Learning to Give Better Talks: One Message per Slide. One Slide per Message.

How to give effective oral presentations | Physicist/Feminist

Cover of Trees, Maps, and TheoremsThere are some fundamental principles of giving effective talks that everyone should learn and try to implement to the best of their ability. Not everybody is going to give great talks, but all of us can give better ones! Towards that end, one of the professional development workshops I organized at the AAS meeting was specifically on giving better presentations. Dr. Jean-Luc Doumont gave a superb 1.5 hour lecture/demonstration on public speaking techniques for scientists. Lucky for us, Vivienne Baldassare summarized the main points of his lecture. You can also buy Dr. Doumont’s book, Trees, Maps, and Theorems (review forthcoming).

Vivienne’s summary of the lecture, while easy to skim, is actually quite dense and includes a ton of important points that are easy to think you understand, but can be quite difficult to figure out how to actually implement. Watching Dr. Doumont demonstrate each technique with his slides, voice, and body was extremely insightful. I am optimistic about bringing Dr. Doumont to future AAS meetings and I hope that many, many more people attend this lecture!

I think if there is one thing that would revolutionize the way many people give talks is learning and embracing the concept of “One message per slide. One slide per message.” Don’t make the audience struggle to figure out what the point of this part of your talk is. Write the message on the slide in one concise sentence. Describe that message verbally. Develop the message visually. Again, easy to say, but not always intuitive to implement.

I think a first step in this direction could be to change our relationship with slide titles. Slide titles should not be what’s on the slide, but the message of the slide. Not, The Distribution of Stars, but, M Dwarfs are the Most Common Type of Star. Not, Observations, but, 100 New High-Resolution Spectra Obtained. This idea is illustrated and described in more detail in the Content Slide section of Dr. Doumont’s
Checklist for Slides.

I have to admit, I was really disappointed with the vast majority of 5 and 10 minute talks I saw at the AAS. How long will it take us as a community to learn how to give better talks? I sure hope it’s in my lifetime, ’cause I don’t know how many more cluttered, text-filled slides I can take!

9 comments… add one
  • Tom Jan 21, 2011 @ 10:53

    I’m in the process of preparing a presentation, and to try and respect the advice of 1 slide = 1 idea, I’m creating a blank presentation with only one sentence per slide, nothing else (except ideas about what to show in the presenter notes), and my plan is to then write the real presentation only once I have managed to write a talk that makes sense with only this one sentence per slide. It’s the first time I try this, but I feel it’s been very helpful so far.

  • DS Jan 22, 2011 @ 17:47

    One reason this shift is slow is the audience has to change too, not just the presenter. I read Duarte Design’s slide:ology in grad school, which I found useful, but if I actually followed their prescriptions, I’m pretty sure a lot of senior scientists in the audience would completely dismiss me. (To be fair, their advice isn’t always 100% applicable to scientific presentations, but still.)

  • P Jan 22, 2011 @ 19:08

    One of the problems is that so many people are accustomed to massive amounts of text/figures on every slide that when they see “nice clean / 1 idea = 1 slide” they tend to dismiss the talk as being too simple/easy. The “pack every slide full of stuff” strategy is horrible to learn from, but it gives the impression of a very complicated and deep talk.

  • Kelle Jan 24, 2011 @ 11:29

    @DS & P: I disagree completely. I’ve been thinking about this for a couple days and I can’t think of one line of (rational) reasoning to support this point of view. To be clear, I’m not necessarily talking about changing the actual content of the talk. I’m not advocating “dumbing it down,” it’s more about changing the presentation and organization to make it easier to follow.

    Throughout, when I say, “busy” I basically mean the same thing as “detailed” and “very complicated and deep.”

    • First and foremost, if you think you need busy slides to impress your audience, you’ve got bigger problems. Your goal should be to impress your audience with your results and knowledge of the subject. This goal can be—and should be—achieved without barraging them with loads of text and illegible figures.
    • I have *never* heard a busy talk get compliments from senior audience members. No one ever says, “Wow, what a great talk! I didn’t follow any of it, but that guy really knows his stuff!” No, people just say (well, imply), “That was awful and a waste of time and I wish I could have just left.” If people have anecdotal evidence to the contrary, please speak up. Further, I think the more “senior” scientists have even more trouble with small text and more than one figure per slide purely since it’s harder to *see* what’s going on.
    • Rather than impress senior scientists, I think that busy talks more often than not, have exactly the opposite effect. Senior scientists have seen lots of talks and they can see through screens of details to find the results (or lack thereof) that you have effectively hidden. Busy talks reflect that the speaker doesn’t really know what their main points are or how their message fits into the bigger picture (and for some us, this may actually be true!). But don’t underestimate people…they are impressed with results, not with the false illusion of complication and depth.
    • I’m not a psychologist (obviously), but there’s a lot of it at play here. I think this reaction stems from people’s FEAR of being perceived as not smart. But I truly think it’s just that…a fear. And this method of reacting to that fear (by filling up slides with content) is not actually to your benefit. I have no idea how to best convince people of this but can only hope that as people gain self-confidence, they will realize that they don’t need cluttered slides to impress their audience. And in fact, could really impress them with a clean and clear presentation.
  • Maryam Jan 24, 2011 @ 12:09

    @Kelle, glad you got Jean-Luc Doumont to give his presentation – he’s excellent! He gave a presentation at UC Berkeley a while back (and I think I had mentioned him to you at some point) and he really changed my fundamental views of giving talks. His book is also very informative, and I’m always trying to remember his many pieces of advice.

    @DS & P: do you have evidence for what you assume senior colleagues think? What you are voicing is I think a common misperception, especially by more junior people.

    I agree with Kelle’s statement — only because who have tons of stuff on your slides doesn’t mean senior colleagues will think you are smart. I usually hear compliments about how elegant or insightful a talk was, not how many technical details or figures/text were covered.

  • Saavik Jan 25, 2011 @ 11:36

    I didn’t attend the AAS so couldn’t make the workshop, but I’ll say that Meg Urry used to do a fantastic ‘How to Give a Talk’ presentation (with a two page handout I still occasionally review), specifically geared at women astronomers (she was then at ST and I was at JHU). She was already pretty senior at the time and emphasized that simple and direct was much better than cluttered and overwhelming with detail.

    She also addressed some of the confidence issues common to women scientists (and I suspect others too). You might consider her for next year. Among the important lessons I learned from her, besides simplicity: ‘fake it til you make it’. Smile – force it if you have to. Look your audience confidently in the eye (practice with your friends beforehand, especially if you’re junior). Never apologize for something you’re putting up – if it requires an apology, it should be fixed or it shouldn’t be in your talk (this one drives me crazy).

    Unfortunately, I think the only way we’re going to improve at this is to practice practice practice in an environment that provides extensive feedback. It might be good to add oral presentations explicitly to the grad school curriculum… except that there are plenty of offenders in our senior ranks, who would presumably be teaching the course.

  • Benny Jan 29, 2011 @ 21:31

    Just to add on the “seniority” issue – I’ve seen “busy” presentations and heard “busy” speeches from senior people too, while other senior people showed great examples of “one slide – one idea” talk. My own mentor is a fine example – one figure, with HUGE labels, and more rarely 2 equally-sized figures, but only when a comparison is needed (i.e. same axes, 2 models or 2 populations …). His talks are always super-clear.
    The real point, of course, is that if you’re asked to give more details, you should be able to easily answer and thus “show off” your depth [maybe by “extra slides” with shortcuts?]. This way it’s clear that the simple/clean presentation is such by choice, not due to lack of knowledge.

    I particularly agree on the “not seeing the big picture” issue. As an observer, I feel that observational talks can be always filled-up with details, without addressing the more crucial scientific issues.
    This practically means loosing ~half the audience, assuming a ~”balanced” meeting…


  • Mordecai-Mark Mac Low Feb 9, 2011 @ 16:55

    As a senior person (tenured department chair) I can assure you that I don’t take simple slides to be evidence of a simple mind. Rather the opposite: simple slides that cover the topic well are much more difficult and impressive. Simple slides that miss the point are of course a problem… but not one that will be solved by throwing more material on to the slides!

    I have been intrigued by the “Assertion-Evidence” design (http://www.writing.engr.psu.edu/slides.html) for slides (as opposed to “Topic-Subtopic”) but have not yet tried to apply it to my own talks. Any comments from people who have?

Leave a Reply

Your email address will not be published. Required fields are marked *