How applicant tracking systems really work, and why Chennai applications disappear

A candidate applies to TCS, Cognizant, Zoho and four mid-market engineering firms in Chennai. Nothing comes back. The assumption is always the same: an algorithm rejected the resume. That is almost never what happened.

Applicant tracking systems do not score resumes and do not reject them. They parse the uploaded file into database fields — name, employer, dates, title, skills — and a recruiter then runs a search against those fields. An application that never appears in the search was not rejected. It was parsed badly, landed with an empty experience field, and no human ever saw it.

What the parser is actually doing

The system reads your file and tries to assign each block of text to a field. It looks for headings it recognises, dates in a standard form, and employer names it can separate from job titles. Where it cannot decide, it guesses, and a wrong guess is silent.

This is why formatting decides more than writing at this stage. A well-written resume in a two-column template can arrive in the database with the entire work history missing, because the parser read the columns as one interleaved block of nonsense and discarded it.

The five formatting faults that cost Chennai candidates interviews

• A two-column layout — the commonest single cause, and most free templates use one
• Creative headings: “My Journey” does not parse; “Experience” does
• Contact details in the header or footer, which many parsers never read
• Tables used for layout, which interleave unrelated text
• Dates written as “2021 – present” with no month, which breaks tenure calculation

None of these is a writing problem. All five are fixable in an afternoon, and together they account for most of the applications that vanish without acknowledgement.

Keywords: what matters and what is superstition

Recruiters search the terms in their own job description. If the advertisement says “Selenium” and your resume says “automation testing tools”, you do not appear. That is the whole mechanism, and it is simpler than most advice suggests.

Two things follow. Write both forms of important terms, because a parser matches strings and not meaning: “Kubernetes (K8s)” matches either query, while one alone matches one. And take the terms from the actual advertisement rather than a generic keyword list.

White text stuffed with keywords is detected, and every Chennai recruiter who finds it treats it as deception. It has ended applications that would otherwise have progressed.

What nobody can promise

There is no single applicant tracking system. Naukri, Workday, Taleo, Greenhouse, Zoho Recruit and a dozen others behave differently, and none publishes a scoring model. Anyone quoting you a “98% ATS score” is quoting a number they invented.

What is genuinely controllable is that the document parses cleanly and contains the terms a recruiter will search. That is the honest limit of the service, and it is enough to change the outcome for most applications.

How Content Factory writes it

Content Factory rebuilds the document single-column with standard headings, month-and-year dates throughout and contact details in the body where every parser reads them. We then align the wording to two or three advertisements you send us, placing both forms of each key term where they read naturally to a person.

We also tell you which of your claims will survive a recruiter search and which will not. An ATS review from us is a written report, not a score, because a score would be invented.

Chennai clients usually take the optimisation with a rewrite, since a cleanly parsed weak resume simply fails one stage later in front of a person. Prices are fixed in writing before work begins.