The Resume Host / The screen

Applicant tracking systems, minus the mythology

They are databases. Most of what is claimed about them is not true.

Recruitment software attracts a large amount of confident misinformation, most of it circulated by services selling optimisation. The systems in general use are principally databases with search and workflow: they store applications, allow recruiters to filter and search, and track progress. In the common configuration they do not score and reject applicants automatically.

Filtering does happen, and it is usually performed by a person applying criteria, or by explicit knockout questions on the form, which are visible to the applicant. The image of a document being silently scored and discarded by an algorithm describes some high-volume configurations and is not the general case.

The practical implications are duller than the mythology. Use words that a human searching for a capability would actually type, which usually means the ordinary term rather than an internal one. Answer the form fields properly, since those are structured data and are what gets filtered on. Submit a format the system can parse, which in practice means avoiding complex layouts, text inside images and unusual file types.

What does not help: keyword stuffing, invisible white text, or padding with terms from the advertisement. Where these are detected they read as an attempt to game the process, and they are detected more often than the people selling the advice suggest.

It is also worth knowing that many systems parse a document into fields and then present the parsed version to the recruiter alongside the original. Where the parsing goes wrong, which multi-column layouts and text in headers reliably cause, the recruiter may see a garbled summary first. This is the practical reason to keep the layout conventional, and it is a much better reason than any claim about keywords.