If I could redo my PhD

• Intelligence isn't a fixed label the universe stamped on your forehead at birth—it's a muscle, and research is just one more thing you get less bad at by doing it badly, in public. Heads up: your taste will sprint out ahead of your skill and then turn around to judge you. You'll be able to diagnose, in forensic detail, exactly why your own work is mediocre long before your hands can do anything about it. That gap isn't a sign you don't belong. It's the engine. (It's also extremely annoying.) Keep going until your hands catch up to your eye.

• Don't burn so much fuel on what other people think—they're reviewing an early cut while you've already rewritten the whole thing. And I say this with enormous love: nobody is thinking about you nearly as much as you'd hope or fear. Everyone is the harried lead of their own feature-length production, frantically learning their own lines, convinced the spotlight is on them. The audience you're agonizing over is, at this exact moment, mostly squinting at its phone wondering if that last text came across weird.

• Make peace with where you are right now—it's a stage, not a verdict, even when the verdict feels like it's already been handed down. Progress is slower and lumpier than anyone admits in their tidy little success story, and today's version of you is just a rough draft of tomorrow's, typos and all. So do your best in the moment and let the rest unfold instead of trying to storyboard your entire life in advance like a maniac. You're not behind. You're in progress.

First, the reassuring part: most advisors are decent people who want you to do well, and a good one is among the best things about a PhD. None of what follows is about unmasking villains—it's about fit, which is a much more useful and much more common question. Most of the time you're just telling an overcommitted-but-caring mentor from someone who'd be a poor match for how you work. Different people need different things from a supervisor; the trick is knowing what you need and reading honestly for it.

So, some things to actually look at.

• Prestige, and whether it answers emails. Scout the academic stars—the people who bring momentum and ideas that make you sit up straighter. The catch nobody prints on the brochure is that the most dazzling are often dazzling because they've said yes to everything, which can leave the time for you looking like a calendar invite that keeps getting moved. Prestige is wonderful. Prestige that answers your emails is better. This isn't a character flaw in anyone—it's arithmetic, and it's why the question is fit, not merit.

• Where their work lands. Top-tier journals, or a quiet graveyard of proceedings nobody cites? A strong track record tells you two encouraging things: they can clear the bar you're hoping to clear, and they know the game's unwritten rules well enough to teach them. You want your ambition pointed at someone whose ceiling sits comfortably above your floor.

• How they share credit. Do they co-author with students? Often a lovely sign—but the byline alone won't tell you much, because co-authorship norms are genuinely all over the map between fields and even between groups. So don't read the name; read the work. Pull up a few recent student-coauthored papers and look for where the student clearly drove it—a method they built, a section that's unmistakably theirs, an idea that's their own. When you can see students leading, it says something warm about how this person works with people junior to them. That's the signal worth chasing, and it's visible if you actually read.

• How the work gets funded. Some advisors pull in competitive grants, where strangers read the idea and wager real budget on it; others build careers on smaller awards, fellowships, or commissioned work. Both are completely legitimate—they're just different bets, peers-on-the-question versus a-funder-on-the-activity, and different fields reward different mixes. Worth remembering that a CV is a story someone curates, not a neutral ledger—yours will be too, someday. Read everyone's generously, including your own.

• Which rooms they're in. Conduct a respectful, scholarly Google-stalk of their collaborators. (Everyone does this; we just don't admit it at parties.) Who someone works with tells you which rooms they're invited into, and an advisor in good rooms can, eventually, hold a door open for you. You're not just picking a person—you're marrying into an academic family, in-laws and all.

• And the thing the title hides: career stage. A senior advisor brings gravity, stability, and a contacts list three decades deep—and sometimes a schedule so fragmented you'll come to know them mainly through their out-of-office replies. A junior advisor is often hungrier and more hands-on—and also running their own white-knuckle sprint toward tenure, which makes their attention a renewable but genuinely scarce resource. Either can be wonderful. Plenty of miserable students have brilliant famous advisors; plenty of happy ones have advisors you've never heard of. The real question is never how impressive someone is. It's whether they have the bandwidth and the genuine desire to mentor someone who isn't them.

• If you want to work on real problems, resist the urge to spend an entire career inside one disciplinary silo. The good stuff tends to live in the gaps between fields—the unclaimed territory where the methodology police haven't set up checkpoints yet. Most of the best ideas arrive sideways: you're staring at one problem and a completely different one falls out of your pocket.

• Start with good sources, and define "good" generously. Serious investigative journalism counts—reporters often map a problem years before the literature catches up. And don't sleep on the grey literature: think tanks, policy institutes, advocacy shops. Some of it is special pleading dressed as research, sure, but buried in there are datasets, field reports, and questions that will never reach a journal—because nobody gets tenure for them, not because they aren't worth asking.

• Train yourself to think across fields, because the borders are far more porous than the departmental org chart pretends. A tool built for one problem is often quietly perfect for an unrelated one:

  • Health-insurance math—pricing rare, expensive events—is also flood-risk math; a hospital stay and a flooded basement are the same actuarial animal.

  • The industrial-organization toolkit, built to dissect markets and market power, untangles an agricultural supply chain without breaking stride.

  • Sociology keeps handing you the variable that actually matters, even when it's cheerfully agnostic about what caused it.

  • A jump in environmental sensing—satellites, cheap sensors—quietly rewrites how some field with nothing to do with the environment collects its data.

• And go to the talk in the other department—yes, the one you have no business attending. You won't follow every slide, and that's fine; that's half the point. You'll leave with a new way of seeing your own problem, or at minimum a better question than the one you walked in with. A better question is never a wasted afternoon.

• Actually engage with the thing, even when the thing has the narrative momentum of drying paint. Sit with the argument instead of skimming past it, and watch which ideas make the room lean forward and which quietly die the moment someone pushes back. A concept that can't survive one hard question is telling you something.

• Whoever you are, take up the space. Ask the question—even the one you're sure is too basic, especially that one, because half the room is quietly wondering the same thing and only the bravest person says it out loud. Visibility isn't vanity; the people who get remembered are simply the ones who were in the conversation. You don't need to be the smartest voice in the room. You need to be a voice in it.

• Don't panic when half of it sails over your head in real time. Experts skip steps—not to lose you, but because the thing they're skipping has been obvious to them for fifteen years and they've forgotten it was ever hard. The gap between "I didn't follow that" and "I'm not smart enough for this" is mostly just the steps they left out; fill them in afterward, where the real learning lives anyway. And breathe—everyone in that room felt exactly this lost at some point, including the person at the front pretending they didn't. The people who make it through aren't the ones who never struggled; they're the ones who kept their sense of humor about it.

• And while you're there: scout. Every seminar is a low-stakes audition for future mentors, collaborators, and the speaker you'll email next week saying "I really enjoyed your talk." Expanding your circle isn't a cynical side quest—it's half of what the room is for. Sit near someone interesting. Say hello after. That's the whole technique.