A long guide to giving a short academic talk (2022)

To this list, I'd add one more thing: pay attention to talks that you see, writing notes not just on the material, but on things you liked and disliked about the presentation style, the slide format, etc. Avoid the things you dislike, and employ the things you like.

This applies to all manner of things.

You may notice that speakers who talk to the screen are unengaging. So avoid doing that.

You may notice that strong speakers have a moment near the beginning when they grab the attention of the audience, and that this makes it easy to stay awake for the whole talk. So try to find a way to do that -- a way that fits with your own work, of course.

You will certainly notice that talks that go overtime make for a terrible experience for everybody. So don't do that. (You'll need strategies. The simplest is to have "landing points" in the talk, so that if you see that time is running short, you can skip some material to get to a landing point. Obviously, the conclusions slide ought to be a landing point.)

And so on. Learn from the impression that other speakers make on you.

>You will certainly notice that talks that go overtime make for a terrible experience for everybody. So don't do that. (You'll need strategies. The simplest is to have "landing points" in the talk, so that if you see that time is running short, you can skip some material to get to a landing point. Obviously, the conclusions slide ought to be a landing point.)

I won't say I've never gone a few minutes over when there's a break coming up and the audience is engaging. But, yeah in general, it's especially rude when the next speaker is waiting to get setup, and that time they're left with 1 minute to plug in will be the time something goes wrong.

You should have a decent sense for pacing of the talk anyway but, as you say, you should also know which material lends itself naturally to "in the interests of time, I'm not going to spend a lot of time on this topic today but we can talk afterwards if you're interested sort of thing."

Conference talks are also different lengths and they've tended towards shorter over time. So, even if you can adjust on the fly to some degree, consider not using the same deck for a 30 minute talk and a 50 minute one. (I'm actually a fan of talks that are on the shorter side in general.)

I have mixed feelings about Q&A in the session itself in general for a lot of reasons. But you need to keep things from getting too diverted or getting down the "more a comment than a question" path. (Which is part of the mixed feelings--it's hard to cut off without seeming a bit rude.)

I've used this approach for slides and websites. Any time I see a great design, I'll take a screenshot and save it in a folder called "PowerPoint/Web Ideas". I'll even include some shots from typography style music videos. Then whenever I have some free time and am bored, I'll try to recreate it in PowerPoint.

Most of the designs are TOO MUCH for teaching slides, but its fun practice since making slides is sort of my job. But the practice gives me a better understanding of what I can do with presentations, building templates, messing with color theory, etc. There's a science/art to making a good lecture presentation and I'm enjoying the artistic expression, while still being educational task.

also part of paying attention to other talks...I always make a reference to one or two talks that just happened or one that I know will be coming up.

It's one more point of context for the audience, and people always love it when others talk about them.

Unfortunately in grad school I noticed a lot of cargoculting bad presentations. People would copy people that were good scientists but bad presenters because “that’s what good scientists are doing”.

Related: there’s a worry that if you make the presentation too easy to understand, people won’t think you’re smart. Inscrutable presentations must mean that the presnter is a genius operating on a higher level /s