How to give a good oral presentation at an academic conference?

Today, I will discuss the topic of how to give a good oral presentation of your research at academic cconferences. This is a very important topic since if a presentation is interesting, more researchers will be interested in your work and it may bring several opportunities such as collaborations with other researchers.  So let’s get started!

The key points to giving a good presentations are the following:

1) Prepare your presentation early

The first key point  is to prepare your presentation well in advance before the date of the presentation. In other words, you should not create your PowerPoint the night before.  Preparing your presentation in advance will let you avoid underestimating the required time for preparation and for rehearshing, and thus ensure that you  have a good sleep before the presentation. If your presentation is ready a few days before the date, it will also give you more time to rehearse your presentation, which lead us to the next point.

2) Rehearsal is always needed

One should always rehearse before giving a presentation, although the amount of time necessary may vary. Typically, I will rehearse my presentation at least a few times in the hotel room the day before giving a presentation at a conference. But depending on how much experience you have, it may be safer to rehearse a few days before the date.  If you are new to doing academic presentations, it is recommended to first rehearse a few times by yourself  and then to rehearse at least once in front of your research advisor or your colleagues. You may also practice about how to answer questions with your colleagues, because most academic presentations are followed by questions from the audience.

If your native language is not English, it is important to rehearse even more. This will help you to feel more comfortable speaking in English and it will allow you to avoid searching for your words and hesitating during the live presentation. Besides, you may practice pronuncing some  difficult words before the presentation. I have seen a few times that a presenter did not know how to pronunce a word or pronunced it incorrectly, especially for math terms.

3) Know your audience 

Another important point is to know who your audience is. Are you presenting in front of experts in your very specific field or a crowd of researchers that may not be very familiar with your topic?  For example, if you are working on data mining algorithms, it is not the same if you present your work at a data mining conference where everyone is familiar with at least basic data mining concepts, an artificial intelligence conference or a general computer science conference. It is important to know your audience because in general, any kind of oral presentations should be adapted to its audience. For example, if you are sure that the audience is familiar with some basic concepts in your field, you do not need to explain them. Or if you think they are not familiar with some topics you may give further explanations.

4) Define your goal, and organize your presentation well

A related point is to clearly know what is the goal of your presentation.  What do you want the audience to understand and remember about your work? If you had an infinite amount of time, you could describe your research in very deep details. However, in reality there is always a time limit. For example, at a conference, organizers may give you 15 minutes to present your research. So what should you present during these 15 minutes? Personnally, my goal in such a conference presentation is to give an overview of the research. I want the audience to clearly understand the problem that I want to solve, then I give a rough overview of the technical solution with emphasis on what is new but without going in deep technical details or mathematic formulas. And then, I present experimental results and conclusion.  My goal is to make the audience understand the problem, what are the main originalities and the results. If they want to read the deep technical details, proofs of theorems, etc., they may read the conference paper, ask me questions or discuss with me after the presentation for more details. I have often seen some researchers showing slides full of mathematic formulas with many different symbols. In some field, it may be ok to do that. But insome other fields, you may lost your audience if you give too many technical details such as that.

A good organization for your presentation is important. To present a paper at a conference, the presentation may roughly follow the sections of your paper. The key parts are:

• Introduction, related work, problem statement:  introduce yourself, the topic, and then slowly introduce related work, their limitations and then the problem that you want to solve. You may show some example of the problem that you want to solve
• Your solution: explain how you have solved the problem
• Experimental validation: this is to show that your solution is good to solve the research problem
• Conclusion: A good conclusion should summarize the key points of your presentations (problem, key characteristics of your solution, key experimental results) and then you may talk about limitations and future work. The conclusion is very important. You can think of the conclusion as what are the key points that you want your audience to remember.

5) Test your presentation and prepare solutions for technical problems

Preferably, you should test your presentations in the environment where you will present before the presentations to avoid issues such as software incompatibility issues. Some common problems are:

• fonts in your Powerpoint presentation are not available on the provided computer (you may solve this problem by embedding them).
• videos in your presentation are not playing properly (e.g. problems with codecs, etc. – generally it may be better to avoid using videos),
• the provided computer uses a different software such as Open-Office and your presentation don’t look the same (a solution is to bring with  a PDF version of your presentation with you, just in case)
• you may need a special cable to  plug your laptop to the video projector (e.g. for Apple computers)
• …

Personnally, I bring my laptop, and a USB stick with a Powerpoint and also a PDF version of my presentation, just in case. I may also send one copy of my presentation to my e-mail just in case. Also, I try to arrive a little bit earlier  to be able to test my presentation on the provided computer if I cannot use my laptop.

6) Make your presentation interesting and interact with your audience

I have seen many boring presentations  at academic conferences, not because the topic was boring but because of a poor presentation. Some key points to make a good presentation:

• Arrive early at the presentation. Just before starting your presentation, look at the audience, smile at them. This will help you get more comfortable and kill the nervousness. Also breath well. If you are well-prepared,  everything will go well.
• Talk to your audience. Do not read your presentation. Otherwise, the audience will quickly lose interest.  If you are using Powerpoint, you should not put very long sentences in your slides. You should instead only write some very short lines  and use them as key points to remember where you are in your presentation rather than as a text to be read. The text may just be some part of sentences rather than full sentences.
• Always look at your audience.  Look around at the audience as you keep talking by making eye-contact with the audience to keep them interested. Do not always look at the same person (e.g organizer) but look at the whole audience.  You may sometimes give a quick look at your laptop to see at which slide you are in your presentation.
• Never turn your back to the audience. Turning your back to the audience is one of the worst thing that a presenter can do. In some cases, it may be ok to turn your back for a few seconds to briefly show something using a laser pointer  but never do it for more than few seconds, and preferably use the mouse instead of turning your back.  
• Avoid doing a presentation while sitting. Some people like to sit to give a presentation. But standing will  result in a better presentation and you can move more freely and people will see you better if you stand up than if you are sitting. Also stand straight and firm. Have confidence. Avoid unecessary movements that can annoy the audience such a playing with a pen in your hand while talking. Make sure that you are not standing in front of the screen so that people can see your slides.
• Avoid using too complicated words or a too large vocabulary. Try to explain your research using simple words. This will allow everyone to better understand, especially for non native speakers.
• Don’t speak too fast.  The goal is not  to explain your whole paper in a short amount of time. But it is to give the main idea about your research. If people are interested in your research they may talk with you after the presentation or read your paper. So carefully select the content beforehand so that you don’t have to speak very fast.  Usually, as a rough estimate, I prepare 1 Powerpoint slide for each minute of my presentation (but I also rehearse to check if the time is fine).
• You may use some techniques such as changing the tone of your voice to emphasize some words and break monotony. You may also consider telling a short story in the introduction to raise interest in your problem.
• Analyse the reaction of your audience. If they look like they are not understanding, you may need to give a few more explanations. If you are not sure, you may ask if everyone understood your explanation (e.g. “is that clear?”). If you feel like they are bored because explanations are too simple, maybe you can skip some basic explanations.
• Speak loud enough.

7) Other advices

A few more advices:

• Keep your presentation simple. from a technical point of view. It may be tempting to use all the features offered in Powerpoint such as animations, transitions, sound effects, videos, etc. But most of these will just distract your audience and should thus be avoided. Sometimes, animations may be helpful. But only use them when they are really necessary to help understanding.
• Take the time to design your Powerpoint slides well.  For example, rather than just copying and pasting illustrations from your paper into your slides, you may adapt them so that they look better for a presentation. You should avoid putting too much text on a slide or too many bullet points. Slides should also have a clear title. Do not use too many different fonts, colors. Adding some pictures may make your presentation more interesting. But use them well, when they help to understand  and do not put too many of them.
• It is great to spend time to create a great Powerpoint. But rehearsing is as important.  Therefore, avoid the mistake of spending too much time to create a great Powerpoint but not rehearsing enough.
• After a presentation, the audience should generally be given a few minute to ask you some questions. Make sure that you understand a question well before answering it. If someone ask you a question and you don’t understand it, you may ask them to repeat it. Besides, if a question is too complicated to answer or if the person is not able to answer you may just say to come talk to you later (e.g. during the coffee break) to discuss it with you.

There is obviously much more that could be said on this topic. If you  present your work for the first time at an academic conference, it is normal to be nervous. But the solution is to prepare yourself well. If you are well-prepared everythng should be fine.

—–

That is all I wanted to write for now. If you like this blog, you can tweet about it and/or subscribe to my twitter account @philfv to get notified about new posts.

Philippe Fournier-Viger is a professor of Computer Science and also the founder of the open-source data mining software SPMF, offering more than 52 data mining algorithms.