Here's my dilemma. I met a post-doc at a conference a while back, and told him about the project I was working on at the time, and he subsequently invited me to give a talk in a session at an upcoming (next week) conference. I accepted, since it was my first invitation to speak at a conference, and overall I was/am excited at this prospect, and at the prospect of collaborating with him in the future. However for various reasons (mostly time commitment on my part), this particular project - which is really a side-project to my main dissertation work - hasn't panned out like I hoped it would when I accepted the talk. Needless to say, I don't feel confident that my talk will make a good contribution to the session, since my results are very lack-luster and my expertise in other aspects isn't enough to make up for this. Since I'm a PhD student and this is a side project, I don't have old work to "fall back on" and my advisor is impartial.

So which is better: cancel the talk and apologize to the organizer profusely, perhaps suggesting a future collaboration, or suck it up and give a woefully under prepared talk, risking embarrassment to myself? I'm leaning towards cancelling, because I feel like a poor talk makes me look bad to everyone who listens, while cancelling only makes me look bad to the session organizer.

  • 18
    I suggest you distinguish between "preparing the talk" (that implies building presentation slides, and is something which can be done as late as while traveling to the conference) and "having content/findings to present" (this may or may not be critical, also depending on the conference and goal of the session). May 23, 2016 at 7:10
  • 3
    For what it's worth, if your talk has been scheduled and this schedule makes it way on the internet (where everything lives forever, except for links), you might have people asking about it in upcoming years. I had to cancel an August 2008 talk in late June 2008 (I wound up being in the middle of the 2008 Iowa floods and lost my apartment . . .), but even though this was nearly two months before the talk, the title is on the internet in some places and I've probably gotten at least 3 requests in the years since then for copies of slides or an associated manuscript (has not yet been written). May 23, 2016 at 14:08
  • 2
    I'm not familiar with your topic, but can't you discuss something like "possible findings?", "what triggered you to start looking into your idea?", "background to the problem" and since the results did not come as you would liked them to be, how about you present your problem, results and ask for "advice, suggestions"? or things that you might have overlooked! I'm presenting these options in case if a week is not long enough for the organizer to get a new presenter (maybe too late?).
    – The Guy
    May 23, 2016 at 17:05
  • 23
    There is a huge difference between a poor or unfinished talk, and a well-prepared talk about poor(?) or unfinished research. You should strive to never give the former -- the latter on the other hand can make for a great talk. Negative results are an important part of research.
    – user168715
    May 23, 2016 at 19:38
  • 1
    You've never seen talks that start by saying "more questions than answers"?
    – yo'
    May 25, 2016 at 6:53

6 Answers 6


When this happens to me, I just describe it as "work in progress" or a "research attempt" and present what I can with what I've got. What's wrong with that? That's how research works.

"...hasn't panned out like I hoped it would when I accepted the talk." You can say this during your talk, and explain why it didn't pan out, and ask the audience if they can help. Maybe someone in the audience (your next co-author perhaps?) knows how to proceed, and has the expertise you're lacking. If nobody can help, maybe that's an indication that there is an obstacle (which is important to know).

You've said you're a PhD student, without expertise in the side-topic. Feel free to mention it in your talk "I'm a PhD student in [some subject] at [some university] under [someone]."

You said a post-doc invited you to give a talk (which I expect is not in the sense of an "invited speaker"), who presumably knows you're an early PhD student, and presumably knows what early PhD student talks are sometimes like. My suspicion is that this question is more about inexperience and lack of confidence than an inability to give a reasonable talk due to the work being unfinished.

  • 30
    Also, negative results are also results. May 24, 2016 at 2:30
  • 8
    This is a great answer. We're increasingly getting to the point where everything (including science) needs some kind of positive, groundbreaking result in order to be considered worthwhile. This is a very, very bad path. May 24, 2016 at 18:50
  • 17
    This is pretty much exactly what I ended up doing. The talk went well enough - I was able to discuss the problem and show a few "preliminary" results. Overall, it was definitely worth the experience to figure out how to deal with my confidence and uncertainty issues! Always a learning experience, this PhD thing.
    – icurays1
    May 24, 2016 at 19:38
  • Also, depending on the conference: if you cancel your slot might be just vacant, meaning the stage will be empty for 15 minutes (to not jeopardize session hopping)
    – lalala
    Nov 7, 2019 at 18:03

Perhaps you could try option 3: send your colleague the slide deck of your presentation, or a detailed summary, and ask him whether he thinks it would be "good enough."

Obviously, you wouldn't just drop this on your colleagues lap; you would explain your situation, much like what you have explained here, and see what he thinks. I have no idea whether this is a good idea in your neck of the woods or not; however, I certainly wouldn't hesitate to ask for an opinion before canceling outright or not holding up my end of a perceived bargain.


I would encourage you to reconsider your premise. In many cases it is possible to give an excellent, polished talk on unfinished research. This would be appropriate and welcomed by the audiences at many (not all) conferences.

As others have suggested, I think it is worth contacting the session organizer. If you do proceed, I would spend most of your time talking about the problem you're working on, why it's interesting, and give an example or two. You could then spend a limited amount of time discussing the progress you've made.

Circumstances may vary, but when I've gone to conferences I've often enjoyed talks like this, and gotten a positive impression of the speaker.

  • 7
    This is exactly what I'm planning to do - I'm glad someone else thinks it's a good idea! After speaking with the session organizer, he thinks my problem and approach are interesting enough to talk about even in the pre-results stage.
    – icurays1
    May 23, 2016 at 19:01

I agree that you should cancel the talk, for pretty much the reasoning that you explained, except that I think you are overestimating the extent to which canceling the talk will make you look bad to the session organizer. And I don't think you need to apologize "profusely" (although an ordinary apology is certainly in order).

As you said, things simply didn't pan out the way you had anticipated - this is quite normal and understandable in cutting-edge research, and I'm sure the session organizer will not find this very unusual, will appreciate your setting high standards for yourself, and in any case would consider a canceled talk (which at least leaves him with a small chance to fill the slot with another last minute speaker) much preferable to a lousy talk by an unprepared speaker, which may also reflect poorly on his credibility as a session organizer.

Overall, a canceled talk is a fairly minor snafu of the sort that conference and session organizers have to deal with pretty frequently, so don't worry about it too much.

  • 5
    I'm going to talk with the session organizer tomorrow. If he thinks I have enough to proceed, I'll go ahead, otherwise I'll cancel it. Thanks for the response!
    – icurays1
    May 23, 2016 at 6:53
  • 9
    -1. I see no reason to cancel the talk. This is a first talk by a Ph.D. student! He could talk about pretty much anything as long as the talk itself is decent. I rather have a Ph.D. student give a good talk about unfinished research than---what is more common---an awful talk about his important results. May 24, 2016 at 1:41
  • 1
    +1 for pointing out canceling is usually minor, but -1 for agreeing with the OP's premise that weak results will mean a bad talk. May 24, 2016 at 17:35

My opinion is that you should give the talk. 1. You've made a commitment to the organizer. Keep it. 2. It will be great experience for you. 3. As others have pointed out, it's ok to let the audience know the research isn't finished or as you envisioned. Share that. They understand. 4. Be open to finding help and/or connections among the extraordinary resources that will be your audience. 5. It's normal to be scared before your first (or fiftieth) presentation. See 2. 6. You can do it.


While it is not impossible to give a beautiful talk on unfinished research, I personally feel acquiring such qualities is NOT a natures' gift. You need to have a lot of experience and in-depth understanding to give such a talk. If you DO have all such qualities, you should GO AHEAD. Otherwise, ending up with a sloppy presentation will let your audience under-estimate your potential.

You must log in to answer this question.

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