Career changers don't mostly fail because they're too old. They fail because they try to compress a career arc that a 22-year-old gets six years to walk into about eighteen months, and then quit in month fourteen when the maths stops working. The industry isn't really the problem. The plan was.
The misdiagnosis
Talk to anyone who tried to move into technology in their late thirties and bounced out around 2023 or 2024, and the story you'll get is some version of “the industry is ageist”. Sometimes that's genuinely true. More often it's the polite version of a thing nobody wants to say at a dinner party, which is that the plan was always too tight and the first knock finished it off.
The bigger thing, underneath, is that they were trying to be a senior engineer in two years. Nobody actually told them to. The alternative, being a junior for three or four years while their former peers run departments, felt unbearable. So the plan got compressed. And then, predictably, it broke.
What expectation compression looks like in practice
It usually shows up as one of these.
- Skipping the first role. Applying straight to mid-level cloud engineer positions out of a 16-week bootcamp, on the basis that “I've managed teams before”. The hiring manager looks at the CV and sees a graduate-equivalent technical profile with no engineering tenure. The interview is short.
- Targeting salary, not progression. Holding out for £55k roles because that's what the old career paid, and turning down the £32k entry role that would have made the next jump real. Eighteen months later, no offers at either level.
- Stacking certs as a shortcut past tenure. Three AWS certs, Security+, Network+, ITIL, all inside a year, all before a first technical role. The cert stack is then asked to do a job it was never designed to do, which is replace four years of engineering work on a CV.
- Refusing the title. Won't take a “junior” or “analyst” or “associate” title because it feels like a step backwards. Then misses the role that would have been a step forwards two years out.
Why the market isn't really being ageist about this
The median engineer on a UK technology team is older than the LinkedIn carousels imply. Plenty of mid-thirties career changers get hired into first roles every quarter. The ones who get hired share one specific posture, which is that they accept the first role's seniority on the role's terms, and don't spend the interview quietly negotiating against it.
Hiring managers aren't looking at a 38-year-old and thinking “too old”. They're looking at a 38-year-old applying for a junior role and trying to work out whether the person will accept being managed by a 29-year-old team lead called Megan who joined as an apprentice in 2019. That's the actual question. The candidates who answer it cleanly get the job. The ones who can't hide that they think they should be Megan don't.
Which is, oddly, easier to fix than ageism would be. You can't change your date of birth. You can absolutely stop wincing when somebody mentions the £30k figure.
The compression that actually works
Some compression is legitimate, and it's worth being honest about where.
Career changers can often skip the bottom of the first role faster than a graduate can. A 38-year-old who's run a logistics depot will become a senior helpdesk technician inside nine months instead of eighteen, because they already know how to triage workload, push back on stakeholders, and document incidents. That's real compression, and it's legitimate.
Career changers can also skip the “graduate optimism” phase. They don't need to spend a year working out that promises made in onboarding don't always land. They already know. That saves them from the slow disillusionment that costs graduates a year of performance.
Where compression doesn't work is the technical depth itself. Six years of engineering tenure produces a kind of fluency that 16 weeks of focused study, however intense, doesn't replicate. Not because the career changer is slower. Because the time itself, the exposure to multiple production incidents, the slow accumulation of failure modes, is the thing being learned. You can't book it.
The honest sequencing for a 35-year-old career changer
Plan for three roles, not one. The bootcamp brochure plans for one. That's most of the problem.
Role one is the entry job. £24k to £35k depending on postcode and route. It lasts 12 to 18 months. You take it on its own terms, not as a stepping stone you're visibly tolerating. The candidates who can't do that get filtered out at the interview, and usually decide later that the filter was ageism.
Role two is the consolidation job. £35k to £50k, more responsibility, a specialism beginning to form. This is where the old-life skills start to compound. The 23-year-old colleague gets promoted faster on raw technical merit. The career changer often out-influences them sideways into the more interesting work, because they've already done ten years of meetings.
Role three is the destination, or the next big specialism choice. Three to four years in, salary back to or above where the old career left off, doing work that genuinely interests you. This is where the change starts to pay back the cost of having started over.
Anyone telling you the destination is reachable in 18 months is either selling a £6,000 bootcamp or hasn't made the move themselves.
The conversation to have with yourself
Not whether you can make the change. You almost certainly can. The market is open to it, the routes are documented, and people like you do this every year.
The harder question is whether you can spend three years being more junior than feels comfortable, on a lower salary than feels comfortable, while your former peers are running departments, and not flinch out of the plan in month fourteen, when role one was meant to roll into role two and it hasn't.
If you can, the change works. If you can't, the change doesn't work, and no amount of certs or bootcamps will fix that.
When to walk away from the plan
Not every career change should happen, and not every career changer should move into technology. Some honest disqualifiers.
- You cannot accept the salary or title of a first role on its own terms, even for eighteen months. That is not a weakness. It is a signal that the timing is wrong, or that the household economics genuinely do not allow the gap. Forcing it produces the bitterness that gets blamed on the industry later.
- You are making the move because your current industry is shrinking, not because you want to do the work. Technology is not a refuge. The people who thrive are the ones who would be doing some version of this on evenings and weekends anyway. If you cannot name a specific part of the work that interests you, the move is unlikely to stick.
- You need the destination salary inside two years to service debt, a mortgage, or family commitments. The three-role arc is not a conservative estimate. It is the honest one, and anyone promising faster is either naive or selling something.
Where this connects on POST
The companion essay, The second beginner problem, covers the household economics and identity side of the same move. The pathways page lists the four entry routes career changers most commonly take. The Career Verdict will give you a written call on the specific move you're planning.