How to give a good research talk?

Use all Seminars on Training Voice, Body and Style

Before you start, make sure that you use the excellent seminars in the Weiterbildung program at TU Darmstadt. Especially these excellent seminars:

Also, Jan has a lot of books on this topic!

Core Steps

  1. Start with something engaging: Funny story, Personal anecdote, Interesting statistic (that the audience doesn't know). The first thing you say determines whether the audience will be leaning in to listen or picking up their phone to check email.
  2. Explain the significance: What's the broader impact on your field? What's the future impact on society? Why should anyone care? It doesn't matter if you're presenting to experts in your field or a broad audience. Succinctly and clearly explain why your work matters.
  3. Make it easy to understand: Don't blind the audience with science! Use simple visuals! Minimize jargon! Define jargon if needed! The most knowledgeable scientists/engineers explain concepts in the simplest terms. They aren't out to impress you. They prefer clarity.
  4. Make it about the audience: What interests the audience? What do they get from listening? What info do you have that would be of value to them? Often bad talks just summarize what a speaker did. They assume the audience will be interested. Good speakers # Be authentic: If you're funny, be funny - but don't force it! If you're heartfelt, be heartfelt! If you're quirky, be quirky! Find your own voice and presentation style. There's no single correct style. Authenticity shines through and resonates with audiences. Being considered inauthentic is a researcher's death.
  5. Don't oversell: Be candid and precise! Mention limitations of your work! Be generous in acknowledging the work of others! The best researchers communicate significance without hype or over-generalizing. Otherwise, you lose trust and credibility with your peers.
  6. Don't put down others: Don't denigrate prior research/tech to motivate yours. Discuss benefits and limitations of prior work. Then explain what your work adds. This is a common mistake amongst early career researchers and insecure senior scientists. But it's unnecessary
  7. Anticipate questions: Preempt expected questions or objections. Add more or less clarity based on what the audience needs. If you know your audience, you can anticipate what they'll be thinking and how they'll be feeling during your presentation. Then craft your message.
  8. Provide a glimpse of the future: What comes next? Will there be a follow-on study? When might the science or tech be ready to translate into societal applications? If you've given a good talk, the audience will want to know about the future. Don't leave them hanging.
  9. End with your takeaways: What do you want the audience to remember or do? Tell them! Be explicit. I advise 3 takeaways max. An audience won't remember more. But if your takeaways are a resource, like this list, then longer is ok. Just enumerate and summarize at the end