Organizing Your Talk

Probably THE most common problem with technical presentations is poor organization. As an expert in your field, it's easy to neglect the forest for the trees. You're intimately familiar with the connections between your ideas, but the audience is not. Therefore in your presentation, you need to tell a focused, cohesive story and guide the audience through that story in a clear and logical way.

Two excellent approaches for organizing talks are the Hourglass Structure and the Pyramid Principle. These methods are complementary -- the hourglass deals with the flow of information, and the pyramid deals with logic. So I strongly recommend using both!

The Hourglass Structure

If you want the audience to understand and appreciate what you're saying, you can't dive immediately into the minutiae. You need to establish context and connect technical details to the bigger picture. The hourglass structure shows you how to do just that.

This conceptual framework depicts how broad or specific the information should be at any given point in your talk. Effective presenters start with the broad big picture, then progressively transition to the narrow technical details, and connect back to the big picture at the end. Here's the hourglass structure for a typical research story:

Background info / motivation: Explain why this topic matters and provide an appropriate amount of background information. This might include a discussion of your research area in a broad sense, explanations of core concepts, and/or a brief review of relevant literature.
Specific research question / knowledge gap: This is where you transition from the big picture to the specific project you're going to present. Where did previous work leave off? Was an important question left unanswered? Was there a big problem that couldn't be solved before?
Your solution: What did you do to solve the problem / answer the question? This is the narrowest part of the hourglass but is probably where you'll spend the majority of time in your talk.
Major findings / results: Recap your main results and explain how the details all fit together to support your central message (more on that below).
Significant and broader implications: How does this work contribute to the field? Why does it matter? You may think it's too repetitive to talk about the broad significance here since you already discussed it in the intro. But the audience doesn't necessarily remember! Connecting back to the big picture helps end your talk on a strong, memorable note.

Audience-Specific Adjustments

When tailoring your presentation for specific audiences, adjust the hourglass accordingly. Giving a talk for the general public? Use a wider hourglass, broadening the big picture to ensure that it's something everyone in the audience cares about. Presenting at your research group's weekly meeting? You can probably use a narrower hourglass, omitting basic background info at the beginning and longer-term implications at the end.

Nested Hourglasses

I recommend applying the hourglass structure not just to your overall presentation but to each individual section. For example, if you did three sets of experiments, present each one by first explaining the purpose of that experiment, then showing the detailed results, and lastly stating the key takeaway of those results. This ensures the audience can follow along with your entire presentation even if some of the nitty gritty details are over their heads!

The Pyramid Principle

This concept was developed by Barbara Minto for the world of business consulting, but it's really an excellent tool for any kind of professional communication. There are two key ideas here. First, your presentation should have only ONE take-home message, which is at the top of the pyramid. Second, all information should be structured such that "ideas at any level in the pyramid [are always] summaries of the ideas grouped below them." It ends up looking like this:

Top of pyramid: Central message. I define a central message as a one-sentence answer to a question about a topic. It's not just a topic statement (e.g., "I'm going to tell you about my work on quantum computers") -- it needs to answer a specific question (e.g., "I developed a way to reduce error in quantum computers by...").
Second tier: Sections. This is the internal structure that groups ideas together. In research presentations, it's common for each section to deal with a specific set of experiments, a particular question, or a step in an iterative process. The boxes in this tier should state the key takeaway for each section, and those takeaways should directly support the central message.
Lower tier(s): Technical details. Fill in the specific info that will be included in each section of your talk. Aim for only 3-5 boxes in any grouping (although it's fine to expand into lower tiers as appropriate). And always check for alignment, i.e. does each detail support the section takeaway? If not, move that detail to the appendix.

Tiers are Crucial

A common pitfall is the "data dump," i.e., presenting every single thing you've done because you want to show how productive you've been. Unfortunately, this only creates confusion for the audience and obscures your central message. In essence, you're missing the second tier and instead connecting the central message directly to a barrage of details. That's not a pyramid -- it's Cthulhu. Avoid.

Top-down vs. Bottom-up

When you first start developing a presentation, it's very rare to already have your ideas clearly organized. More commonly, the pyramid is built from the bottom up. I.e., start with your detailed data, then figure out how to group it all in a logical way and define what conclusions can be drawn from each grouping. If the overall central message is already known, you might take a top-down approach instead. Either way is fine, and the end goal is the same.

There's a lot more to the pyramid principle that I'm not covering here, including the importance of defining the explicit logic of your organization. You can learn a little more here, or better yet, check out Minto's book.

This article by Dr. Robyn Javier is licensed under CC BY 4.0