A comprehensive guide to creating slides

Creating slides is an interesting aspect of Science. Focusing on communicating what you have clearly without overwhelming your audience, while maintaining enough depth on the subject is an art in itself. There is a fine balance between design making it look good, communication making efficient use of words and sceince tying every slide. The possible slides that one can make is potentially infinite, but just like any art form, the slides you make will most certainly have your trademark style in them.

Talking about slides, we also need to talk about figures, but those are for a different blog. As of now, I am in the process of creating slides for my qualifying exam and I wanted to record this for anyone in the future that wants a compact guide on how to structure slides, how to communicate clearly, and what to expect in a preliminary presentation ( although this applies to any situation ).

The structure should roughly contain the following sections for a full presentation:

  • Motivation
  • Gap and Problem
  • Objectives
  • Approach
  • Results
  • Conclusion

Establishing a structure is essential even before one starts with creating the slides. This gives a good framework to build around and improve incrementally. The structure should follow an hourglass shape - The start should be a problem statement wide enough for anyone to understand and provide motivation for why someone should even be interested in your talk. This should be followed by the introduction of ideas in the field of interest. The middle part, which will ideally be the largest part of the presentation should be very narrow, focusing on the specific approach / idea / method that is the crux of your presentation. Finally, the Results should again be broader, calling back to the motivation at the beginning of the presentation and address that gap or motivation that was provided.

One important thing grad students have is providing updates and slides on updates, which usually occurs every week or on some regular basis. These should also follow the similar hourglass shape, but focus more on the results if there are interesting results or on the method if feedback on the methods is the aim of the presentation. The key to a good update slide is not to just jump in assuming everyone has the context of what you have been working on. It is very easy to get bogged down on the details which most people won’t care about, but it’s tempting to add these. You would have been working on this same topic for weeks / months but others don’t and won’t spend much time on the details. These presentations should always have a crisp introduction of what the aim is, what has already been done, and what the slides are going to cover ( replacing the broad overview for a general presentaiton)

Finally, this is something that I really like to see in a presentation. Add a summary slide at the beginning and at the end of the presentation, that is a very high level overview of your entire presentation. Cover key findings and important approach directions. All my presentations have the principal components of what I want to convey along with a paragraph of aims. This grounds the presentation and gives a bite sized piece for people to remember. ( This approach is called bookending )

I will add further ideas and tips to making slides more interesting below

  • Titles: Every slide title should always have some key takeaway. If your audience were to read only one phrase from each slide, that phrase should go to the title. For example, a title such as “Model Architecture” is a fine enough title and will generally convey what you want to. But a title that says “X architecture models global interactions” is a much better title. It covers what architecture is used, why it was used. This conveys what the slide will eventually cover in a crisp manner. I think writing good titles is an artform in itself.

  • Text content: Too much text in a slide is never good. This I think is global and people intuitively understand this. The same is true for figures too, a single slide should not have more than two figures at max. If there is any more figures to add, that should go to a new slide.

  • Presenting to an audience: But repeating sentences that are on the slides when presenting them to an audience is not a good ideas. You want to add and explain what the words mean, never repeat the words as this creates a cognitive load on the listener reading the same content and repeating the same content. ( although this is fine for key points and important details )

Addressing questions: Most of the questions in a presentation won’t come from the top or the bottom part of the hourglass. Most questions should arise from the middle part, in the implementation and the results section. Questions will test you as a person and your qualifications with regards to the subject. It’s always good to be prepared with a set of questions you can expect from the crowd you will be presenting to. It’s always okay to say, you don’t know the answer to a question yet.