Maybe It's Not the System...

Every week there's a new post about how "hiring is broken." Recruiters ghost you. ATS eats your resume. AI rejects you in six seconds. And the conclusion is almost always the same: the system failed me.

Here's an uncomfortable alternative: the system didn't change nearly as much as the pool of people swimming in it did.

The math got brutal - and it's recent

In 2024, the average job opening pulled in roughly 250 applications, and employers needed about 180 applicants to make a single hire. By 2025-26, entry-level roles were routinely seeing 400+ applications per posting, and the applicant-to-interview ratio had collapsed to 2-3% — down from over 15% back in 2016. Job seekers today typically send 30-80+ applications before landing an offer; some data puts the number as high as 400-750 for a single offer in the toughest fields.

That's not a "the system is rigged" number. That's a supply-and-demand number.

The pool really is denser

This is the part people skip. College enrollment climbed roughly 380% between 1940 and 1990, and it kept climbing after that — more degree-holders, more master's degrees, more people qualified for the same shrinking set of "good" roles. Layer that on top of years of tech layoffs pushing senior, experienced people into the same applicant pools as newer grads, and you get more qualified competition per opening than existed even five years ago.

The result: recent college grads hit a 5.7% unemployment rate in late 2025 — a three-year high — and underemployment (grads working jobs that don't require a degree) topped 41%. On LinkedIn-tracked metros, some markets now see 100-400 applicants per job ad within a week. Openings themselves have also shrunk — U.S. job postings fell around 12-29% between 2022 and 2024 depending on the measure. Fewer seats, more people qualified to sit in them.

ATS and AI aren't new. The panic is.

Applicant tracking systems have existed since the 1990s. Fortune 500 adoption has sat above 97% since at least 2023, and mid-size company adoption climbed steadily for a decade — nothing sudden there. So why did the complaints explode in just the last three years?

Because something did change in that window: generative AI. Since ChatGPT went mainstream in late 2022, both sides of the hiring process armed themselves. Roughly 83% of companies now use AI to screen resumes before a human sees them. At the same time, job seekers started using AI to mass-apply — one recent estimate puts LinkedIn at over 11,000 application submissions per minute, up 45% year-over-year, and about 38% of job seekers now openly admit to mass-applying with AI tools. Recruiter workloads jumped 26% in a single quarter as a result.

Greenhouse's CEO has called this the "AI doom loop": candidates use AI to blast out applications, employers use AI to survive the flood, and the signal-to-noise ratio for everyone gets worse. That's a real, documented dysfunction — not a conspiracy, but not purely a skill-issue either. It's worth naming honestly: ghosting is up (roughly 53-61% of candidates report being ghosted, a multi-year high), and a majority of recruiters admit their own companies post "ghost jobs" that were never real openings. The system genuinely is straining under new load.

So where does that leave "just get better at it"?

This is the part worth sitting with rather than skipping past. The data supports two things at once, and they're not actually in tension:

• The market is objectively harder than it was three years ago, for reasons largely outside any individual's control — more graduates, fewer openings, AI-flooded pipelines on both sides.

• Within that harder market, adaptation still moves the needle. Candidates who tailor their resume to match the job title see roughly 3.5x higher interview rates. Recruiters say generic, obviously AI-written applications get rejected more often (62% report this), while thoughtful, tailored use of AI is treated very differently than mass-spam use.

Refusing to learn how ATS parsing or AI screening works isn't a moral failing, but it is a competitive disadvantage in a pool this dense — the same way refusing to learn spreadsheets was a disadvantage in the 2000s. The tools aren't going away, and the applicant pool isn't shrinking back down. The people who spend energy understanding the system tend to outperform the people who spend the same energy being angry at it.

Keep moving

None of this means the frustration is fake or that every rejection is your fault. Ghost jobs are real. Ghosting is up. The process genuinely broke down in places. But "the hiring system is broken" and "the talent pool tripled while openings shrank" can both be true — and only one of those two things is something you can actually do something about today.

So learn the tools. Tailor the application. Expect a numbers game, not a meritocratic straight line. And keep going — not because the system is fair, but because the ones who adapt to it are the ones still standing when it thins out again.

Sources: CareerPlug 2025 Recruiting Metrics Report · HiringThing 2025-26 Job Application Statistics · Jobscan 2025 ATS Usage Report · Federal Reserve Bank of New York, Labor Market for Recent College Graduates · St. Louis Fed, "Recent College Grads Bear Brunt of Labor Market Shifts" (Aug 2025) · Indeed Hiring Lab, "For New Grads Looking for Work" (2026) · Greenhouse 2024-25 State of Job Hunting Report · Criteria 2025-26 Ghosting Report · Fortune, "AI Doom Loop" interview with Greenhouse CEO Daniel Chait (Jul 2026) · Ars Technica / NYT, LinkedIn application-volume data (Jun 2025) · High5Test Resume Statistics 2024-25

<Language refined by GenAI>