I'm an undergraduate student majoring in physics and mathematics, who is preparing to begin applying to graduate programs in physics. In doing so, I've reached out to a few people for advice, and I've received a lot of conflicting opinions on various topics. Among them is the importance of reaching out to potential PI's, and how much doing so potentially affects my admissions odds.

  1. Firstly, how and when should one reach out to a potential PI? I've "cold-called" professors plenty of times under other circumstances, however I'm not sure what the etiquette here is. Should I write multiple faculty members in the department? What and how much information should I provide about myself? Does it make sense to reach out to faculty at highly-competitive "reach" schools (I'm planning to apply to Columbia, for instance, and I imagine they receive tons of emails, so it seems a bit silly to "cold-call" random Ivy-league faculty)?

  2. Secondly, how much does it affect my odds of admission? I've read and been told that communicating with a faculty member can effectively guarantee admission, if you can "sell yourself" such that they are interested in taking you on as (and presumably funding you as) a student. Is this true? Does not reaching out to potential PI's put me at a significant disadvantage?

I'm not sure what the etiquette is here, nor how much effort I should be putting into this.

  • It depends on your subfield of physics. (Theory or experiment? If experiment, big collaborations or small?) Ask grad students who are in your subfield.
    – knzhou
    Jul 15, 2023 at 2:56
  • Are you applying to masters or PhD programs, and are you in the US (which I believe you are based on your flags)? Part of the conflicting information is that in most places besides the US you apply to a PhD program after a masters and apply to work directly with someone, then you look at getting a position at the university (which may automatically come with the PhD funding if you applied to an open position). But when applying for a masters you don't need to reach out to someone and you're probably just going to go through some admission process where PI's aren't so important. Jul 15, 2023 at 18:50
  • In the US the PhD and masters is combined, so its often like applying to masters elsewhere in the world. However (to my understanding, not having proper experience with US academia), different universities have different admission processes. Some systematic where you need to hit certain characteristics and then find a supervisor during the first couple of years, so PIs don't tend to have as much influence. Others I believe have the PIs eyeing certain candidates during the admission process and there a PIs word may carry more weight. Jul 15, 2023 at 18:54
  • @NAMcMahon Correct, I'm applying to PhD programs in the US. Jul 16, 2023 at 0:18
  • @knzhou Thanks for the advice. I'm interested in doing research in theory. Unfortunately, it's been a bit difficult to find grad students to speak with as I'm attending an undergraduate-only institution. Jul 16, 2023 at 0:18

3 Answers 3


I get a dozen such emails per month (and this isn't a lot). In most case the sender has clearly not done their homework and know nothing about my research program: I ignore those.

In rare cases (< 5% of the time), the sender has actually cared (at least pretend correctly) to have read some of my papers, in which case I will usually reply to the email.

On balance, reaching out will make limited difference. If your application is not competitive, nothing can save you. If your application is borderline, then a good word from a prospective thesis director will never hurt: they can single out an application and say they would be interested in supervising X is the file is acceptable and the resources are there. This in no way replaces the formal requirement to apply: I know of no system where supervisors directly accept students to a program. Also, even if you contact someone, this person may choose not to get involved in the admission process and let chips fall where they fall. In other words, emailing ahead of time is a low-return proposition, but may on occasion make an enormous difference.

Finally, there are plenty of other posts on this site on how to write a proper email when "cold-calling". You should never email "random" faculty: do your homework and be sure that your email conveys your professional interest in the specifics of the research program. No lollypop statements like "Since I was 5 I always wanted to go to Columbia" or "I would be the honour of my life to work under your supervision": these are immediately flagged as false compliments and leave a bad impression.


Yes, contacting a prospective PI or several PIs is definitely a good idea, and this may critically affect your odds of admission -- provided you do it correctly.

To understand what "correctly" really means, put yourseld in the shoes of a professor looking for graduate students. The professor, of course, wants the candidates to be hard-working, creative, intellectually brilliant -- that's obvious. Above that, the professor always wants candidates to show initiative and to know exactly what they want. Strong would be a candidate who already has interest in and some knowldege of the area in which they would like to work under the tutelage of this professor.

For example, a scholar working on gravitational-waves detectors would be pleased to receive from an applicant an e-mail stating that (s)he has read this scholar's recent works on a certain type of detector, and would like to join a groups designing these devices. Such an e-mail will serve the purpose of demonstrating that the applicant is mature and has made a well-informed choice.

I am aware of several cases when making such a contact played a decisive role in the admissions process.


I wouldn't do that. I would just apply and express interest in who I wanna work with. Once you're accepted, then you'll decide who you'll work with. But, I'm not in physics (thankfully!!!!!!!!!), so perhaps it may be different.

In my field of public policy for example, I can reach out to anyone anywhere, since our research isn't tied to a lab, but in the harder sciences, who you work with is literally based off where you're located, as far as I understand.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .