How to cold-email professors, physicians, and mentors

Why cold email is the highest-leverage skill you have

Think about what a connection actually does for a student. It gets your name in front of a busy person with a small amount of credibility attached. That's it. A well-written cold email does roughly the same job. It costs you fifteen minutes and nothing else, you can send twenty of them, and the worst realistic outcome is that a stranger doesn't reply.

Professors and physicians get cold emails constantly, and most of them are bad. They're vague, they're clearly copy-pasted to fifty people, they ask for something enormous in the first sentence, or they contain three paragraphs of flattery and no actual request. This is very good news for you. The bar is low enough that a specific, short, obviously-written-for-this-person email genuinely stands out.

There's a second reason this skill matters more than the opportunity itself. Cold-emailing is the same muscle you'll use to request a letter of recommendation, ask a program director a question, contact a residency, or reach out to a physician in a specialty you're curious about. You're going to use this for the next fifteen years. Start being good at it now.

One mindset shift before you start: you are not begging. You're offering someone reliable, motivated help and asking them to consider it. Most labs are perpetually short on people who show up consistently. Most physicians remember what it felt like to want in. You're not an imposition. You're a candidate.

Finding the right person to email

The single biggest predictor of whether you get a reply is whether the person you emailed was a sensible person to email. Spraying a generic message at every biology professor at your school is worse than sending four carefully aimed ones. Aim first.

Start close and work outward. People with any tie to you, however thin, reply at much higher rates than total strangers. A professor at your own school beats a professor at a school across the country. An alum of your high school beats a random physician. Someone your community college transfer advisor mentioned beats someone you found on Google. Use whatever thread exists, and say so in the first two sentences.

Here is where to actually look.

• Your own school's department pages. Every biology, chemistry, psychology, public health, and neuroscience department has a faculty directory with research blurbs. Read the blurbs. Make a list of five to ten people whose descriptions you can honestly say you find interesting.
• Individual lab websites. Most PIs (principal investigators, meaning the person who runs the lab) keep a site with a People page, a Publications page, and sometimes a "Join Us" page that says exactly what they want from applicants. Read the Join Us page before you write. If it gives instructions, follow them precisely.
• PubMed (pubmed.ncbi.nlm.nih.gov). Free, run by the National Library of Medicine. Search a disease, a technique, or a topic you care about, then add your city or university name to the search to find people doing that work near you. Sort by most recent to see who is actively publishing.
• NIH RePORTER (reporter.nih.gov). Free searchable database of NIH-funded projects. You can filter by institution and by state. A lab with a current NIH grant has money, which often means it has room and sometimes means it can pay you. This is one of the most underused tools available to students.
• ClinicalTrials.gov. Search by condition and location to find trials running at hospitals near you, and see the listed principal investigator and study contact. Clinical research coordinators listed on these pages are excellent people to email.
• Hospital "Find a Doctor" directories. Every hospital system has one, searchable by specialty and language. This is the cleanest way to find physicians in your city who practice what you're curious about. Some directories list an office email or a department contact; if not, the department's administrative office can route your message.
• Your school's alumni network. Handshake, your career center, LinkedIn's alumni filter, or a simple email to your college advising office asking "do we have alumni working in medicine who are open to talking with students?" A shared school is a real connection, and alumni reply far more often than strangers.
• LinkedIn, used carefully. Good for finding people and confirming their current job and background. Less good as a place to send the actual message. If you can find someone's institutional email, use that instead. Institutional emails at universities are usually a predictable pattern, like firstname.lastname@school.edu, which you can confirm from any department page.
• Conferences and program pages. Look at who spoke at your school's research symposium, who runs the summer program you want, or who is listed as faculty on a program you were rejected from. Rejection from a program does not disqualify you from emailing the people in it.

Reading a paper well enough to mention it honestly

The line that makes a cold email work is the one that proves you looked at this person's actual work. You do not need to understand a paper the way a graduate student does. You need to understand it well enough to say something true and specific about it. Faking this is worse than skipping it, because researchers can tell instantly.

Almost every abstract has the same five parts, even when it isn't labeled that way: why the question matters, what the gap is, what they did, what they found, and why it matters. Read it slowly with that structure in mind and most of the fog clears.

Do not stress about the jargon. Look up two or three terms, accept that you won't get the rest, and focus on the shape of the argument. Here's a process that takes about twenty minutes per paper.

1. 1Pick a recent paper. Go to the lab's publications page or PubMed and choose something from the last two or three years where the PI is the last author. Last author is usually the person running the lab.
2. 2Read the title and rewrite it in your own words in one sentence. If you can't, pick a different paper.
3. 3Read the abstract twice. First pass for the shape, second pass with a browser tab open to look up two terms you don't know. Two, not twenty.
4. 4Write down, in plain language, what question they were trying to answer and what they concluded. Two sentences maximum.
5. 5Read the figure captions. Figures carry most of the actual findings, and captions are usually more readable than the results section.
6. 6Read the last paragraph of the discussion. This is where authors say what they couldn't answer and what they want to do next. It is the richest source of honest, specific things to say.
7. 7Write your one sentence for the email. Reference the specific finding, method, or open question, and say what it made you think or want to know. If you can't do this honestly, that's real information: this lab may not be the right fit for you.

What a good cold email actually contains

The whole email should fit on a phone screen without scrolling much. Aim for 150 to 200 words. Every extra paragraph reduces your odds, because the person reading it is triaging their inbox between patients or between meetings, and a wall of text gets marked unread and forgotten.

There are six moving parts, in this order.

• Subject line. Specific and boring beats clever. "Undergrad interested in your work on macrophage metabolism" or "Shadowing request from a Rutgers student interested in EM." Never "Hello" or "Opportunity" or "Quick question."
• Who you are, in one line. Name, year, school, what you study. Nothing else yet.
• Why you're writing to THEM specifically. This is the make-or-break sentence. It's the paper detail, the shared school, the person who referred you, the clinic they run. If this sentence would work in an email to anyone else, rewrite it.
• What you're asking for, stated plainly. "I'd like to ask about joining your lab." "I'm asking whether I could shadow you for one day." "I'd like 15 minutes to ask about your path." Don't bury it. Don't be coy. Ask by the third paragraph at the latest.
• What you bring, briefly and truthfully. Coursework, a technique, a language you speak, a job that proves you show up on time for two years straight. Working thirty hours a week at a grocery store and holding a 3.6 is a genuinely impressive credential. Say it.
• An easy yes. Make the smallest possible version of your request the default. Offer specific windows of availability, offer to do it on Zoom, offer to take a five-minute call instead, and say you'll work around their schedule. "Would you be open to a 15-minute conversation, or could you point me to whoever in your lab works with undergrads?" gives them two ways to help you without much effort.

Three templates you can copy and paste

Replace everything in brackets. Do not send any of these without changing at least the specificity line, and never send one where a bracket survived into the final draft. Read each one out loud before sending; anything that sounds like a robot wrote it, cut.

TEMPLATE 1: To a professor, about joining a lab

Subject: Undergrad interested in your work on [TOPIC IN PLAIN WORDS]

Dear Dr. [LAST NAME],

My name is [YOUR NAME] and I'm a [YEAR, e.g. second-year] at [SCHOOL] studying [MAJOR]. I read your [YEAR] paper in [JOURNAL] on [ONE-CLAUSE DESCRIPTION OF THE FINDING], and the part that stayed with me was [ONE SPECIFIC DETAIL: a method, a surprising result, or an open question they raised in the discussion]. I'm writing to ask whether you have room in your lab for an undergraduate research assistant.

What I'd bring: [ONE OR TWO REAL THINGS, e.g. "I've finished organic chemistry and I'm comfortable in Python" or "I've worked 25 hours a week at the same job for two years and I've never missed a shift"]. I'm looking for something long-term rather than one semester. I could commit [NUMBER] hours a week, on [DAYS/TIMES], starting [DATE OR TERM].

Would you be open to a 15-minute conversation in the next few weeks? If your lab is full, I'd be grateful if you could point me toward anyone in the department who is taking students. I've attached a one-page resume in case it's useful.

Thank you for your time,
[YOUR NAME]
[SCHOOL EMAIL] | [PHONE]
[SCHOOL], Class of [YEAR]

TEMPLATE 2: To a physician, about shadowing

Subject: Shadowing request from a [SCHOOL] student interested in [SPECIALTY]

Dear Dr. [LAST NAME],

My name is [YOUR NAME] and I'm a [e.g. junior at Lincoln High School / sophomore at State University] here in [CITY]. I found you through [the [HOSPITAL] provider directory / my school's alumni list / [NAME], who suggested I reach out]. I'm writing because I want to understand what [SPECIALTY] actually looks like day to day before I commit years of my life to this path, and you practice it here.

I'm the first person in my family to consider medicine, so I don't have anyone at home I can ask about this. What I do have is [ONE OR TWO REAL THINGS: a job, a GPA in your science courses, volunteer hours, a language you speak with patients].

Would you be willing to let me shadow you for one day, or even half a day, at whatever time works for you? I'm available [DAYS AND TIMES, including early mornings and weekends if true], and I can complete any hospital paperwork, immunization records, background check, or HIPAA training your site requires ahead of time. If shadowing isn't possible where you work, I'd still be grateful for 15 minutes on the phone.

Thank you for reading this,
[YOUR NAME]
[SCHOOL EMAIL] | [PHONE]

TEMPLATE 3: Asking for an informational chat

Subject: 15 minutes? [SCHOOL] student asking about your path into [FIELD]

Dear [Dr./Mr./Ms./Mx.] [LAST NAME],

My name is [YOUR NAME] and I'm a [YEAR] at [SCHOOL]. I came across your profile through [our alumni directory / LinkedIn / the [ORGANIZATION] website], and what caught my attention was that you [SOMETHING SPECIFIC AND TRUE: started at a community college before your PhD, moved from bedside nursing into health policy, run the [PROGRAM] at [PLACE]].

I'm trying to figure out whether [FIELD OR PATH] is right for me, and I don't have anyone in my family or my school who has done it. I'm not asking for a job or a referral. I'd just like to ask three or four questions about how you got there and what you'd do differently.

Would you have 15 minutes for a call in the next few weeks? I'm free [TWO OR THREE SPECIFIC WINDOWS, e.g. "weekdays before 9am or after 6pm, and anytime Saturday"], and I'll work around whatever is easiest for you. If email is simpler, I'm happy to send the questions in writing instead.

Thank you,
[YOUR NAME]
[SCHOOL EMAIL] | [PHONE]
[SCHOOL], Class of [YEAR]

Your email address, signature, resume, and attachments

These details are small and they matter. They're the difference between reading as a serious student and reading as spam.

Use your school email address, always. A .edu or a school district address gets past spam filters that eat Gmail messages, and it signals immediately that you're a real student at a real institution. If you're at a high school without individual student emails, use a personal address that is firstname.lastname or firstinitiallastname at Gmail. Nothing else. Retire the address you made when you were twelve.

Put a three-line signature at the bottom of every email: your full name, your school and expected graduation year, and your phone number. That's all. No inspirational quote, no image, no long list of clubs.

• Attach one resume, one page, as a PDF. Never a Word doc, because formatting breaks. Name the file YourName_Resume.pdf, not resume_final_v3.pdf.
• Say the attachment exists in the body of the email, in one short clause. People don't scroll for attachments.
• Don't attach transcripts, writing samples, certificates, or letters unless asked. Extra attachments read as trying too hard and increase the chance the whole thing gets filtered.
• Your resume can include your job. It should include your job. Cashier, warehouse, home health aide, caring for siblings: these show reliability, and reliability is what labs are actually short on.
• If you have no research experience, don't apologize for it in the email. State what you do have. Nobody expects a first-year to have a publication.
• Send between 8am and 5pm on Tuesday, Wednesday, or Thursday if you can. Not because there's magic in it, but because a Friday evening email lands under a weekend of accumulated inbox.
• Proofread the person's name three times. Check the spelling, check whether it's Dr. and not Mr. or Ms., and check that you didn't leave the previous recipient's name in from your last email. This one error kills more cold emails than anything else on this page.

Following up: when, how many times, and exactly what to say

Most students never follow up, and a large share of the replies students do get come from a follow-up rather than the original email. The first email got buried under two hundred others during a grant deadline. That's not a rejection. It's an inbox.

The rule is simple: follow up twice, then stop. Reply to your own original email so the whole thread is visible, rather than starting a new message. Keep it to three or four sentences. Never express irritation, and never imply they were rude for not replying.

Follow-up one, seven to ten business days after the first email:

"Dear Dr. [LAST NAME],

I'm following up on the note below in case it got buried. I'm still very interested in [SPECIFIC ASK], and I'm happy to take whatever time you can spare, including a short call or just a reply here. If now isn't a good time or the lab is full, please just say so and I'll stop writing.

Thank you,
[YOUR NAME]"

Follow-up two, two to three weeks after that, and this is the last one:

"Dear Dr. [LAST NAME],

One last note from me. I'm still hoping to [ASK]. If there's someone else in your department or on your team who would be a better person for me to contact, I'd be grateful for a name, and that would be just as helpful.

Either way, thank you for your time, and I'll leave your inbox alone from here.

Best,
[YOUR NAME]"

After that, move on. Two follow-ups is persistence. Four is a problem, and academic and clinical departments are small worlds where a reputation for pestering travels. The correct move after two follow-ups is not to keep pushing on one person. It's to send five new emails to five new people.

Silence is not personal, and what to do when someone says yes

Set your expectations before you start, or the silence will convince you that you don't belong. Plan on most of your emails going unanswered. That is the normal experience for everyone who does this, including graduate students and faculty emailing each other. It is not a referendum on you.

Here's what's actually happening on the other end. The professor is on sabbatical. The lab has no funding for new people this year. The physician's hospital doesn't allow shadowing for liability reasons. The email went to a spam folder. They read it on their phone at 11pm, meant to reply, and never did. Almost none of the reasons have anything to do with your worth as a student.

So work in batches. Send eight to ten emails in a sitting, track them in a simple spreadsheet with columns for name, date sent, follow-up date, and outcome, and judge yourself on emails sent rather than replies received. Sending is the part you control. If you send thirty thoughtful, specific emails across a semester, something will land. Students who send three and conclude the door is closed are drawing a conclusion from a sample too small to mean anything.

When someone does say yes, your speed and specificity in the next 24 hours matter enormously. This is where students lose opportunities they already won.

1. 1Reply within 24 hours, ideally the same day. A four-day delay after they said yes signals that you weren't that interested.
2. 2Give concrete availability, not "I'm flexible." Flexible puts the scheduling work back on them. Write "I'm free Tuesday and Thursday after 2pm, Friday mornings, and any time on weekends" instead.
3. 3Confirm the logistics in writing: date, time, exact location including building and floor, what to wear, what to bring, and whether you need to check in at a front desk or security.
4. 4Ask early about paperwork. Hospitals often require immunization records, a TB test, a background check, HIPAA training, or a signed observer agreement, and these can take two to three weeks. Ask on day one so you don't lose the date.
5. 5Send a short calendar-confirming email 48 hours before: "Looking forward to Thursday at 7am at the 3rd floor nurses' station. Let me know if anything changes."
6. 6Show up fifteen minutes early, silence your phone, and bring a small notebook. Take notes by hand, not on a phone, which always reads as texting.
7. 7Send a thank-you email within 48 hours afterward that names one specific thing you learned or observed. This is what turns a one-day shadow into a mentor who remembers you next year.
8. 8Keep the relationship warm with a two-sentence update every few months. "I got into the biochem sequence and I'm still aiming at this. Thank you again for the day in clinic." People write letters for students they've heard from recently.

Mistakes that cost people the reply

• Mass BCC or a visible list of recipients. Instantly fatal. Everyone can tell, and it tells them you don't actually care where you land.
• Flattery with nothing under it. "I am deeply impressed by your groundbreaking research" says you read a webpage headline. "Your 2023 paper found X, and I've been wondering about Y ever since" says you read the paper.
• A wall of text. If your email is more than about 200 words, cut it. Your life story goes in a conversation, not an introduction.
• Asking for too much up front. "Can I join your lab, get a paid position, and have you write my letter of recommendation" in email one is too much. Ask for a fifteen-minute conversation and let it grow.
• Misspelling their name, using the wrong title, or calling a PhD "Mr." Check the department page for how they list themselves.
• Attaching five documents, or attaching a Word file that opens as gibberish on their phone.
• Apologizing for existing. "Sorry to bother you, I know you're incredibly busy, I'm probably not qualified." Cut every one of these sentences. Be polite and brief, not small.
• Being vague about the ask. If the reader has to guess what you want, the easiest response is no response.
• Sending from an unprofessional address, or from a personal account when you have a school one.
• Writing like a chatbot. Short sentences, real specifics, and one honest reason you're writing beat any amount of polish.
• Giving up after four emails. This is the most expensive mistake on the list, and it's the most common one.