How to give a talk

Although meetings and conferences are about far more than just giving public talks, they are of course a central theme. For the scientist, it’s a chance to present his information to his peers to disseminate his ideas and research, and of course trigger feedback and new research. For the audience, it’s an opportunity to gain access to information and ideas perhaps years ahead of their formal publication, (and some will never be published) and to get access to people and branches of research far outside what they can get in their own institutions of even countries or continents if it’s a big meeting. It seems odd then that I had had to sit through a great many talks that were obtuse, dull, confusing or apparently pointless. It seems a simple enough skill (especially in a profession at the core of which is communication and for which lectures and teaching form such a prominent position) yet is clearly one to be valued and improved if possible. (Image courtesy of Luis Rey).

There are naturally good and bad speakers, and it’s always going to be hard for someone who works on gastralia in theropods to interest someone working on the behaviour of Permian fishes, but the simple errors and mistakes we can at least try to deal with. As ever in these guides, most of this information is simple, straightforward and probably obvious – but enough people don’t seem to realise it so I might as well put down my 2c:

You will of course have an abstract sent in and accepted if you are going to be speaking at a meeting, though of course you might also have to give departmental talks, or lectures when visiting other institutions and so on, but the main thing of course is to have a clear idea of what you want to talk about and what message you want to get across. Write your talk with these things in mind – have two or three key points you want to get across and make sure they are prominent and clearly expressed. As with papers LINK different people write things in different ways so don’t be bothered about your style provided you produce a good set of slides and a well rehearsed talk. Personally I write the two together, putting the slides together on the computer and mentally assembling a monologue to fit them. I don’t write anything down about what I want to say, and I don’t use notes when speaking, but I know others who write down their words then build a slideshow around them, or vice versa.

Do make sure the talk is smooth and logical. Treat is somewhat like a paper, so you introduce the problem, what you wanted to test / analyse (or what you discovered that impacts on old ideas), how you analysed it, what the results were and what that means. Remember that you only have a limited amount of time and you can’t go into great detail (and it will be a technically minded audience) so keep things brief. If necessary repeat your main points as a summary to make sure they get across.

Do try to make it interesting. Of course there will probably be people there who are not at all interested in what you are saying, (sad, but true) but you must engage everyone as best you can. Cutting out the jargon or in-jokes from your field will stop you losing people. Don’t just string together a bunch of photos and talk about each one in turn, even if they are all good examples of the points you are making.

As for the slides themselves, a few simple rules can really improve the presentation of slides to make them more accessible. Use a large and simple font so everyone can read the text easily. Keep each slide clean and simple – better to use three slides for just a few seconds each to show two graphs than cram three onto a single slide where no one can see them properly. Similarly, don’t write too much, people have to read what is written *and* listen to you talk, put up a slide with 150 words on it while you are talking as well means people will certainly miss something. Keep the background either simple or faint, so that it does not distract form the words or images you want them to focus on. Don’t use distracting things like sound effect, or weird slide changes and wipes.

Make sure the title slide includes the names and affiliations of all the people involved (especially where limitations mean they were not listed in the abstract). Do include an acknowledgement slide for funding, other colleagues etc.

Practice your talk, especially if it’s your first. Make sure you know what you want to say, how and when and the points you want to emphasise. Ideally rehearse in front of a critical audience like a few colleagues and ask for help – they will be able to spot things you never would and improve both the presentation and the presentation. When speaking, face the audience (not the computer or the board behind you, even when pointing things out on the slides), and speak slowly and clearly and if there is no microphone, loudly. Try not to move around too much or give wild and exaggerated gestures, though equally don’t stare at your feet and speak in a monotone – half of getting the audience to listen is just making eye-contact (as it were) and speaking to them, rather than at them.

Make sure that what you say matches the slides – it’s really hard to follow a talk where there are lots of words on the screen but the person is talking about something else, or there are only five bullet points, but he lists six factors etc. If you do have a list of points for example, go through them in the same order as they appear on screen. Ensure that you have enough time for each slide, both to talk about it, and the audience to absorb what is on the screen.

As for actually giving the presentation, first off make sure you have right file format for the talk. There’s nothing worse than turning up with a file the computer can’t read. If you can take along a back-up copy to your basic CD / flash stick (I e-mail myself a copy so I can access it easily at short notice in case of loss).

Prepare for questions (again rehearsals with people will help you spot the likely ones, so you can prepare your answers), and if you do get something you can’t answer or are not sure about, there is nothing wrong with saying you need to think about it, or offer to discuss it later – better that than guessing.

Finally and above all, make sure you stick to the time. Make sure you know how long you have to give your talk and that you talk lasts that long – at big meetings if you overrun, you’ll get hauled off, at small ones they’ll let you run late and you’ll be resented for making other people late and miss their coffee breaks.

Once more, this is short, probably not very sweet and based on my own experiences as a speaker and listener, and I hope it provides some information and ideas. In general scientists are excellent speakers, but public speaking can be intimidating for anyone. As with the posters LINK I suspect bad talks are generally a result of people doing them at short notice with too little time given to preparation and rehearsal. I once saw a colleague write his talk for a meeting on his laptop in the coffee break before his talk. He did the whole thing in 20 minutes and actually gave a great talk, but I would not consider it a model for other to follow.