The question "should I go back to school?" is almost always really three different questions: Do I need a credential? Do I need specific knowledge? Do I need a career pivot? The answer to each is different, and mixing them up leads to expensive decisions that solve the wrong problem.
What school actually buys you
A graduate degree or professional program buys you four things: a credential (the piece of paper), specific knowledge (what's actually taught), a network (the people you meet), and a signal (the selective institution telling the market something about you). The relative weight of each varies dramatically by field and by what you're trying to accomplish.
For medicine and law, the credential is non-negotiable — you can't practice without it. For MBAs, the signal and network often matter more than the knowledge. For technical fields like software engineering or data science, the knowledge and portfolio frequently matter more than the degree. Understanding which of the four you actually need is the first question to answer.
The opportunity cost that rarely gets calculated
Full-time graduate programs don't just cost tuition — they cost the salary you would have earned during those 1–3 years, and sometimes more. A $60,000/year MBA program over two years isn't a $120,000 decision. When you factor in 2 years of foregone income at $80,000, it's a $280,000 decision, before interest on student loans.
This doesn't mean the decision is wrong. It means the return on that investment needs to be proportionally significant. A credential that increases your annual income by $10,000 doesn't justify a $280,000 outlay. A credential that moves you into a field where you'll earn $50,000 more per year might, over a decade.
When going back to school is genuinely worth it
- ◆You're trying to enter a field that requires a specific credential (law, medicine, clinical psychology, certain engineering specialties).
- ◆You're changing careers and need credibility in a new domain — and the degree is the fastest path, not the only one.
- ◆You're targeting a top-tier institution and the network/signal value of that institution specifically is the ROI.
- ◆You've identified a specific income jump and the math clears with honest numbers, not optimistic ones.
- ◆You have clear funding that won't require decades of repayment.
When it probably isn't
- ◆You're going back to school because you're unhappy in your current role and don't know what else to do.
- ◆The field you want to enter doesn't actually require the degree — it requires experience and portfolio.
- ◆You're choosing a program primarily for the experience or the break, not for specific career outcomes.
- ◆The income premium of the degree is unclear, or clearly modest relative to the cost.
- ◆You're taking on significant debt for a program at a school where the credential doesn't signal much to employers.
The career change case
If you're switching fields, a degree is one option — but often not the fastest or most cost-effective. Bootcamps, certifications, freelance work, and building a portfolio while working have moved people into tech, design, finance, and consulting without a second degree. The question isn't whether you need to build credibility in the new field — you do — but whether a degree is the most efficient way to do it.
The honest test: find people who are doing what you want to do. How did they get there? If most of them have the degree you're considering, that tells you something. If most of them got there another way, the degree may be solving the wrong problem.
Questions to answer before deciding
- ◆Can I name three specific jobs I want after this program, and does this credential meaningfully help me get them?
- ◆Have I talked to recent graduates about actual outcomes, not just average reported salary increases?
- ◆What's the total cost including foregone income, and what does my realistic income bump need to be to justify it?
- ◆Is there a faster, cheaper path to the same outcome?
- ◆Am I solving a career problem or a life satisfaction problem? (Both matter, but they have different solutions.)