This website uses cookies

Read our Privacy policy and Terms of use for more information.

MARKET COMMENTARY

"Entry level" that quietly wants five years

By Ric @ Jobric · November 2026

The posting that says "entry" and means "experienced"

Here's a listing you've probably already met this fall.

Title: Associate Analyst, entry level. Then, four lines down: "3 to 5 years of experience preferred." Then a stack of four tools you'd need a prior job to have touched. Then a degree, and a bonus line about internship experience "a plus."

Read it twice and the contradiction doesn't resolve. It's labeled for someone with no track record and written for someone who already has one. You are not misreading it. You are not underqualified for a role you were supposedly built to fill. The rung got raised while the sign on it stayed the same.

That gap is the whole story, so let's read it honestly. No clean villain, no tidy headline. Just what the data actually says and what it means if you're the one updating a resume tonight with one internship and a graduation date on it.

Why the rung got raised

The mechanics are less dramatic than the panic headlines and more frustrating because of it.

  • Fewer junior postings. When the bottom of the ladder thins out, the roles that remain absorb more expectation.

  • Leaner teams. A team that lost a headcount doesn't post for "someone to train." It posts for "someone who can start Tuesday."

  • More applicants per opening. When hundreds of people apply to one posting, the employer can keep adding requirements and still fill the seat. So they do.

None of this is a plan. It's just what happens when demand for juniors softens and supply of applicants climbs. The "5 years for entry level" line isn't a company being cruel. It's a company writing a wish list because the market lets them, and discovering nobody's there to tell them it's absurd.

The data, read without the hype

This is where most takes pick a narrative and run. I'd rather hand you the contradiction and let you hold it, because the honest read is a contradiction.

Entry-level postings have genuinely thinned. Analysis from Revelio Labs puts US entry-level postings down roughly 35% since early 2023, with junior software and data roles down as much as about 67%. The World Economic Forum covered it this March under a plain enough framing: the first rung of the career ladder is narrowing fast. That's not a vibe. That's the bottom rung getting narrower, hardest in exactly the fields new grads were told were safe.

Early-career headcount is softer where AI adoption is heaviest. A Harvard working paper by Seyed Hosseini and Guy Lichtinger tracked 62 million workers across 285,000 firms and found that at companies hiring "GenAI integrator" roles, junior employment fell about 7.7% relative to non-adopting firms over the six quarters from early 2023. Senior employment at those same firms kept growing. Notably, the decline came from slower hiring rather than people being pushed out.

And here's the part worth sitting with: that still isn't a closed case. The design is good and the finding is real, but "junior hiring slowed at firms that adopted AI" is not the same claim as "AI replaced juniors." Those firms were also restructuring and reorganizing for other reasons.

There's now a serious paper making that case directly. The Broken Ladder: AI, Remote Work, and Early-Career Hiring, by Peter John Lambert and Yannick Schindler, circulated in May 2026 through the LSE's Centre for Economic Performance, works from 243 million new-hire records and 407 million job postings across the US, UK, Canada, and Australia. Their finding isn't just that the AI story is unproven. It's that remote work is the stronger predictor of the decline in early-career hiring. Two credible papers, two different culprits. Anyone selling you certainty here is selling you something.

Meanwhile, grad-hiring intent actually ticked up. NACE's Job Outlook 2026 Spring Update put projected hiring of new graduates at +5.6%, revised up from the flat 1.6% its fall survey projected. That's 185 employers surveyed in February and March, so hold it lightly. But it points the opposite direction from the doom narrative.

And the lived experience is harder than any single number. The New York Fed's Q2 2026 read put recent-grad unemployment at about 5.6% and underemployment at 42%. That underemployment figure is the one that stings: it means a large share of grads who are working are working below the level they trained for. (That's the NY Fed's recent-graduate series for bachelor's holders aged 22 to 27, not the national rate you'll see Friday mornings. Different number, different thing. I won't blur them.)

So what's true? Postings thinned. The junior tech door narrowed hardest. AI is a suspect with real evidence against it and a credible alternative suspect standing right there. And hiring intent is up even as the day-to-day feels brutal. All of it at once.

What it means if you're graduating into this

The temptation is to read all that and conclude your degree failed you. It didn't.

The degree isn't the problem. The matching is.

Think about what actually happens when you apply. A posting lists a 5-year number you can't have yet, a filter screens for that number, and you're gone before a human reads the one internship that's genuinely relevant. You didn't lose on fit. You lost on a keyword that was priced for someone a half-decade older than you.

That's a matching failure dressed up as a qualifications failure. And it's fixable, because the thing being measured (a years-of-experience string) was never a good proxy for whether you can do the work.

Reading roles for fit instead of filtering you out

This is the part where Jobric does something useful, so I'll be plain about it.

Jobric scores roles against your actual profile before you ever see them. Not against who can clear an inflated keyword bar. For someone with a short resume, that changes the game, because a short resume isn't a weak one. It just needs to be read for what's in it.

Two things matter most when you're new:

  • The fit score plus the AI Fit Analysis. You get a percentage and a written breakdown: what aligns, what's a stretch, and why. When a posting says "entry" but means "experienced," the analysis tells you whether it's genuinely reachable or whether that 5-year line is load-bearing. That's the difference between a smart application and 200 hopeful ones.

  • Curated matches, ranked by fit. Fewer roles, better aimed. You stop scrolling past a thousand listings written for other people and start seeing the handful where your actual profile competes.

You still have to do the work and write the application. But you get to spend that effort on roles where a short resume has a real shot, instead of burning it on postings that quietly filtered you out at line four.

What the data doesn't tell you

A few honest limits, because the trust is the point.

  • We don't know how many of those "entry-level" postings are even real. Some are evergreen listings, some are pipeline-building, some are ghosts. "Down 35%" counts postings, not genuine openings.

  • The AI question is open. I showed you the evidence and the counter-evidence on purpose. If someone tells you AI definitely did or definitely didn't eat entry-level work, they know less than they're claiming.

  • NACE's +5.6% is a 185-employer projection, not a promise. Intent is not a signed offer.

  • Averages hide you. "Grad hiring up" and "42% underemployed" can both be true, and your field, your region, and your timing decide which one you're living in.

The forward look: NACE updates again heading into the spring cycle, and the next NY Fed recent-grad read posts in November. If the hiring-intent number holds while postings stay thin, that tells us the roles are getting more selective, not more plentiful, which is exactly the squeeze you're feeling. We'll track both.

Until then: a short resume isn't a weak one. It just needs to be read for fit, not filtered for a number you can't have yet. Jobric's free Seeker plan does that reading for you. It's a real plan, not a trial.

That's the update. Now go do something that isn't job searching.

Ric @ Jobric

Sources