This semester, I decided to replace the final exam in one of my upper-level graduate courses with a short 15 minute presentation on a scientific paper related to the course topic. To give the students some guidance, I provided a list of my tips for given short conference presentations, and I thought: why not share it here? Then at least I will have an easy place to point students in the future. At this point, I have about 30 years of experience giving presentations at conferences and elsewhere, so hopefully this information will turn out to be useful.

If you’re not interested in tips for scientific presentations, you can safely ignore this post. 🙂

Guiding principle #1: A 15-minute conference presentation is not intended to give detailed information, it is more of an advertisement. You want to provide enough information for the audience to understand “Why was this work done?” (background), “How was it done?” (methods), “What did we learn?” (results). Your goal is to convince the audience that they should go read your printed publication on the subject, where they can get all the details.

Guiding principle #2: You should structure your presentation like you are telling a story. There should be a narrative flow throughout the presentation that leads the audience naturally from slide to slide. One way I’ve put it in the past is that every slide should ask a question that the next slide answers. Even many professional scientists will give a talk as a list of isolated results with no strong connection between them. Constructing the talk with the “ask a question/answer a question” approach to building your slide deck will give the slides a natural flow and avoid awkward pauses where you look at the next slide and try to remember why you put it there at all.

Beyond those guiding principles, here’s a list of random tips to keep in mind:

1. Each slide should have one major point that the slide is built around, probably mentioned in the title of the slide. Make slide titles descriptive, not just something like “method” or “results”! Obviously, you can talk about more than just one thing on a slide, but it should be built around that one central idea.
2. My personal guideline is to spend one minute per slide. This means, allowing for a couple of questions, you should have no more than around 12 slides, leaving 3 minutes for questions (or a few extra minutes in case you end up taking more time than you expect to get through the slides).
3. Don’t bother with a list-like “overview” slide to explain the structure of your talk. Again, keep in mind the idea of “telling a story”: if you set up your talk in an organized way, you shouldn’t have to list in advance what you’re talking about. (Think of tables of contents in books — in fiction, it is pretty rare to have an explicit table of contents these days, and even rarer for someone to read the entire table before reading the book.)
4. To tell the story, think of using about 3-4 slides to describe the background of the work, 3-4 slides to explain the methods used in the work, and 3-4 to describe the results of the work. Obviously, the number of slides used in each category can vary depending on the talk. (I don’t usually include the title slide in my slide count, unless for some strange reason I plan to talk a lot while staying on the title slide.)
5. Keep mathematics to a minimum, only including equations that are necessary to the method and results. Avoid including lengthy mathematical derivations of results unless those derivations are the whole point of the talk. If you do describe derivations, keep the descriptions brief and only hit on the major steps taken.
6. Be sure to include on the slide, in print, any important points you want to make. It is tempting to keep a separate set of notes on the side to refer to when giving the talk, but anything you definitely want the audience to see and hear should be listed explicitly on screen. (Keep in mind that audience members will have their minds wander while you present and might miss that very important point that you only spoke aloud. Yes I’m talking about me.)
7. Speaking of words on the slide, find a “Goldilocks” zone where you keep text short and to the point but not so short that it is vague. Font size for me is usually 20-28, with the high end preferred but the low end used occasionally when more information must be provided. (References on slides can be in smaller font, like 14 points.)
8. Speaking of references: be sure to include references for any external work you talk about. Also, make sure there is a clear distinction in the presentation between what other people have done in the past and what you are doing.
9. Be flexible when speaking, and think ahead about what you could “cut” from your presentation on the fly. Sometimes one ends up spending more time than planned in the early part of the talk, resulting in being short on time for the latter part. My former postdoc advisor once said “there is a fine line between a scientific presentation and a hostage situation,” and you don’t want to be the cause of the latter. I usually identify one or two slides that have interesting results but are not essential for a key understanding of the work (This is also important in case someone decides to ask questions in the middle of the talk, which usually is frowned upon but people do it anyway.)
10. If you’re not used to giving presentations, do at least one out-loud practice talk! It could just be for yourself, and doesn’t need an audience. This is important because the words we have in our head often aren’t the words that are going to come out when we’ve got the pressure of an audience.
11. Explain any acronyms or jargon if they aren’t reeeeeeally familiar to the audience. Also, avoid using too many acronyms or jargon, because it is easy for the audience to get lost when buried beneath a bunch of random letters.
12. If you’re super nervous about giving the presentation, keep in mind that it’s only fifteen minutes! Even in the worst case scenario where an audience member decides to start arguing with you on slide one, which almost certainly isn’t going to happen, that’s only fifteen minutes of your life.
13. For short conference presentations, I view “conclusions” slides very much like I do “overview” slides: if you’ve told a good story to the audience, you shouldn’t have to go back and review it literally minutes later. A “conclusions” slide can be replaced with a slide describing future research or the implications of the current research to the state of knowledge of the field.
14. Finally: don’t be afraid to say “I don’t know” to questions. It is impossible to be prepared for any possible question an audience member might ask, and it’s okay to not have the answers to everything. If you don’t know, just let the audience know and don’t try to bluff your way out of it. The audience is there to learn something, and if you don’t know the answer, you don’t want to mislead by pretending you do.

These are just some random suggestions, and I’ll probably have more to add to the list in the future!