The Seven-Second Resume Study, and What It Does Not Say

The Seven-Second Line Has No Checkable Source

The figure is old, widely repeated, and consistently attributed to eye-tracking research commissioned by a job-board vendor rather than an independent lab. No methodology stating its sample, its year or its scope is cited alongside it, which is the whole problem with it.

So this article does not repeat the number as a measured fact. A figure nobody cites a method for is not evidence, however many pages carry it, and the honest response is to say so rather than to quietly adopt whichever version sounds most precise.

Call this typical, not exceptional. Statistics about hiring circulate the same way rumors do: a figure appears once, sounds specific enough to be plausible, and gets copied from page to page without anyone tracing it back to a method. The seven-second line is simply an unusually durable example. Treat any single, precise-sounding number in this domain the same way until you've seen the source that backs it: useful as a prompt to ask a question, not as a fact to plan a resume around.

Why a First Pass Is Real, Even Though the Number Isn't

A fast first pass genuinely happens, which is why the seven-second line felt true, just not with the precision it implies. In a Harvard Business School and Accenture survey of 2,275 senior leaders in Germany, the UK and the US, fielded in early 2020, 63% reported using a recruitment management system, 69% at employers over 1,000 workers and 75% in the US, and over 90% of those users said they filter or rank candidates with it before a person looks.¹

That is employer self-report about their own configuration, in three countries at one moment, so read it as how the pile gets handled rather than a current global rate. Employers are also being told something more uncomfortable about that pass than applicants usually hear: how do you know if your resume screen rejects the wrong people argues that a screen cannot be validated on the people it kept, because the ones it got wrong were rejected and no record of them will ever exist. Worth being precise, then, about what that software actually does, because it is not a stopwatch on a human reader. Workday's administrator documentation tells employers that parsing results can vary with a resume's format and word order, and that for best results resumes should not contain images or image-based styles.² Greenhouse publishes a named list of what breaks a parse in its own system: files over 2.5MB, letters with spaces between them, graphics, a resume uploaded as an image, a columned layout, and abbreviated job titles among them.³ Greenhouse also states the consequence, which is smaller than the folklore: when a parse fails, the resume stays attached to the candidate and someone types the details in by hand. So this stage is software deciding whether the page becomes readable data, on one vendor's parser at a time, and it is a different problem from a reader's attention.

Even the US government's federal resume guidance, which governs hiring through USAJOBS and nothing else, carries something that sounds similar and is not the same thing: a self-review prompt asking whether an agency could find your main credentials within roughly ten to fifteen seconds. It is written as advice for the writer to check their own layout against, not as a measured finding about how long any hiring official actually spends.⁴ It is easy to mistake a writing tip for a research finding when both arrive as a small number attached to the word 'seconds'.

Design for the Read That Follows the First Glance

A resume that survives the first pass gets read properly later, and that is where the hiring process spends its time. In SHRM's 2021 benchmarking data from its member organizations, the median time to fill a nonexecutive role was 44 calendar days from requisition to accepted offer, and the median organization gave five of those days to screening and seven to interviewing.⁵ The spread was wide: a quarter came in under 28 days, a quarter took over 73.

Whatever happens in the first few seconds a file is open, it is a small fraction of a process that runs for weeks and depends, later, on the same document holding up under a much closer read. That is the case for treating the first pass and the second pass as different design problems rather than one number to optimize against. Make the top third of the page and a clear section order carry the outline, because a first pass, however long it actually runs, does favor a page a reader can orient inside quickly. Then write the rest of the page for the reader who gets there: the interviewer checking a specific claim, the hiring manager comparing two finalists, the reference call confirming what the page said. Hollowing out the content to win an unverified first glance costs you the read that runs longest, and it is worth checking whether the resume template you pick actually matters before spending more hours on layout than on substance.

The first pass rewards clarity: a heading structure a reader can scan, dates that line up, nothing hidden behind an image. The second pass rewards specificity: a real project, a real number, a decision you made and what happened next. A resume built only for the first pass reads as empty once someone actually sits with it, and a resume built only for the second can lose a reader before they ever reach the good part. Neither trade is necessary once you know both reads are coming.

See a sample report >