu, is Turkey's Personal Data Protection Law No. 6698 of 2016. It sets the conditions for processing personal data, a CV included, in Turkey, and gives the data subject rights of access and erasure.

One-time credit pack

A plan adds a fixed number of credits to the balance on every payment, monthly or yearly, and renews until it is cancelled. CitedCV sells three sizes, Starter, Pro and Max, priced on the pricing page.

PAR formula

The PAR formula writes a bullet as problem, action, result: what was wrong, what you did, what changed. It produces the same content as the XYZ formula in story order, and suits interview answers as well as CVs.

Parser

A parser is the component of an applicant tracking system that reads a CV file and turns it into fields: name, contact details, employers, titles, dates, education and skills. It reads single-column text with familiar headings reliably, and tables, columns and images poorly.

Parser safety

Parser safety is the property of a CV that lets Workday, Greenhouse, Lever and similar systems extract its fields without loss. It comes from a single-column layout, standard section headings, real text rather than images, and plain dates.

Provenance

Provenance is the record of where something came from. On a CV it means that each line can be traced to the fact it was written from, and it is the property a cited CV makes visible.

Quantified bullet

A quantified bullet is a CV line that states a result with a number and a unit, such as "cut invoice processing from 12 to 4 days". The number must come from the candidate's own record, not from guesswork.

Reverse-chronological order

Reverse-chronological order lists roles from the most recent backwards. It is the default order Harvard, MIT and Prospects show and the one parsers handle best, because the dates decrease down the page.

Round-trip test

A round-trip test exports a CV to PDF or DOCX, parses the file the way an applicant tracking system would, and compares the result with the original fields. A CV passes when nothing is lost or reordered.

Single-column layout

A single-column layout places every section in one vertical flow with no sidebars, text boxes or side-by-side columns. It is the layout parsers read most reliably, because they read top to bottom.

Standard section headings

Standard section headings are the labels parsers are built to recognise: Experience, Education, Skills, Projects, Certifications. An inventive label such as "My journey" costs the section its place in the parsed record.

Summary vs objective

A summary is a two or three line statement of what you have done and can do. An objective states what you want; Harvard and MIT treat any opening statement as optional, and most guides now prefer a summary or none.

XYZ formula

The XYZ formula, popularised by a former head of people operations at Google, writes a bullet as "accomplished X, as measured by Y, by doing Z". It forces a result and a measure into every line and exposes the lines that have neither.

Questions people ask

Sources

1. Harvard Office of Career Services: create a resume, CV or cover lettercareerservices.fas.harvard.edu/resources/create-a-resume-cv-or-cover-letter/
2. MIT Career Advising and Professional Development: how to write a resumecapd.mit.edu/resources/how-to-write-a-resume/
3. Prospects: how to write a CVwww.prospects.ac.uk/careers-advice/cvs-and-cover-letters/how-to-write-a-cv
4. Greenhouse support centresupport.greenhouse.io/
5. Lever help centrehelp.lever.co/
6. Workdaywww.workday.com/
7. Europass: create your Europass CVeuropa.eu/europass/en/create-europass-cv
8. Huang et al., A survey on hallucination in large language models (arXiv 2311.05232, 2023)arxiv.org/abs/2311.05232
9. Regulation (EU) 2016/679 (GDPR) on EUR-Lexeur-lex.europa.eu/eli/reg/2016/679/oj
10. Law No. 6698 on the Protection of Personal Data (KVKK), mevzuat.gov.trwww.mevzuat.gov.tr/mevzuat?MevzuatNo=6698&MevzuatTur=1&MevzuatTertip=5