I met a faculty (from a university in the UK) at a conference. She liked and praised my work and we are in touch after conference. She has an impressive profile and her research work is similar to mine. Is it okay to take a recommendation letter from her to seek postdoctoral position overseas? Will it help?

  • Comments on some answers seem to indicate this may vary by field. You may want to add that detail to your question. It's also recommended to wait at least 24 hours before accepting an answer as best answer, open questions encourage more answers to be posted.
    – Aubreal
    Commented Apr 22, 2019 at 16:25
  • 1
    What's this recommendation letter for? Applying to grad school or "something else"? Commented Apr 22, 2019 at 18:44
  • Its for a post-doctoral position.
    – LekhaS
    Commented Apr 23, 2019 at 6:38

5 Answers 5


It’s best for people applying for postdocs to have at least one letter from a well-known researcher in the field who is not at their school. It is quite typical that this person has only seen you talk once or twice and has read your work. Of course your advisor’s letter will have more detail and first-hand experience, but an outside perspective on the value of your research is a valuable addition. You should definitely ask for a letter from this professor!


There's nothing stopping you asking for one. However, you need to consider what would be written in such a letter.

In short, you can ask for a letter, but be careful how you use it. What could the person say on your behalf if a potential employer contacts them to get more information about their assessment of you?

For instance, it doesn't seem that you have a professional relationship - you haven't co-authored papers, nor have you worked together in the same institution, nor have you served on the same committees, etc. Any of these things is not necessary, of course, but these are the sort of things a referee would cite as evidence that they are in a position to recommend you or your work.

However, it sounds like your contact does know of your work. However, are they a world-leader in your field? Do they have the respect of your community? When they say good things about you or anyone else, do others pay attention?

How will a reader interpret a reference letter that is based only on a familiarity of your published work, and a single meeting at a conference?

  • 17
    How will a reader interpret a reference letter that is based only on a familiarity of your published work — At least in my field, the fact that someone knows and is willing to praise your work despite having no professional relationship makes the evaluation significantly more credible, and therefore more useful. For faculty hiring, letters from people with existing professional relationships are viewed with suspicion; for tenure and promotion cases, such letters are simply forbidden.
    – JeffE
    Commented Apr 22, 2019 at 9:25
  • 1
    Yes @jeffe that's fair enough, hence it is a question for the poster to try and determine the value of such a letter in context. One concern for example is the suitability of a candidate for a teaching or supervision role. It is hard to assess someone for this based on research outputs alone.
    – Nicholas
    Commented Apr 22, 2019 at 10:21
  • 4
    @JeffE For tenure and promotion the situation is fundamentally different; one would expect a tenure applicant to have developed a research program beyond what was done at the PhD level, and that the work of the candidate would have received enough accolades to be recognized by colleagues who are not current or recent collaborators. I’m not sure this higher threshold is realistic for postdocs. Asking people who do not know you well for reference letters is dangerous. Commented Apr 22, 2019 at 12:10
  • If the position requires teaching then they will ask for a letter that specifically addresses teaching. The research letters are not expected to address teaching. Commented Apr 22, 2019 at 12:25
  • 4
    @ZeroTheHero In my subfield (theoretical computer science) the best postdoc candidates have developed independent research reputations. Of course, that can't be "beyond the PhD level"—by definition!—but that's a red herring, as is "receiving enough accolades". The right benchmark is whether they have done enough high-quality, high-visibility, high-impact work to be recognized by independent, arms-length colleagues. The best postdoc candidates in my field most certainly have.
    – JeffE
    Commented Apr 22, 2019 at 22:02

You can definitely ask her for a letter of recommendation. It cannot hurt I assume.

However, considering that the person does not know you much, the letter might not have much weight in the eyes of whoever reads it.

Since there is not much to lose, ask for the letter. If you receive it, use it as you see fit. There is no guarantee whether it will help or will be useless.

  • 1
    Well, there is a small risk that it hurts your relationship with the acquaintance and that interviewers particular remember a let-down moment when they first see you have a recommendation from someone important but then see it's very superficial / based on low insight. I'd feel at least a bit weirded out if someone I spoke to once during a conference and maybe exchanged a mail or two asked for a reference letter. An application that includes a reference from someone famous which then turns out very weak content-wise, might also look like aiming at the stars without any substance. Commented Apr 22, 2019 at 19:07
  • 1
    So while I agree that, given a good CV otherwise, it can be an added benefit, I would see some risks as well. Also depends on the circumstances, if the position matches a lot with the discussions one had, it might come off more as, "look you want to fill a position in that area, which is exactly where this young person wants to go and has apparently some knowledge in, Research in that direction makes sense", then it's more a reference for the research proposal and less for the candidate. Commented Apr 22, 2019 at 19:08
  • @FrankHopkins: I totally agree, there is always some associated risk in any endeavor (including drinking some water ;) ). I thought I need not explain that to someone searching post-doc anything.
    – virolino
    Commented Apr 23, 2019 at 7:21

Most reference requests come, implicitly or explicitly, with the expectation that the referee will describe in what capacity and for how long she/he has known the candidate. As a result, references from people you’ve met recently or only have passing acquaintance usually do not carry much weight.

Ask yourself how you would start a reference letter if the situation were reversed.

Dear Selection Committee, I write supporting the application of [candidate]. We met once at [this conference] a few weeks ago and exchanged emails afterwards. He seems quite reasonable and full of good ideas but I’ve never seen him in a lab, I have no idea of his academic performance, cannot comment on how well he works in a team etc.

and how you would react receiving such a letter.

It might make a difference in a post-doc application if the position is offered by a colleague of that person, so that this person can supply a final supporting opinion to other strong reference letters.

The situation would become different if you were to have conversations with this person over an extended period of several months, were invited to her lab for some sort of visit or internship, started a project or co-applied for some grant, and generally develop your initial meeting into something much more substantive.

  • 2
    This is not really true of research letters in math, which aren’t mostly focused on how well they know the applicant but instead on how well they know the applicant’s work. Commented Apr 22, 2019 at 12:22
  • @NoahSnyder I will grant you that math is different, and in more than one way. Commented Apr 22, 2019 at 12:23
  • 3
    JeffE seems to be saying CS is similar. What field is your advice coming from? Have you been involved with postdoc hiring? Commented Apr 22, 2019 at 12:28
  • 2
    @NoahSnyder Yes of course and my field is physics, though I’ve interacted with chemists. Words from people who don’t know a candidate are cheap. Maybe I’m reading this differently than you but whenever I am asked for a reference or I see reference letter, how well the referee knows the candidate is crucial to provide context beyond the CV. Nobody wants to hire a bright but socially abrasive candidate. Granted former thesis directors may have a vested interest but someone you meet at a conference would be very suspicious, and indeed people would see this as desperation. Commented Apr 22, 2019 at 12:57
  • @NoahSnyder I can only claim that theoretical CS is similar. But I do claim that.
    – JeffE
    Commented Apr 22, 2019 at 22:05

All the answers here assumed that this recommendation letter is for graduate school and if someone doesn't know you very well, its weight maybe is less than someone that know you very well such as your adviser. But, I want to look at this situation from other views, which is not very familiar for all academic people. I believe if that professor will be convinced to sign off on a recommendation letter for you, it means you have a really good scientific reputation that resulted to that someone as an independent researcher endorse your scientific work in a recommendation letter and I think it has even more weight than a recommendation letter from a dependent person (e.g. your advisor). But be cautious cause this is not the generally accepted view in routine academia specially for applying to graduate school but it may have some advantages for other purposes. This other purposes could be convincing someone outside of academia to understand your research and be convinced that it has substantial merit. But again it depends pretty much on your special purpose from getting this recommendation letter, which is not specified explicitly in your question.

  • I believe if that professor will be convinced to sign off on a recommendation letter — If a professor merely "signed off" on a recommendation letter, that letter is almost certainly worthless. But if you can convince a professor to actually write a strong recommendation letter, you're in business.
    – JeffE
    Commented Apr 22, 2019 at 22:08
  • @JeffE Exactly! That's my main argument here. When I'm saying "singing off on a recommendation letter", I mean a strong recommendation letter otherwise you should put it in trash... Commented Apr 22, 2019 at 22:10
  • 1
    That's exactly opposite to my purpose... First of all, these days nobody will write a recommendation letter from scratch for someone else even your advisor. At least, it's rare I think, so people, if they accept to sign off on a letter, will tell you to prepare a draft letter and they will review it. The important point here is that if you prepared a draft for someone he or she sign it off without changing it otherwise I would not use it at all. Commented Apr 23, 2019 at 3:30
  • 2
    First of all, these days nobody will write a recommendation letter from scratch for someone else even your advisor.This is simply false. I write at least 20 recommendation letters every year, mostly for graduate admissions, but also for postdoc searches, faculty searches, tenure, promotions, and various awards/honors. And yes, I write them myself. This is a normal and expected part of my job. At least in US academia, any letter that smells like it was written or even read by the applicant immediately poisons the application.
    – JeffE
    Commented Apr 23, 2019 at 9:25
  • 1
    Only 20+ years of reading and evaluating graduate-school and faculty applications. Faculty writing recommendation letters themselves is not rare (in the US); it's completely normal, expected, common behavior (in the US). Applicant-written letters are relatively easy to detect, and even when they're not, they tend to be badly written, because people who don't regularly read recommendation letters don't know how to write good recommendation letters.
    – JeffE
    Commented Apr 24, 2019 at 6:14

You must log in to answer this question.

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