How to Email a Professor About PhD Positions (With a Template That Gets Replies)

Ten seconds and a very full inbox

Here's what happens to most of these emails. A professor opens the laptop on a Tuesday in November, sees forty unread messages, and starts triaging. Somewhere in that stack are six from students they've never met, each opening with some version of "I am very interested in your research and would be a great fit for your lab." They read the first line, maybe the second, and archive. Ten seconds per email, and most of that is spent deciding you're not worth an eleventh.

I've sent dozens of these over the years (a master's RA spot first, then PhD positions), and sat next to advisors clearing their inboxes, listening to the running commentary. It's bleaker and simpler than you'd think. During application season, a well-known professor in a popular field gets several of these a week, and they blur together because they are, genuinely, almost identical.

The good news is buried in that same fact. The bar is on the floor. When nearly every email is interchangeable flattery, the one that shows a real person did real homework stands out like a lighthouse. You don't need to be brilliant. You need to be specific, brief, and clearly not a mail merge. That's a far easier target than impressing a famous scientist.

The one thing a generic email cannot fake

If you take nothing else from this, take this: prove you actually read their work. One accurate, specific sentence about a recent paper does more than three paragraphs of praise ever will.

Think about why. Flattery is free. Anyone can write "your work on graph neural networks is fascinating and aligns perfectly with my interests." It costs nothing, means nothing, and every professor has read it two hundred times. But "your 2024 paper on message passing over dynamic graphs (the part where you froze the edge weights after the third layer) made me wonder whether the same trick would hold up on temporal knowledge graphs" cannot be written without doing the reading. It carries proof of effort inside itself. That's the whole game.

So before you write a word, find one or two of their recent papers (recent matters, because interests drift, and what they were known for in 2016 may not be what they're funding now) and read enough to have one honest thought. Not the whole paper. The abstract, the figures, the conclusion, the "future work" line. You want a single concrete point you can connect to something you want to do. Find it and you have the spine of your email; everything else is scaffolding around that one sentence.

If you can't find that sentence, if their work genuinely doesn't connect to anything you care about, that's useful information too. It might mean they're the wrong person to email, and I'd rather you learn that now than after you've joined the lab.

What actually goes in the email

A good cold email to a professor is about five short paragraphs and comes in under two hundred words. Longer than that and you're asking a busy stranger to do work. Here's what each part is for.

The subject line. This is the first filter, so make it specific and human. "Prospective PhD student" is invisible. "Prospective PhD student: your work on X, my background in Y" gives them a reason to open. Name their method or your shared topic. Some people put the cycle in there too ("Fall 2026 PhD applicant"). The subject line alone should signal this isn't spam and isn't generic.

The opening line: who you are, in one breath. Name, what you're doing right now, your field. "I'm a final-year computer science undergrad at [university], focused on natural language processing." That's it. No "I hope this email finds you well" preamble. They need to place you in about four seconds and then get to why you're writing.

The why-them paragraph. This is the sentence you earned by reading. Name the specific paper and the concrete idea, then connect it to what you want to do. This is where most emails die and where yours will live. Two or three sentences. You're not writing a literature review; you're showing one genuine point of contact between their work and your head.

The what-you-bring line. One real, relevant thing. A method you know, a project you built, a result you got. "Last summer I built a retrieval pipeline for a legal-tech startup and wrote the evaluation harness myself." Keep it true and relevant. Do not inflate. Professors have finely tuned detectors for a padded CV, and the moment they catch one exaggeration they reread the whole email as marketing. One honest concrete thing beats five vague impressive-sounding ones.

The ask. Small and clear. You usually want to know two things: are they taking students for the coming cycle, and would they be open to a short conversation. That's a low-cost yes. Don't ask them to evaluate your CV in detail, don't ask for funding guarantees, don't propose a forty-five-minute call. "I wanted to ask whether you're taking new PhD students for Fall 2026, and if so whether you'd be open to a brief chat" is easy to answer even for someone with forty unread messages.

The sign-off. Short. Thank them for their time, say your CV is attached, done. One PDF, one or two pages, sensibly named. No ten-page statement of purpose, no writing sample they didn't ask for, no zip file. If they want more, they'll ask, and that request is exactly the reply you're hoping for.

A template you should take apart

Here's a skeleton. It's deliberately plain, because a fancy template is still a template. Use it to see the shape, then rewrite every single bracket in your own voice.

Subject: Prospective PhD student: your work on [their method or topic], my background in [your area]

Dear Dr. [Last name],

I'm a [current role, for example a final-year MSc student in X at Y University] working on [your narrow area]. I've been following your group's work, and your recent paper on [specific paper or finding] stood out: [one honest, specific sentence about the idea and how it connects to something you want to do or already know].

[One real thing you bring: a method, a project, or a result, in a single sentence. Keep it true, and keep it relevant to them.]

I wanted to ask whether you're taking new PhD students for [cycle, for example Fall 2026], and if so whether you'd be open to a short conversation. I've attached my CV for context.

Thank you for your time,
[Your name]

Now the warning, and I mean it: if you keep the rhythm and only swap the bracketed words, it still reads like a template, because it is one. What makes an email land is the part you cannot template: the actual specific thought about their actual specific paper. Treat the brackets as prompts, not fill-in-the-blanks. If your finished email could be sent to a second professor by changing three words, you haven't finished.

Does emailing even help? It depends where you are

This is the part most guides skip, and skipping it leaves students demoralized. Whether your email matters (and how much) depends enormously on your field and your country. Anyone who gives you "always email professors before applying" as a universal rule hasn't looked outside their own corner.

Start with a big chunk of US programs, especially in the sciences and social sciences. In many of them, admission is decided by a committee, not by any single professor. Someone you email can't simply "accept" you; they can advocate, but the cohort is assembled by a group weighing the whole pool. Your email isn't a key that opens a door. It does two smaller, real things: it signals genuine fit to someone who might later be in the room when your file comes up, and it tells you whether they're even recruiting this year. Both are worth having. Neither is a guarantee, so don't read a warm reply as an admission or a cold silence as a rejection. You apply to the program either way.

And in some of those programs, professors simply won't answer cold emails at all. It's departmental culture, sometimes even explicit policy, and it says nothing about you. I emailed people who never replied and later found out they had a standing rule against pre-application contact. You still submit the application. The email was never the thing standing between you and the offer.

Now flip to the other world. In computer science and much of engineering, and across most of Europe and the UK, individual professors carry far more weight. PhD positions are often tied to a specific grant with a specific advisor, and that advisor is effectively hiring. Here, contacting them before you apply isn't just allowed; it's frequently expected, and applying cold without a word can look like you didn't do your homework. A European PhD advertised on a funded project is closer to a job posting than to a US-style cohort admission. The email sits much closer to the actual decision.

So find out your field's norm before you assume. Ask a current PhD student in the department. Read the program's admissions page. Notice whether positions are advertised individually with named supervisors or admitted as a class. Then set expectations to match. I wrote a whole piece on reading these signals in how to tell if a professor is actually accepting PhD students. It's worth reading before you send anything, because half of "why won't they reply" is really "they weren't recruiting and you couldn't have known."

The mistakes that get you deleted

Most rejected emails aren't rejected on the merits. They trip one of a handful of wires and get archived before the professor reaches the good part. The ones I see constantly:

• The obvious mass send. A visible BCC, a "Dear Professor" with no name, or a body so generic it could go to anyone in the department. If they can tell at a glance that four hundred other people got this, you're done.
• Over-flattery. "Your groundbreaking, world-renowned research has inspired me since childhood." It reads as insincere and slightly desperate, and it eats the space a real thought should fill.
• Asking for too much. Funding promises, an admission guarantee, a long meeting, detailed feedback on your whole application. You're a stranger. Big asks from strangers get ignored.
• The giant attachment. A ten-page statement of purpose, a full thesis, three writing samples. Send a short CV and nothing else unless asked.
• Typos and the wrong name. Calling Dr. Chen "Dr. Smith" because you forgot to change it from the last email tells them exactly what you did. "Mr." or "Ms." instead of "Dr." or "Professor" is a small cut that still bleeds. Proofread, then check the name again.
• The vague ask. "I would love to discuss opportunities." What opportunities? What are you actually asking them to do? Give them one clear, answerable question.
• The wall of text. Six dense paragraphs with no breaks. Nobody reads it. Short paragraphs, white space, done.

None of these are about intelligence. They're about care and restraint, which is exactly the signal a professor is reading for, because care and restraint are what working with you for five years will require.

When to hit send, and when to follow up

Timing matters more than people expect. Send during working hours on a weekday: a Tuesday, Wednesday, or Thursday morning tends to beat a Monday, when the weekend backlog buries you, or a Friday, when your email waits until Monday and gets buried anyway. Soft advice, not physics, but a message read at 9 a.m. midweek has a better shot than one that arrives Saturday night.

Bigger picture: send with enough runway before the application deadline. A few weeks to a couple of months ahead is a reasonable window, early enough that they're still forming plans for the cohort, late enough that the cycle feels real to them. Too early and it's not on their radar; too late and the answer is "we've already made our decisions." I've gone deeper on getting this window right in when to contact professors and how the PhD timeline works, because it shifts by country and it's easy to be off by a month in the wrong direction.

If you hear nothing, one follow-up is fine. Wait about ten days, reply on your own thread so the context is right there, and keep it to two or three sentences: a light nudge, the original question restated, nothing guilt-trippy. Something like "I know this is a busy season, just wanted to check whether you're taking students this cycle." If the second email also gets silence, stop. A third follow-up doesn't read as persistence; it reads as someone who can't take a hint, which is a worse impression than never having emailed. Silence is an answer too.

What a good reply looks like, and what to do with it

Say it works. A professor writes back: "Thanks for reaching out. Yes, I'm taking students this year. Can you tell me more about your experience with X?" This is the moment, and plenty of people fumble it by going quiet for a week or answering with a novel.

Reply fast, within a day if you can, and answer the actual question specifically. If they asked about your experience with reinforcement learning, tell them the concrete project: what you did, what broke, what you learned. Three or four sentences, not three paragraphs. Match their length: if they were brief, be brief; if they opened a real technical thread, go a little deeper, because now you're having a conversation rather than pitching. When they offer a call, come with two or three genuine questions about the work, not "so what's the lab like." You want them to hang up thinking you'd be interesting to have around.

If the reply is a polite no (not taking students, no funding this year, moving institutions), thank them and mean it. A gracious no today is a warm contact in two years. People remember who was decent when the answer was disappointing.

Handling silence is its own skill. It will happen a lot, and it'll feel personal even though it almost never is. Don't send a wounded follow-up. Don't read it as a verdict on your worth. Log it, move on, keep your list moving. The professors who ghost you aren't a signal about your future; they're a signal about their inbox. And if the ghosting has you second-guessing whether someone was ever a realistic advisor to begin with, that's worth thinking about honestly: how to find a PhD advisor gets into building a list of genuinely plausible fits, which makes the silences sting less because you had good reasons for each name.

PhD, postdoc, or a master's-slash-RA spot: what changes

The skeleton holds across all of these, but the emphasis shifts.

For a PhD email, you're selling potential and fit more than a finished record. You haven't done a dissertation; nobody expects you to have one. Show that you can read their work, think about it, and that you have the raw materials: a relevant project, a method, a genuine question. The subtext they're reading for is whether you'll still be motivated and functional in year four.

A postdoc email is a different animal, and if you're reading this as an undergrad it's not yet your problem. But briefly: it's much closer to a straight job application. You have a track record now, so lead with it. Specific publications, a clear research direction, a concrete story about how your work extends or complements theirs. The "what you bring" section grows; "prove you read their work" becomes "prove our two programs genuinely connect." Funding source and start date are fair to raise early, since postdoc hiring is usually tied to specific grants.

For a master's position or an undergraduate RA spot, scale everything down. The ask is smaller, so the email can be lighter. You're often asking to help with existing work rather than launch your own, so emphasize reliability and relevant skills over grand vision. "I've taken the machine learning sequence, I'm comfortable in PyTorch, and I can commit ten hours a week next semester" is precisely what an overloaded lab wants to hear. Professors staff these roles more casually and more often, so your hit rate can be higher, and a good RA email is often the thing that turns into a PhD conversation a year later.

The part that takes real time

The honest catch is that the good version (the specific, researched, genuinely-not-generic email) takes real time per professor. Reading a paper closely enough to have one true thought, then shaping it into four tight paragraphs, is maybe thirty to sixty minutes each. Multiply that across a list and it's a lot of blank-page staring, which is where people quietly fall back on the mail-merge version that doesn't work.

That's the specific problem ApexApply tries to take off your plate: it drafts a personalized first version grounded in each professor's actual research and your own background, so you spend your time editing and sharpening instead of facing an empty screen. It won't have the one genuine thought for you (that part is still yours, and it should be), but it can get you to the point where having it is the only work left.

Whatever you use to get there, the rule underneath doesn't change. Be a real person who did the reading, ask for something small and clear, and send it to people who are plausibly a fit. That's the whole thing; everything else here is detail around it.