- You should probably give up on the idea of presenting data. In a typical scientific presentation, you're expected to show data slides--figures, graphs, equations, etc. The problem in an Ignite talk is that there just is not enough time. When I was making a rough draft of my slides, I wrote out the ideas that I wanted to convey on each slide. But once I started practicing, I realized that I had time for just two sentences per slide (and that was speaking at a faster than usual pace). In the end, I included just one slide containing a figure from a paper. I was able to explain it in two sentences, otherwise I would have had to scrap it. Any complex analysis is definitely out.
- Instead, use lots of photos, and minimal text. Once I stopped stressing about presenting data, I was able to focus on ideas instead, and how I could illustrate my ideas with pictures. I always feel lucky in that I work with a charismatic study species in the field, so it's not difficult to find captivating photos. Apart from my title slide, I included text as little as possible. I did make an exception for my 11th slide (halfway through), which was my only all-text slide. I used this slide to reinforce my main question, and to allow me to breathe (really, it's hard).
- Avoid animations. It's hard enough to practice 15 second timings. When you add in animations that start on the 3rd second, or the 10th second, I think it gets too complicated. Reduce the amount of thinking you have to do by having static slides.
- Practice much more than you would for a normal talk. In my experience, scientists are busy people, which sometimes means that presentations are being put together and practiced (or not) until the last minute. Part of this comes from knowing that even if you stumble a little in your talk, you have 15-20 minutes (for a normal talk), and you'll be able to find your place again, or rephrase something if it doesn't come out quite right. In an Ignite talk, there is no room for error. You can't afford the time it takes to repeat one sentence, or otherwise you'll be playing catch up with your next slides. So I practiced...a lot. It took me hours to get through the first half of my presentation successfully. For some reason, the second half came together much quicker (although maybe I was just exhausted at that point). Some of the online advice I found said to not get caught up in the exact phrasing for each slide, and to rather have a point that you want to get across. This didn't work for me--I practiced until I had the phrasing down pat each time. It got to the point where I think I could have given the script of my talk, without the slides behind me, in the correct 15 second intervals. I even ran through the whole thing in my head when I woke up at 3:30 am with presentation anxiety. The upside of all this practicing was that when I gave my talk for real, I was able to run on a combination of autopilot and adrenaline. The five minutes passed in a total blur.
One more thing that I think would help future presenters: (if people are willing to share) ESA should start videotaping Ignite talks and posting them to the Ignite website. When preparing, I would have liked seeing how other scientists approached their talks. Maybe others have found successful ways to incorporate data (perhaps by "hacking" the format, e.g. repeating a slide to give yourself more time, although ESA told us not to).