How to Tell If a Career Is Actually Good — Before You Commit Years to It

Every week, thousands of people type some version of the same question into Google: is nursing a good career, is accounting a good career, is software engineering still worth it. The answers they get are almost always from someone with a stake in the answer — a school, a bootcamp, a recruiter, or a stranger on Reddit describing one personal experience as if it were a trend.

Here is a framework you can apply to any career yourself, using public data. Five tests. If a career passes four of them, it is probably a good choice for someone. Whether it is good for you is the fifth test, and it is the one most people skip.

Test 1: Is it growing or shrinking?

The Bureau of Labor Statistics publishes ten-year employment projections for every occupation. This is the single most useful number in career planning and almost nobody looks at it. A career growing 20% will have openings, upward mobility, and salary pressure in your favor. A career shrinking 5% means competing with laid-off experienced people for fewer seats.

Fastest-growing occupations in our data right now:

• Solar Photovoltaic Installers — +48% through 2034
• Wind Turbine Service Technicians — +45% through 2034
• Nurse Practitioners — +40% through 2034
• Nurse Anesthetists — +38% through 2034
• Data Scientists — +36% through 2034
• Business Intelligence Analysts — +36% through 2034
• Clinical Data Managers — +36% through 2034
• Prompt Engineer — +35% through 2034

And the ones actively contracting — not careers to avoid entirely, but ones where you should have a specific reason for entering:

• First-Line Supervisors of Office and Administrative Support Workers — -3% projected change
• Switchboard Operators, Including Answering Service — -3% projected change
• Telephone Operators — -3% projected change
• Bill and Account Collectors — -3% projected change
• Billing and Posting Clerks — -3% projected change
• Bookkeeping, Accounting, and Auditing Clerks — -3% projected change

Test 2: Does the pay justify the entry cost?

Compare median salary against what it costs to get in — in both money and years. A career paying $52,000 after a six-month certificate is a better financial deal than one paying $68,000 after four years and $80,000 in debt. Run the actual math over ten years, not the starting-salary comparison everyone makes.

The trap: comparing your potential salary to your current one instead of to the total cost of the switch. A $15,000 raise that costs $60,000 to obtain takes four years just to break even, before interest.

Test 3: Can it be automated or offshored?

Ask two questions about the daily work: does it require being physically present, and does it require judgment that changes case by case? Work that is repeatable, rule-based, and screen-only is the most exposed to both AI and offshoring. Work that is physical, relational, or high-consequence is the most protected.

This is not about whether AI will “replace” the job — it rarely happens that cleanly. It is about whether AI makes one person able to do the work of three, which quietly shrinks hiring for everyone else.

Test 4: Is there a path, or just a job?

Some careers have visible ladders: clear next roles, predictable salary progression, transferable skills. Others are flat — the job at year ten looks like the job at year one, with slightly better pay. Neither is wrong, but you should know which you are signing up for.

The question to ask people already in it: what do the people five years ahead of you do? If nobody can answer, that is the answer.

Test 5: Does it fit how you are actually built?

This is the test that determines whether you last, and the one no salary chart can answer. A career can be growing, well-paid, automation-proof, and full of upward mobility — and still be the wrong career, because the day-to-day work fights your nature.

Concretely: how much unstructured time can you tolerate before you feel unmoored? Does conflict energize you or exhaust you? Do you need to see the result of your work, or can you contribute to something abstract? Would you rather go deep on one problem for a year or touch twelve things a week?

People rarely quit careers because the salary was wrong. They quit because the daily texture of the work was wrong, and no amount of growth data warned them.

Putting it together

Score any career you are considering: growing, affordable to enter, hard to automate, has a ladder, fits your wiring. Four out of five is a strong bet. Three is workable with a clear reason. Two means you are choosing with your hopes rather than the evidence.

Tests one through four you can research in an afternoon. Test five is the hard one — which is exactly why we built an assessment for it.