Blog · Job applications
We build a resume optimization tool, so this is against our own commercial interest to say: most of what you have read about applicant tracking systems is marketing, and the single most-repeated statistic about them was invented by a company selling the cure.
You have seen it. "75% of resumes are never seen by a human." It appears in career advice columns, on LinkedIn, and in the sales copy of nearly every resume tool on the market.
It traces to a 2012 sales pitch from a company called Preptel, which sold resume-optimization software. No study was published with it — no methodology, no sample size, no data. The Interview Guys traced its origin and found the company shut down the following year. The number outlived it by well over a decade.
A sales pitch
Not a study. The number that convinced a generation of job seekers their resumes were being binned by robots was produced by a company whose product was the solution to that fear.
When the source of a fear is also the seller of its remedy, that's worth noticing. It's worth noticing about us too — we sell the remedy.
An applicant tracking system is, for the most part, a database with a search box. Workday, Greenhouse, Lever, Taleo, iCIMS — they exist so a recruiter handling 400 applications can store them, search them, tag them, move them through stages, and not lose anyone in an inbox.
What they do not generally do is score your resume and bin it below a threshold. There is no industry-standard score. No major vendor publishes one for candidates to optimise against, because there isn't one to publish. Survey work reported by HR.com found the overwhelming majority of recruiters review applications manually, using filters to prioritise rather than to eliminate.
The realistic worst case is not a robot rejecting you. It is a recruiter searching for "Kubernetes", getting sixty results, and reading the first twenty — and your resume said "container orchestration".
Three things, none of them exotic:
Tailoring your resume to a job description is worth doing — but for an unglamorous reason. It isn't that you're defeating a filter. It's that a resume written in the posting's own vocabulary is easier for a searching recruiter to find, and easier for a reading one to say yes to in the fifteen seconds they give it. Those are both human problems.
That's a smaller claim than the industry usually makes. It also happens to be the one that survives contact with how hiring actually works.
Our resume optimizer does exactly this and no more: it reads your resume against a specific posting and suggests wording changes, each shown as an accept-or-reject decision so you can check every one against what you actually did. It won't give you a score, because the score would be made up. Free, no account needed.
Sources: The Interview Guys on the origin of the 75% claim; HR.com and HiringThing on how recruiters actually use these systems.