The bottom rung of the career ladder is being sawn off, and the numbers are not subtle. Entry-level job postings are down about 35 percent since 2023, and in some technology and data roles closer to 67 percent. Recent graduates are underemployed at roughly 43 percent as of December 2025, the highest since the pandemic, and their unemployment rate, 5.6 percent, now sits above the national rate of 4.3 percent. The class of 2026 faces projected hiring growth of 1.6 percent, which is not growth once you count the larger graduating class.
The usual telling blames one thing, AI, and moves on. The more useful question is what a person does about it, because a ladder that is losing its first rung is not going to grow the rung back. We went looking for the entry rung in our own data to see how thin it has become.
A caveat we will make before anyone else does: postings undercount entry work, because an entry role often just prints the job title with no rank attached. Our seniority read is the explicit signal, the words junior, senior, and lead, so the true entry share is higher than 0.13 percent. But the direction is unmistakable and it agrees with the national figures. The labeled bottom rung is thin to the point of absence, and the market that used to hire people to learn on the job is hiring people who already learned somewhere else.
What breaks when the rung breaks
Entry-level work was never really about the work. It was where a person learned what good looks like: how to notice when something does not add up, how to push back without blowing up the meeting, how to see what their output does three desks downstream. Automate the straightforward tasks that juniors used to cut their teeth on, and you do not just remove a job. You remove the training that turned that job into the next one. That is the part the displacement statistics miss. The BLS (US Bureau of Labor Statistics) can count the vanished postings; it cannot count the vanished apprenticeship.
This is a Career Half-Life problem, not a young-person problem. A ladder with no first rung does not only trap graduates. It removes the normal way anyone crossed into a new field, which was to enter at the bottom and climb. If the bottom is gone, the only way in is sideways.
The way in went sideways
Here is the move the panic articles miss. When you cannot enter a field at the bottom, you enter an adjacent field where your skills already reach, and you pivot on coverage rather than seniority. Our whole instrument measures exactly this: how much of one occupation’s demanded skills another occupation’s people already hold. Where that coverage is high, the crossing does not need a junior rung, because you are not starting over. You are arriving with most of the job already done.
| Side door | Skill coverage | What it means |
|---|---|---|
| Accountant → Financial controller | 81% | Almost the whole skill set already carries |
| Accountant → Bookkeeper | 77% | An open lane while junior-accountant roles thin out |
| Software engineer → Backend developer | 63% | Same tools, a narrower title with its own openings |
| Data analyst → Business-intelligence developer | 51% | The reporting stack, one seat over |
| Data analyst → Product analyst | 47% | Analyst skills, pointed at a product team |
None of these is a fantasy leap. They are the crossings where our corpus says the skills mostly transfer already, so the move is a matter of naming and evidence, not of starting at the bottom of a ladder that no longer has a bottom. The route pages lay out the exact gap for the most-searched crossings, and the graph on the front page runs it for any starting point you give it.
What to do if the front door is closed
Stop applying to the vanishing rung. Four hundred applications to entry-level roles that increasingly do not exist is a strategy optimized for the market of 2019. Instead, inventory the fields adjacent to the one you want, the ones where your current skills already cover half or more of what the postings ask, and enter there. Then pivot from inside, where you can see the target job and hold the evidence that you can do it. The BLS and the World Economic Forum agree there will be work: the Forum projects 92 million roles displaced and 170 million created by 2030, a net gain. The catch is that the created jobs will not be entered the old way. They will be entered sideways, on skills, which is the door that is still open.
PivotHop pipeline, July 2026 run: 77,443 live job postings across 153 occupations. Seniority is read from explicit title signals (junior, senior, lead), which undercounts unlabeled entry roles; 65 of 50,660 seniority-tagged postings carry a junior signal, and two of 154 occupations reach a junior band. Skill coverage is the destination’s demanded-skill coverage by the origin’s profile, top 20 skills per occupation from posting text. External figures: entry-level decline and recent-graduate underemployment and unemployment from 2025 to 2026 reporting (Fast Company, CNBC, Washington Monthly, NACE outcomes); displacement and creation projections from the World Economic Forum Future of Jobs. Run your own starting point on the front-page instrument.