I recently read that the format of a modern talk came be traced all the way back to the Greeks, in particular to Cicero. According to him, a persuasive talk follows the sequential set of steps:
Invention. An empty canvas or page can be scary for fine art painters or writers, but what is worse is after the first stroke lands on the canvas or the first word on the page, a world of possibilities shrinks dramatically. After the second stoke or word there are no more possibilities, the path has been laid out. This is no different in scientific talks, though here there is no canvas or page. I know what you are thinking, but no. No PowerPoint. Presentation Gurus like to use sticky notes (a.k.a. Post-It) to start the creative process because it allows a more flexible brainstorming process. Whether you start with sticky notes, a blackboard, a whiteboard or window, or a scratch-book, remember that crafting your message involves the audience's needs, background and exceptions. I have addressed this issue in a past post.
Arrangement. Now that you have the content, give it some structure. The basic structure is the following: Ethos first, then logos, finally pathos. Ethos refers to the reputation, logos to the rationality of the argument, and pathos to the emotions. In a scientific talk the ethos is leveraged by the moderator, or at least should be. A quick sketch before of a guest professor or researcher citing patents, affiliations, number of published papers and books, and obtained awards is intended to raise the his/her ethos, basically saying that there is some level of credibility. For graduate students building his/her ethos is more difficult, but basically if the talk is not flawed in the content, s/he would have raised- or at least have not hurt- his or her ethos. The purpose of the ethos is to have the audience on your corner. In the classical approach, narration follows ethos. You have entered the logos sector. You present your points and logically supporting them. You narrate in a scientific storytelling manner.
Allow me to elaborate some more on that term, scientific storytelling. If the term makes you feel uncomfortable because of the association of the word storytelling with the word fiction, i.e. not true, let me be clear. Science has nothing to do with "the Truth". As professor Paul Grobstein writes
Scientific statements are not either claims or approximations to ‘Truth,’ but provisional stories, reflecting human perspectives, that get progressively less wrong. Science is therefore fundamentally not about security but about doubt, not about knowing but about asking, not about certainty but about skepticism. Scientific stories are written not to be believed but to be understood, made use of as appropriate, and revised. Narrate your finding and your observations as it were a story, a non-fiction story that this, but ultimately a story.
Pathos is a complicated in scientific talks, for emotions in science are a misunderstood phenomena. However, ending your talk in a high note is something you want to achieve, that's where pathos comes in, moving the audience in the emotional plane in your call to action, whatever this might be.
Style. Very close to narration is he story telling is style. Very scientific discipline and audience have their own set of stylistic parameters and rules. Synchronizing these two is a challenge, but here are some guidelines. Proper language: words that suit the occasion, audience and scientific discipline. Let me give you an example. Reading this 5*(1/5) = 1 aloud, you would say at some point 'one over five', which is completely fine. However, if you replace the 5 for a matrix (a rectangular arrange of numbers) named A, you don't say 'one over A' but 'A inverse'. Proper language. However, watch out for jargon and not so common abbreviations. In the field of linear algebra most people will know that SVD stands for Singular Value Decomposition, but the very similar Eigenvalue Decomposition is not always abbreviated as EVD.
Related to proper language is clarity. Should be obvious, shouldn't? Sadly, in many scientific talks clarity is the exception rather that the rule. Working yourself through clarity you might step on the Curse of Knowledge:
This is the Curse of Knowledge. Once we know something, we find it hard to imagine what it was like not to know it. Our knowledge has "cursed" us. And it becomes difficult for us to share our knowledge with others, because we can't readily re-create our listeners' state of mind. Also be mindful that in a multimedia presentation, such as today's scientific presentations involving a stack of visuals, details clutter rather than clarify. Let the KISS principle guide you: "Keep Simple Stupid!"
Another aspect of style is vividness. Can your words recreate imagery in the audience minds? A good way to achieve this is using metaphors and analogies. Here is an except from Sir Winston Churchill's book The River War
The Soudan is joined to Egypt by the Nile, as a diver is connected with the surface by his air-pipe. Without it there is only suffocation.Rehearsal. I already talk about it. Check out this past post on the topic.
Delivery. It's your turn. Your name has been called. It's delivery time. I'll dedicate a whole post to this topic soon.
 Grobstein, P. Revisiting Science in Culture: Scienceas Story Telling as Story Revising. Journal of Research Practice, Vol. 1, Issue 1, Article M1, 2005.
 Heath D. and Heath C. Made to Stick: Why Some Ideas Survive and Others Die. Random House, 2007.
* Based and inspired on the book Heinrichs, J. Thank you for arguing. Three Rivers Press, 2007. Chapter 23: Give a Persuasive Talk.