Job descriptions speak in requirements. Resumes speak in achievements. The space between them is where hiring breaks down.
You have probably heard the statistic. Men apply for a job when they meet only 60 percent of the qualifications. Women wait until they meet 100 percent. It gets quoted in leadership books and conference talks as proof that women need more confidence. It traces back to an offhand answer a Hewlett-Packard executive gave in an internal interview decades ago. Nobody has ever found a study behind it, because there isn’t one.
The interesting part is what happened when a writer named Tara Mohr got skeptical enough to actually check. She surveyed more than a thousand professionals and asked the ones who had skipped applying for a job why they had done it. Confidence barely showed up as an answer, for men or for women. The single most common reason, given by roughly four in ten of both, was some version of this: I didn’t think they would hire me since I didn’t meet the qualifications, and I didn’t want to waste my time. People were not lacking confidence. They were reading the job posting exactly as it was written, and taking it at its word.
That finding matters more than the myth it replaced. A job description and a resume look like two halves of the same conversation. They are not. A job description is written in requirement language. It says what you must already have. A resume is written in achievement language. It says what you have already done. Nobody sat down and decided to build two documents that answer different questions and hand them to each other as if they matched. It just happened, over and over, at every company, until it became how hiring works.
Job description: “Experience leading enterprise-wide digital transformation.”
Resume: “Led SAP S/4HANA implementation across three manufacturing facilities.”
Read as people, those are an excellent match. Read as strings of text, they share almost no words at all.
Job description: “5+ years of project management experience.”
Resume: “Led fourteen technology programs totaling 18 million dollars.”
That is arguably far stronger evidence of the capability than the years-of-experience language the posting used, but nothing in a keyword match is built to see that a dollar figure and a program count answer the question better than the literal phrase it was searching for.
Everyone adjusted
Once a requirement gets read as a real bar, and clearing that bar has an obvious payoff, people respond exactly how you would expect. A 2026 survey of recent job applicants found 93 percent admitted to embellishing or misrepresenting something on their materials. The honest part of that study is not the number, it is the reasons behind it. The top drivers were competitive pressure, a sense that nobody was really checking, and a flat belief that telling the truth put you at a disadvantage. Six in ten people who padded their resume said they would not have gotten the job if they had described themselves accurately. They were not wrong. They were responding rationally to a document that punishes honesty.
This creates a problem that spreads past any one resume. Economists have a name for it: adverse selection. Once faking a signal becomes common and cheap, honest signals lose their value along with the fake ones, because a reader can no longer tell them apart. Employers respond the only way they can, by trusting resumes less and adding more filters, which raises the payoff for the next round of embellishment. We have seen this same spiral before with AI-written applications. Employers tightened their filters to deal with the flood, which pushed job seekers to lean on AI even harder just to get through, which produced more of the flood. Embellishment works the same way. Each round of distrust makes the next round of exaggeration more necessary, not less.
It would be nice if tighter keyword matching solved this. It does not, and it was never going to. A meta-analysis of hiring methods going back decades found that resume-derived information like years of experience or education level barely predicts who actually performs well on the job, somewhere around a tenth to a fifth correlation with real performance, a range researchers describe as barely useful. Structured interviews do far better, somewhere around half. Later research revised the exact numbers down a bit, but the ranking held. A resume was never a reliable signal to begin with. Sharpening the filter on an unreliable signal just filters faster.
And there is a separate problem sitting underneath all of this that no amount of verification will fix. Courtesy. The ability to stay calm with a difficult client. Reading a room. Genuine empathy. A candidate who actually has these things can write “strong interpersonal skills” on a resume with complete honesty, and it will carry almost no information, because it is the exact same sentence an unqualified person would also write. This is not a trust problem. Background checks and reference calls solve the trust problem. This is a format problem. Some things simply cannot be shown on paper, no matter how honestly the paper is filled out. They can only be observed.
None of this gets fixed by design. A bad hire that turns into a legal problem is visible, it has a name attached and a cost on a spreadsheet. A strong candidate lost to an inflated requirement or an unreadable resume never shows up anywhere. Nobody gets called into a meeting over the hire that never happened. So the system keeps defaulting to whatever protects against the visible failure, even when everyone involved can see the process is broken.
There are two real fixes, and they answer two different halves of the problem. The first is writing job descriptions as an honest skills-needed exercise instead of a wishlist copied from an old template, so the bar being described is the bar that actually matters. That shrinks the incentive to embellish, because there is less daylight between what is asked and what is real.
The second fix is the one no document can replace. It has to happen before a resume ever lands on a hiring manager’s desk, not after. A hiring manager gets a resume and maybe two interviews to figure out who someone really is. That is not enough time for anyone to stop acting. This is where a real hiring partner earns its place, not by screening harder, but by already knowing the answer before the search even starts.
TM Floyd Talent’s recruiters build relationships with candidates over years, across multiple assignments, long before any specific role exists to fill. By the time someone gets presented for a job, they have already been watched doing real work, more than once, under real pressure, with people who had no reason to perform for an audience. That is the version of a person a two-interview process almost never sees. Closing that gap was never going to be a smarter document. It is time spent finding out who someone actually is, well before the stakes are high enough for anyone to fake it.
Sources
The debunked “60 percent men, 100 percent women” statistic and what the real survey found instead: Mohr, T., “Why Women Don’t Apply for Jobs Unless They’re 100% Qualified,” Harvard Business Review, 2014.
The research finding that no study actually supports the original statistic: Investigating Gender Differences in Qualification Thresholds, graduate thesis, Bowling Green State University.
Resume and interview embellishment rates and the drivers behind them: The 2026 Trust in Hiring Report, GCheck.
Predictive validity of selection methods including resumes, unstructured interviews, and structured interviews: Schmidt, F. and Hunter, J., “The Validity and Utility of Selection Methods in Personnel Psychology,” Psychological Bulletin, 1998.
