Entry level hiring feels harder than the employment statistics suggest, and both impressions are correct. This page explains why they diverge, where the first rung actually moved to, and what that means for anyone trying to find one now.
The short answer
Not in total, but the first rung has moved. Overall employment is projected to rise from 170.3 million to 176.2 million between 2025 and 2035, an increase of about 5.9 million positions. What has changed is where entry points sit and how many exist in the specific fields graduates target.
The gap between those two facts is the whole story. There are more jobs and fewer of the particular jobs people were told to aim for, which is why the national numbers and the experience of anyone job-hunting right now feel like they describe different worlds.
Who this is for
- Graduates and school leavers trying to work out whether the market is genuinely worse or whether it feels that way.
- Career changers trying to work out which fields still have a viable way in for someone starting over.
- Parents and advisers giving guidance that was formed when the entry landscape looked different.
- Anyone told to learn a skill that was recommended when it led into a tier of work that has since become considerably thinner.
Key benefits of understanding the shift
- You stop applying into the wrong tier. Much of the frustration comes from competing for roles that have genuinely contracted.
- You can see where the entry points moved to, which is mostly healthcare, trades and licensed fields.
- You can ask better questions in interviews, particularly about how an employer actually trains people.
- You avoid attributing a structural change to personal failure, which is common and unhelpful.
- You can plan around a real constraint rather than around a general sense that things are bad, which makes the effort you do spend count for something.
How it works: where the entry points actually went
The clerical tier that absorbed enormous numbers of first jobs is shrinking, and the numbers are unambiguous. Tellers are forecast to decline 13 percent to 2035, bookkeeping and accounting clerks 6 percent, and information clerks 2 percent. These were classic first jobs requiring no degree, and there are steadily fewer of them.
Meanwhile hands-on fields are adding positions quickly. Medical assistants grow 13 percent, adding roughly 107,600 roles. HVAC technicians grow 11 percent. Both are genuine entry points with training routes that do not require a degree.
So the total holds up while the composition changes. The entry tier did not disappear. It relocated out of offices and into healthcare settings, building sites and vehicles, which is a real change in what a first job looks like even though the count holds steady.
Within professional fields the change is different and less visible. The jobs still exist but the first rung is thinner, because the well-specified tasks that juniors used to do are the ones automation handles best. That produces a market where mid-level roles are advertised freely and genuinely junior ones are scarce, which is precisely the pattern graduates describe.
The mechanism behind that is worth understanding, because it was not a decision anybody announced. A manager with a well-defined task can either brief a junior, wait, review the work and give feedback, or produce it directly in ten minutes. The second is faster on that task, and nothing at the moment of choosing prices the training that did not happen. Multiply it across every manager and every task over two years and a training route disappears without a meeting about it.
That also suggests it is not permanent. The cost of not training arrives about five years later as a shortage of mid-level people, at which point buying them on the open market is considerably more expensive than having grown them. Several industries have been through this cycle with apprenticeships before, and the pattern repeats. Cold comfort for someone applying today, but it argues against treating the current state as the new permanent shape of the market.
What this means practically, by starting point
The shift lands differently depending on where you are, and generic advice is worth less here than usual.
If you are finishing a degree aimed at a professional field, the junior tier you were preparing for is thinner than when you started. The response that works is arriving with something besides the qualification: a domain you know properly, a portfolio of work you actually did, or experience in an adjacent role that touches the same industry. Applications that look identical to two hundred others fare badly in a thin market.
If you have no degree and are choosing a direction, the entry landscape is genuinely better than it was for the trades and healthcare. Apprenticeship and certificate routes lead into fields growing at two to four times the whole-economy rate, they pay well, and the cost of training is a fraction of a degree.
If you are changing careers mid-life, licensed fields are the strongest destination. The requirement to hold a qualification is what keeps the role human, and prior work experience frequently counts toward entry in ways it does not for school leavers.
If you are already employed somewhere entry level, the question is whether your specific tasks are the specified and repetitive kind. Moving toward the exceptions, the diagnosis and the customer-facing work inside your current employer is usually faster than moving out.
Proof, examples, and objections
The clearest evidence is the divergence within single industries rather than between them.
In technology, software developers and QA analysts are forecast to grow 10 percent while computer programmers decline 7 percent. Same sector, same starting skills, opposite direction. The declining category is the one defined by implementing to a specification, which is what junior work mostly was.
A common objection is that this is simply a weak graduate market that will recover with the economy. Partly true and incomplete. Cyclical downturns compress hiring across all levels; this compresses one level specifically while mid-level hiring continues. That pattern points at a structural cause rather than a cyclical one.
A second objection is that people have always found entry difficult and this is ordinary. Also partly true. What is different is that the difficulty concentrates in fields with formal graduate routes while entry into trades and healthcare has become comparatively easier, which reverses the advice most people received.
A third objection deserves a direct answer: does this mean qualifications are worth less? No, and the data points the other way. Fields with formal entry requirements have held up better than those without, because a licence or certification is a structural requirement that automation does not soften. A qualification opens the first conversations and gives learning a shape that self-teaching rarely matches. What has changed is that pairing it with something demonstrable, whether a domain, a portfolio or logged practical hours, matters more than it did.
For context on how exposure is measured, the Bureau publishes AI exposure categories for 831 occupations and states plainly that exposure “does not imply job loss, productivity gains, automation probability, or wage effects.”
The cost of getting this wrong, in both directions
Two mistakes are common, and they are opposites.
The first is concluding that nothing has changed and applying into a contracting tier repeatedly, at high emotional cost, while attributing the outcome to personal inadequacy. This is the more damaging error because it produces a lot of effort with no feedback loop. Someone sending two hundred applications into a category that has genuinely thinned learns nothing from the silence except that they are failing, which is the wrong lesson.
The second is concluding that everything has collapsed and abandoning a viable plan too early. Professional fields still hire; the route in is narrower and requires more than the qualification alone. Someone who gives up a year before finishing a degree in a licensed profession has made an expensive decision on incomplete information.
The way to avoid both is to check the specific occupational category rather than reasoning from the general mood. The data is published, free and updated on a schedule, and ten minutes with it is worth more than a month of speculation. A field growing at 13 percent and one declining at 13 percent both sit inside the same overall employment picture, and the difference between them decides whether persistence is the right strategy or a costly one.
Your next step
Before applying anywhere else, check whether the roles you are targeting sit in a growing or contracting category, using the occupational data rather than impressions. Ten minutes there will tell you whether you are competing in a thin market or a thick one, and that changes the strategy rather than just the effort.
Whichever route you take, the skill worth building alongside it is knowing how to work with these systems rather than around them: what they handle reliably, where their confident output is wrong, and how to check it. That capability transfers between fields and holds its value as the tooling changes. If you want a structured way in, explore Coursiv AI lessons and check current plan details on the official site.
Then ask every employer you speak to how they train people into the next level, and listen for specifics. An organisation that can name someone who took that route recently has thought about the problem. One that talks vaguely about learning on the job has probably not, and in a market where the junior tier is thinner, that distinction decides how much you will actually learn in your first two years.