How to Change Careers: Move One Variable, Not Your Whole Life

Here's how to change careers without losing years and money: change one variable at a time. Keep your function and switch industries, or keep your industry and switch functions. Never both at once. The boring, adjacent move is the one that keeps your earning power intact. The cinematic leap is the one that quietly costs you years.

Here's the whole plan on one screen:

Step	What you do	Why it matters
1. Split your job in two	Write down your function (what you do) and your industry (where you do it)	You can only hold one still if you can see both
2. Pick one variable to change	Function or industry, not both	Your earning base travels with whatever you keep
3. Measure the skill overlap	Compare your skills to the target role's, honestly	Distance between skill sets predicts the pay hit
4. Find the shortest bridge	Internal transfer, adjacent employer, or a hybrid role	Proof inside a building beats a certificate outside it
5. Budget the dip	Know your runway before you resign	Most switches cost something in year one
6. Make the second move later	Step again once the first move has stuck	Two short steps beat one long jump

Here's what most people get wrong. They believe a career change means becoming a new person, so they copy the stories on their feed: the banker who became a chef, the lawyer who became a designer. Those stories are real, but you never see the denominator, the many people who tried the same leap, spent years below their old salary and drifted back. The real driver of a successful switch isn't courage. It's how much of your existing value comes with you.

Why do adjacent career changes work better than a full reinvention?

Because much of your career value is tied to your occupation, not your industry, and less of it survives each line you cross.

The labor economics here is unusually clear. In Kambourov and Manovskii's study of U.S. workers, five years of occupational tenure was associated with 12% to 20% higher wages, and once occupation experience was accounted for, industry and employer tenure added relatively little. The value you've spent years building travels with the work you do, not the sector you do it in. That's why an industry switch is usually cheap and an occupation switch usually isn't.

The damage isn't really about job titles either. It's about distance between skill portfolios. A study of displaced U.S. workers found wage losses were more closely associated with switching skill portfolios than with switching industry or occupation code. A follow-up found that wage losses following displacement are strongly related to distance between the old and new skill set. The further your new job's skills sit from your old ones, the bigger the cut. Move one variable and the distance usually stays short. Move two and it usually grows.

Survivorship bias makes this hard to see. A reinvention that worked becomes a profile. One that didn't is just a quiet person who is a bit poorer and a bit more tired, and nobody interviews them. You're calibrating your odds against a sample scrubbed of every failure.

What does "move one variable" look like in practice?

It means changing your function or your industry, not both. Here's the rule as a table.

Move	What changes	What carries over	Risk
Same function, new industry	Industry context	Your core craft, skills, seniority	Low
Same industry, new function	Day-to-day work	Industry knowledge, network, relationships	Low to moderate
New function and new industry	Almost everything	Very little	High

Real moves cluster around skill overlap. A study of real job transitions reached 76% accuracy predicting transitions using skill-similarity measures, and one-third of the real transitions it examined were to another job classified as the same occupation. If skill distance predicts where people actually land, the short sideways step is the normal move, not the consolation prize.

Picture a hypothetical marketing manager at a consumer goods company, burned out and eyeing UX design.

Weak version: She quits, pays for a design bootcamp, and starts over as a junior designer at an agency. New function, new industry, new seniority level. Three variables moved, so her capital resets to nearly zero and she spends years below her old salary.

Strong version: She takes a product marketing role at a software company. Same function, new industry. She builds product intuition on the job, works daily with designers and product managers, and 18 months later moves internally into a product role. One variable at a time, and the second move is easy because she has already proven herself inside the building.

That second step is often the cheapest one available, which is why an internal transfer is the pivot most people overlook. The company already trusts your work. You're only asking it to trust you in a new chair.

What are the easiest careers to switch to?

The easiest career to switch to is the one closest to your current skills, not the one on a generic "best jobs" list. Ease is personal. It depends on overlap.

You can measure that overlap instead of guessing. LinkedIn's Career Explorer scores how similar two jobs are on a 0 to 100 scale, using data from over 1 billion members, over 41,000 skills and the last five years of transitions. Its own examples: a restaurant server to customer service specialist scores 62, to event coordinator 58, and to sales coordinator 49. The higher the score, the more of your existing capital does the heavy lifting.

For a second opinion, the U.S. Department of Labor's O*NET database profiles over 900 occupations by their knowledge, skills, abilities and tasks. Pull up your current job and the target side by side and list what's missing. If the gap is two skills, you have a short move. If it's twelve, you're looking at a reset, and you should know that before you commit a year to it.

Overlap with the target role	What it means	What to do
High (most core skills shared)	Your experience counts on day one	Apply directly and rename your skills in their language
Medium (a few gaps)	You need proof of one or two new skills	Close the gaps on the job or on a side project first
Low (most skills new)	This is a reset, not a pivot	Find an intermediate role that sits between the two

Does changing careers mean taking a pay cut?

Not necessarily, and the adjacent path is usually the one that avoids it.

When Pew surveyed 6,627 U.S. adults in early 2022, 53% of workers who quit a job in 2021 and were employed again said they had changed their field of work or occupation in the past year, and 56% of those now employed elsewhere said they were earning more. That's a tight labor market and a self-reported measure, so hold it loosely. It still says that changing direction and earning more aren't opposites.

The longer game favors breadth built in steps. A 2024 study in Scientific Reports found that jobs with tightly coherent skill sets did not reach wages above about $60,000, while jobs with diverse skill mixes ranged from $30,000 to $200,000 in U.S. data. You don't build a diverse portfolio by jumping to an unrelated field and starting at zero. You build it by stacking adjacent skills: a marketer who learns product, an analyst who learns the business side.

The far jump is where the cut lives. If you do take one, know the pivot pay cut is real but usually temporary, and plan for how long yours will last.

Is 37 too old to change careers?

No. Age isn't the barrier. Runway and accumulated capital are, and both argue for the adjacent move rather than against changing at all.

People keep moving well into mid-career. In the Bureau of Labor Statistics' long-running survey of Americans born in 1957 to 1964, people held an average of 12.9 jobs from ages 18 to 58, including 2.9 jobs between ages 35 and 44. And LinkedIn's Work Change report says professionals entering the workforce today are on pace to hold twice as many jobs over their careers as those 15 years ago. Movement is normal. The question is distance.

At 37 or 45 you have more occupation-specific capital than a 25-year-old, which makes a full reset more expensive and an adjacent move more powerful. That's the argument in why a career change at 40 fails when you copy a 25-year-old's playbook: do the runway math, then pick the move that keeps the most of what you've built.

Do you need to find your passion before you change careers?

No, and waiting for it can keep you stuck. Meaning often depends on how you relate to the work more than on which work it is.

In the classic study of 196 employees by Wrzesniewski and colleagues, people split across seeing their work as a job, a career, or a calling, and even a narrow group of 24 college administrative assistants was distributed evenly across Job, Career, and Calling. Same title, same campus, three very different experiences of the work. Hold this loosely: it's a small, decades-old sample. But it cuts against the idea that the right job title delivers meaning on its own.

So name the deep want underneath the change, whether that's recognition, impact or autonomy, and ask whether an adjacent move serves it. It usually does. The fuller argument is in why passion follows mastery, not the reverse: build the competence, then point it.

When is a complete career reinvention the right call?

Sometimes it genuinely is, and pretending otherwise would be dishonest.

If your industry is abusive, your health is breaking, or your values have collapsed against what the work demands, a clean break can be the only sane answer. Wage studies measure wages. They don't measure twenty years of quiet misery in a "safe" adjacent role. A financially worse reinvention can still be the right choice.

Three more cases where this advice fails:

• You're early in your career. Under about five years in, your occupation-specific capital is still thin, so a bigger jump costs less.
• Your sector is dying. Staying adjacent inside a collapsing industry isn't conservative, it's slower decline. The one-variable rule assumes your base sector is viable.
• Your occupation is welded to its industry. A data analyst or product manager can carry their craft almost anywhere. A petroleum engineer or a maritime lawyer often can't, because the occupation only exists in one sector. Know which kind you are before you decide how far you can jump.

The boring move costs you the story. You don't get the dramatic before-and-after. You get a slower, far more reliable climb that doesn't photograph well. Decide which cost you'd rather pay with your eyes open.

What to do now

1. Split your current job into function and industry. Write both down. Circle the one you want to change and commit to holding the other still.
2. Score the overlap before you commit. Run your current and target roles through Career Explorer or O*NET and list the missing skills. Two gaps is a pivot. Twelve is a reset.
3. Look inside your own company first. Ask your manager or HR what an internal move into the target function would take. It is usually the shortest bridge you have.
4. Search for the hybrid role. Look for jobs that sit between your old world and your new one, like product marketing between marketing and product. Filter current openings by the function you want, inside the industry you know.
5. Do the runway math. Know how many months of a lower salary you could absorb before you resign, even if you never need them.
6. Plan the second move, not just the first. Write down where the adjacent role could lead in 18 to 24 months. Two short steps get you further than one long jump.

Thinking about a career change but not sure which move is short and which is a reset? Tell Praxy what you do now and where you want to end up on WhatsApp. I'll split your job into function and industry, map the roles your skills already transfer to, and show you the adjacent first step before you bet a year on a guess.