Presentation Poll: Outline or No Outline?

We have a summer student seminar series in the science and engineering departments here, running two days a week at lunchtime with three students each day giving 15 minute presentations on their summer research projects to other students and faculty.

The student talks are split almost 50/50 overall on whether to provide an outline at the start of the talk or not. About half of the students put up a slide listing the component parts of their talk (“First, I’ll give some motivation for the experiment, then I’ll talk about the apparatus, then…”), and about half jump right into the talk, identifying the sections as they go (“The motivation for this experiment comes from…”). I have my own opinion on this question, but I thought I’d throw this out as an audience poll to see what people think:

Seminar series are most definitely classical environments, so please choose one and only one option.

24 thoughts on “Presentation Poll: Outline or No Outline?

  1. Marginally less pointless than reading out the title of your talk, yourself, after whoever introduces you has already done that 15 seconds ago.

  2. Not only do I use an outline at the beginning, I place the headings from the outline along the bottom of every slide and have a sort of progress bar that fills up as I get to each new topic. I saw someone do that years ago and it was a huge help to me in following their talk, so I incorporated it into my style.

  3. I think that your poll is highly biased, since it gives no good option to support having an outline. After having given a lot of talks, I’ve learned it’s often best to give an outline because people watching your talk have short attention spans, and ultimately will only want to pay attention to the part they are interested in. Giving an outline allows them to know whether or not they want to pay attention. Probably this is less important if fewer people are presenting than if you are used to giving presentations in front of hundreds of people(as is the case in high energy collaborations).

  4. It depends on the outline…

    This generic one is a waste:

    * Motivation
    * Methods
    * Apparatus
    * Results
    * Conclusion
    * Questions

    If you can make the outline into something worth reading or remembering, then they are worthwhile

    * Malaria kills millions
    * Do Fans repel Mosquitoes?
    * Fans are relatively effective repellents

  5. If you have 15 min to fill, and 4 slides, including title and acknowledgments, an outline is a god send. And for sure better than a bloody mission statement.

  6. I strongly dislike most of the talk outlines I’ve seen. I’m willing to believe that there exist good outlines, as Dave X suggests, but these are probably better called overviews. An outline, if its like an outline of a paper, likely includes every section and gives too much trivial detail. We absolutely don’t need to hear that you’ll “give an introduction” in an introduction section and that you’ll have conclusions in a conclusion section or that you’ll describe your apparatus in the experiment section. The general format of a physics talk is so well established that I hardly think we need to be told about it again, especially in a short, time-constrained format. The considerations are similar to the format of a PRL, where there is a space constraint, and where there are no section headings, but where the flow of the paper still follows a rather standard order.

    An outline might be useful if a talk deviates significantly from the standard outline. Otherwise, an overview that touches on the specifics of the “story” of how the whole project came together–that a surprising result from an earlier project led to another project for which you had to design a new apparatus that let you measure something better than anyone had done before–could work. But in most cases, the introductory part of the talk should motivate and naturally lead into the rest.

  7. Outline slides are almost always useless. The ones that work are the ones that don’t sound like outlines, where the introduction/motivation serves as a setup for what will be covered before getting to the conclusion.

    If you need a progress bar to follow the talk, it means you’re just waiting it out to the end either because you’re not actually interested, you’ve already been lost, or it’s just not a good talk.

  8. The Internet has not shortened my attention span, but it has tried my patience. If somebody doesn’t get to the point quickly, I quit. An outline of a fifteen minute talk would suit me fine. I can think about your points before you get to them. And the more thinking, the better, right?

  9. Always make the outline, even if you end up not using it. Otherwise, the risk of your talk turning into an incoherent mess becomes too great.

    One of the serious drawbacks of PowerPoint, compared to viewgraphs (which was the standard technology back in my grad school days), is that PowerPoint slides are an inexhaustible resource. It’s way too easy to make more slides than you need and end up babbling. (Senior tenured professors do this as often as newbie grad students.) Viewgraphs were a scarce resource, so when that was what I used I always made a list ahead of time of what would be on each slide, so that I knew what my talk would cover before I made the first slide (and as a bonus it prevented me from making too many slides for the time allotted). There is a reason movies and TV shows are storyboarded before the first scene is shot–heed that piece of wisdom from people whose job it is to make entertaining presentations.

  10. I wouldn’t object to putting one in, but agree with the “don’t use generic waypoint names” comment above. I would suggest absolutely NOT reading it at all. There really isn’t any reason to have it in there, especially if everyone is more or less following the same format, and I would expect the better presentations to not have it.

  11. I went with “pointless”, but only because “optional” wasn’t an alternative. 15 minutes is rather exactly at the break point between a talk so short that no outline is needed and a talk long enough that the outline and the weight given to different sections may be non-trivial. For longer talks there ought always to be an informative (i.e. non-trivial in the sense discussed above) outline, that should indicate what is going to be talked about at what depth and length. This is more important if there are several topics to be covered.

    Also I am of the firm opinion that talks should never exceed 45 minutes, which is roughly the attention span of most listeners.

  12. Outline or motivation slides in talks of any length are generally useless, I find.

    What I do find useful is to put the Conclusions first. That way the people who doze off after ten minutes still get the key take-home message, and they provide useful motivation for the rest of the talk, in that there’s a “how we get there” structure. You don’t need to structure your talks the way you actually did your research, complete with dead ends, after all. Think of the first slide as being an abstract for the talk — you wouldn’t leave your critical result out of the abstract, after all.

  13. I still shudder at the IEEE talk I gave that required an outline for a fifteen minute talk. I considered just saying, “And this next slide was mandated by the conference organizers. Moving on…” but eventually figured out a way to be can useful by ditching all of the standard motivation, etc. type headings. Still, the minute things that conference obsessed over was infuriating.

  14. Outline. So people in the audience know whether theyshould try to keep awake or they can sleep. Three fifteen minutes talk at lunchtime? On summer projects? Naptime!

  15. Handouts are a detractor and should only be used if the presentation is long and follows a very complex topic. The best scenario is to have all the attention on the speaker and avoid the rustling of papers.

  16. Your students should look at that 3-minute talk as both a model for economy of time and to critique its effectiveness at making a particular point.

    IMO, the talk needs an outline but you don’t have to show that slide to the audience other than to have it up there in the slack time while the speaker is being introduced. The speaker needs to have a clear point, and a clear plan about how to get to that point. Having a slide before slide 1 is a good way to remember that structure.

    The first slide needs to present that point (the conclusion @14 within an overview @8), perhaps with both the question behind the research and the answer they got. “Tell them what you are going to say, say it, tell them what you said.” After that, each slide needs a purpose. Remind the student that if they put the purpose on the slide to show its role in the overall outline/argument, they don’t have to read that part. Everyone will have read it before they get their finger off the clicker.

    But the most important thing is to explain it to the other students, not the faculty. KISS. If the students get it, so will the faculty. Everyone enjoys a talk where they know what is going on.

  17. I don’t think the length of time is the deciding factor; the number of slides is. Those will be correlated, obviously, but the fact remains.

    That said, I always make them, even if I don’t include them in the finished product. It’s useful to organize my own thoughts and keep the presentation from getting away from me. (I also tend to do what a previous poster does, which is incorporate the outline phrases in the slide titles.)

    Having made them, I tend to put them in. It is not a mistake to do so. The common mistakes are:

    A) Including a generic one
    B) Including one too detailed
    C) Dwelling on it

    Even if you’re giving a three hour presentation with a hundred and fifty slides, and your initial outline goes in five levels and spans five slides itself, don’t put that in. Prune it to the highest level. And once you’ve pruned it, don’t dwell on it. Don’t read it. No one wants to hear you read, effectively, the table of contents. Blaze through that slide in fifteen seconds.

    Even for academic-type talks, I do this. For engineering-type talks (which are liable to be design reviews, archived for legal posterity to prove to ourselves, our customers, and our auditors that we did our jobs) it gets vital: Those can run for hours and top a hundred slides, easily.

    This will make your academic skin crawl: If I’m giving a hundred-fifty slide presentation with a five section outline… I’ll even put in transition slides with no other text than the title of the new section as we begin it.

  18. I think that outlines can be very useful (if well done). While they are not always acutely needed
    by senior people who are familiar with the narrow research area or the specific project, they are extremely helpful for younger researchers who may not already know everything.

  19. Novak@19’s last remark does not make my academic skin crawl. Students want to see the same thing in lecture: a clear indication of where you are going next.

    The best colloquium I ever saw in my life (Al Bartlet from Colorado) started with out looking scary: there was a stack of transparencies at least two inches high. That day I learned that the number of slides is less important than how much is on them. Some had only one word, the punch line answer to a rhetorical question. It flowed.

    And that old-school technology (in competent hands) was superior in many ways to Bloatware. He put up two slides in less time than it takes to do those cute fade in tricks that attempt to emulate the same thing, quicker even than some computers can pull up the next screen.

    BTW, there were no slides filled edge to edge with words (see Chad’s comment today about the string theory talk), none with multiple bullet points. Each point got its own page. Clean and neat, which also made it easy to take notes on key points.

  20. I want to second (or third) Novak@19’s transition slides. If you’re changing topics in a talk it’s nice to give a clear visual indication to go with your verbal indication, lest the unobservant be left even more confused. It’s worth the few seconds it takes to put up a two-word slide to accompany your verbal introduction of the new subject. In a typical 50-minute talk I usually have 4 or 5 of those.

    Similarly, an un-belabored outline (a la Dave X @6) isn’t a bad thing either, although I would personally skip it if giving an APS-style 12 minute talk.

  21. CCP and AC:

    In the presentations I often give, the transition slides are useful in the sense of giving people one last chance to ask relevant questions before we move in. So the slide is really saying, “Okay, we’re leving the circuit design part of this, and moving on to the mechanical design part. Questions? Questions? Bueller? No? Okay, mechanical stuff.”

    It’s not that people don’t know where we’re going; but it’s very helpful to let people know when we are actually making a transition.

    My point through all this is that the talks a faculty member is used to giving are not the same as the talks a practicing engineer will be called on to give. Yeah, sometimes I do need the equivalent of a 15-minute presentation that has to not make a VP’s head explode. But more often, I’m giving, or am part of, a multi-hour thing.

    I’m guessing that industry physicists, or maybe even physicists who work on the semi-academic but very large scale programs (ITER, say, or some particle accelerator thing) will be in the same position. If they’re working for an engineering company– as a systems engineer, for instance– they’ll be directly in my situation.

  22. Usually I’d say it’s a waste of time. None of important speeches in the history (by important I mean very strong, influential speeches that changed the history) had an outline. Yet in some of technical presentations it makes sense.

Comments are closed.