Good PhD Student
Bad PhD Student

A self-selection guide

Abhishek Gupta

Associate Professor, ECE | The Ohio State University
Co-Director, IITB-OSU Frontier Center
Founder, Ensemble Control Inc.

Why This Document Exists

A PhD is a highly leveraged, 5+ year decision — one of the largest investments of time, energy, and foregone income you will ever make.

Being a good PhD student is hard. Most struggles are not caused by lack of intelligence — they are caused by habits, mindset, and expectations that were never articulated clearly.

The goal of this deck

Not to scare you away — to help you self-select with clear eyes. Each slide carries one message. If most “Good” columns sound like you (or who you’re willing to become), a PhD may be a great fit.

Format borrowed shamelessly from Ben Horowitz & David Weiden, “Good Product Manager / Bad Product Manager” (AOL, 1997).

A final note, up front

A PhD is a demanding and high-autonomy job. Individuals should make sure they’re up to the challenge — that is what this deck helps you decide.

Summary Points

  • CEO of the thesis — ownership
  • Output over intention — research results
  • Manage your advisor — the most important relationship
  • Communicate in writing — clearly and proactively
  • Marathon, not sprint — lifestyle
  • Know what you don’t know — scholarship
  • Failure is data — resilience
  • Ask why you’re here — motivation

Ownership

A Good PhD Student is the CEO of Their Thesis

Good PhD students…

  • treat the thesis as their company: they drive the vision, the roadmap, and the deadlines
  • know their advisor is a board member, not a boss — advice is input, the decision is theirs
  • show up to meetings with an agenda: “here is what I tried, here is what I found, here is my plan”
  • take responsibility for outcomes, including failed experiments

Bad PhD students…

  • wait to be told what to do next
  • treat the PhD as a job assigned by the advisor — no task, no work
  • have lots of excuses: “my advisor didn’t reply,” “the cluster was down,” “the other group has more funding”
  • point out, after failing, that they predicted they would fail

One message: Nobody will care about your thesis more than you. If you need someone else to care first, self-select out.

Research Output

Good PhD Students Ship

Good PhD students…

  • produce a steady stream of artifacts: drafts, proofs, plots, code, talks — every single week
  • write the paper while doing the research, not after
  • know a submitted imperfect paper beats a perfect one that lives in their head
  • keep experiments reproducible — scripts, seeds, versioned data

Bad PhD students…

  • hoard ideas, waiting for the “big result” before writing anything
  • confuse being busy with producing output — six months of reading, zero pages written
  • rerun experiments by hand and can’t reproduce their own figure from three months ago
  • polish section 1 for weeks while sections 2–6 don’t exist

One message: Research that isn’t written down, doesn’t exist. Ship early, ship often.

Relationship with Advisor

Good PhD Students Manage Their Advisor

Good PhD students…

  • understand the advisor’s time is the scarcest resource in the group — and use it deliberately
  • come with attempted solutions, not just problems: “I tried A and B; I think C, because…”
  • make it easy to help them: send the draft before the meeting, with specific questions
  • ask for feedback early, when it’s cheap to change course

Bad PhD students…

  • hide when things go badly — and surface three weeks later, further behind
  • use meetings as status theater instead of decision points
  • expect the advisor to schedule, remind, and chase them
  • take feedback personally, defend instead of probe, and leave with nothing changed

One message: Your advisor is a resource you manage, not a manager you report to. The relationship is yours to run.

Communication

Good PhD Students Communicate Clearly — In Writing

Good PhD students…

  • send a short written update before every meeting: what happened, what’s blocked, what’s next
  • err on the side of clarity — willing to explain the obvious to make sure it’s understood
  • state what they know, what they don’t, and which is which
  • deliver bad news immediately: a stalled project reported today is a plan; reported next month, it’s a crisis

Bad PhD students…

  • go silent for weeks, then reappear with a wall of confusing results
  • obfuscate — defend gaps in knowledge instead of filling them
  • over-state their progress and burn their credibility
  • assume “my advisor knows what I meant”

One message: In research, unclear writing is unclear thinking. If you hate writing, you will hate the PhD.

Lifestyle

Good PhD Students Run a Marathon, Not a Sprint

Good PhD students…

  • work consistent, focused hours — 5–6 deep-work hours a day, sustained for 5 years, wins
  • protect sleep, exercise, and relationships as research infrastructure, not luxuries
  • accept the stipend lifestyle deliberately — they did the math and chose it
  • take real breaks and come back sharper

Bad PhD students…

  • binge 80-hour weeks before deadlines, then crash for a month
  • perform “lab presence” — long hours, little thought
  • resent the stipend every day while classmates in industry earn 4×
  • let the PhD consume identity, health, and every relationship — then burn out in year 3

One message: The PhD is won by the compounding of ordinary weeks. If your only gear is sprint, five years will break you.

Scholarship

Good PhD Students Know What They Don’t Know

Good PhD students…

  • read deeply and continuously — they know the 10 papers that matter and why
  • can state the difference between opinion, hunch, and proven fact in their own claims
  • seek out the strongest counter-argument to their result before a reviewer finds it
  • learn the math/tools the problem demands, not just the ones they already like

Bad PhD students…

  • cite abstracts they never read
  • ask leading questions of their own data and get biased answers
  • compare their future method to today’s baselines
  • have blinders on — notice the field changed only when their paper gets rejected

One message: Your job is to fill gaps in knowledge — including your own — not to defend them.

Resilience

Good PhD Students Treat Failure as Data

Good PhD students…

  • expect most experiments to fail — that’s what research means
  • extract the lesson from every rejection: “the reviewers misread section 3 → section 3 was unclear”
  • distinguish “the idea failed” from “I am a failure”
  • ask for help before the hole gets deep

Bad PhD students…

  • interpret a rejected paper as a verdict on their worth
  • abandon a project at the first negative result — or worse, never test the risky part at all
  • ruminate for months instead of running the next experiment
  • suffer alone until quitting feels like the only option

One message: A PhD is 5 years of mostly failing, punctuated by discovery. If you need frequent external wins to function, choose a different game.

Motivation

Good PhD Students Know Why They’re Here

Good reasons to do a PhD

  • you have questions that keep you up at night and can’t not work on them
  • you want a research career — academia, national labs, industrial research — where the PhD is the actual license
  • you crave 5 years of protected time to become the world expert in one thing
  • you’ve tried research (REU, thesis, internship) and wanted more

Bad reasons to do a PhD

  • the job market is scary and this postpones deciding
  • your parents / culture expect the title
  • “Dr.” sounds prestigious
  • you were good at courses, so this seems like the natural next level — (it isn’t: research ≠ coursework)
  • someone offered funding and you had no better plan

One message: The PhD rewards intrinsic motivation and punishes borrowed ambition. External reasons run out around year 2 — the questions have to be yours.

Self-Selection

Should You Do a PhD? The Honest Checklist

Ask yourself… Sounds like a PhD Sounds like industry (for now)
Ambiguity “No one knows the answer” excites me I want well-defined problems and clear wins
Time horizon I can work 18 months for one result I need to ship something every few weeks
Autonomy I’ll set my own direction and deadlines I do my best work with structure and a manager
Writing I’m willing to write, rewrite, and rewrite Writing is the part I’d pay to avoid
Money Stipend for 5 years is a trade I accept Foregone salary would eat at me daily
Failure Rejection is feedback; I iterate Repeated rejection would grind me down
The question There’s a topic I can’t stop thinking about I like the field, but no burning question

One message

Neither column is “better” — industry is a great outcome, and leaving research is not failure. Choosing wrong for 5 years is the failure. Score yourself honestly on the next slide.

Self-Assessment: Rate Yourself Honestly

0Fit score10
Move the sliders…

A slider score is not destiny — it’s a structured conversation with yourself. Discuss the low bars with a mentor.

The Final Word

Being a good PhD student is so hard that many students at most universities struggle — not because they lack intelligence, but because nobody articulated the principles clearly.

If the “Good” columns sounded like you

The PhD can be the most intellectually alive five years of your life. Come talk to us.

If they didn’t

That is a successful outcome of this talk — you just saved five years. Industry, startups, and national labs need brilliant people too.

One last message

A really good PhD student doesn’t just survive the degree — they use it to become the world’s expert on one question that matters to them. Make sure the question is yours.