Software that manages the recruitment process from job posting to hiring.
An applicant tracking system (ATS) is recruitment software that holds every application to a vacancy in one database and moves each candidate through defined hiring stages, from advert to accepted offer. It is the system of record for a hire: who applied, who reviewed them, what was decided at each stage and when.
An ATS starts at the approved vacancy: the requisition, its criteria and the advert sit together, and are published to the careers page and job boards from one place. Applications then land as structured records rather than attachments in an inbox, and every candidate sits at a named stage — screening, interview, offer, hired, rejected — which changes only when someone takes an explicit action.
Around those stages sit the working parts of hiring: interview scheduling, scorecards and interviewer feedback attached to the candidate record, the history of what was sent to them, offer approval, and handover of an accepted candidate's details to onboarding without anyone retyping them. Applicants who were not hired stay in a searchable pool for later roles. Everything the recruitment reports draw on — application source, conversion between stages, time spent in each — is a by-product of that stage history rather than a separate exercise at quarter end.
The common belief is that the software reads a CV, hunts for keywords and discards anything that misses them. That is not what the parsing step does. Parsing converts an uploaded document into structured fields — name, contact details, employers, job titles, dates, education, skills — so that applications become searchable and comparable with each other. It is an extraction step, and it makes no decision.
Where an application really is rejected without a human, it is almost always a knock-out question the recruiter configured on the form — right to work, a required licence or qualification, availability for the advertised pattern — answered explicitly by the candidate. Match or ranking scores order a list; they do not remove anyone from it.
The genuine failure mode is quieter than the myth. Multi-column layouts, tables, text boxes, details buried in headers and footers, inconsistent date formats and text embedded in an image all parse badly, producing a record with the wrong employer against the wrong dates, or with the skills field simply empty. Nobody rejects that candidate. They just never appear in the searches a recruiter runs, which is why neither side ever notices.
Recruitment produces decisions about people who never become employees and so have no ongoing relationship in which to raise a concern. If one of them later challenges how they were treated, the only evidence of how the decision was made is whatever was recorded at the time. An ATS is where that lives: the criteria advertised, who was screened out at which stage and by whom, the reason recorded against them, and the interview notes and scores behind it.
Consistency counts for as much as the record. The same questions and the same scoring applied to every candidate for a role is defensible in a way that an exchange of emails between two managers is not. The same store also carries the data-protection side: candidate records are personal data gathered for a stated purpose, held on a defined basis and deleted or anonymised when that basis ends, with any agreement to stay in a talent pool recorded alongside. Retention and access requests are answerable when applications sit in one system and close to unanswerable when CVs are scattered across individual mailboxes. Equal-opportunities monitoring data, where it is collected, is kept apart from the assessed application so that it plays no part in the decision.
An ATS records a process; it does not supply one. The stage data is only as truthful as the habit of moving people through it — if a manager interviews on Monday and updates the record a fortnight later, the time-in-stage report is fiction and the real bottleneck disappears from the chart. Decisions taken elsewhere and back-filled afterwards do the same damage to the audit trail.
Two other traps are structural rather than behavioural. Every extra stage feels like rigour, lengthens the funnel and gives candidates another point at which to walk away, so each one should earn its place. And time-to-hire means nothing until the organisation agrees where the clock starts, because moving that start point changes the number without changing anything a candidate experiences.
Rarely, and not on the basis of CV wording. Automatic rejection happens only where a recruiter has switched on knock-out questions in the application form — right to work, a required licence, availability for the advertised hours — and the candidate answers in a way that disqualifies them. Parsing extracts information from a CV; ranking scores sort a shortlist. Neither removes an applicant from the pipeline on its own.
Single column, real selectable text rather than an image or a design-heavy export, conventional section headings, employers and dates written in a consistent format, and contact details in the body of the document rather than in a header or footer. Tables, text boxes and multi-column layouts are the usual cause of a mangled record — and a mangled record is what stops a candidate surfacing in a recruiter's search months later.
An ATS is application-centric: it tracks people who have applied to a specific vacancy through the stages of that vacancy. A recruitment CRM is relationship-centric: it holds people who have not applied — sourced prospects, past finalists, event contacts — and manages contact with them until a suitable role exists. Many products now do both, but the distinction matters when choosing one, because an ATS with no CRM side leaves proactive sourcing to spreadsheets.
Explore our full platform with a 14-day free trial on Standard. Manage employees, post jobs, hire faster, and manage your tasks effortlessly with our all-in-one platform.