Giving a Good Talk
Three Things #82: August 20, 2023
I love giving talks. There’s something magical about being in front of a room full of people that are paying attention, listening to you, and actually interested in what you have to say. I’ve always felt this way, maybe because I find communication interesting in and of itself, but I feel doubly this way today because we spend so much of our time staring at screens and communicating through mediums like chat apps and social media, and distractedly multitasking. The chance to face a room full of people with no intermediation and no filter is just really, really unique and special, and it’s even better when I get to share something I’m really excited about with the audience.
The other thing I love about giving talks is that it’s an excellent opportunity to organize and structure my own thoughts and to test my understanding of a topic. Giving a talk really means telling a story. I like hearing stories and I like telling stories. The best talks are just captivating, interesting stories, well told—although, granted, most tech talks aren’t nearly so skillfully written or delivered—and who doesn’t like listening to interesting stories? I often choose topics that I have only a basic understanding of, but that I hope to understand more deeply, because giving a talk on the topic requires that I get smart about it quickly. Good examples of this are my talk at ETHBuenosAires ages ago on scaling Ethereum, my talks at ETHDenver in 2022 and 2023, and my EthCC talks on the use of Bitcoin in El Salvador (2021, 2023).
I’ve given dozens of talks on dozens of topics over the years, and over time I’ve become more comfortable and more confident. Friends and colleagues occasionally ask for advice so I thought I’d share some ideas here.
Thing #1: Keep it Spicy
If I had to give a single piece of advice on giving a talk—if you remember one thing from this issue—it should be this: pick a topic that you’re genuinely curious and passionate about and present it in a creative way. If you get nothing else right but you nail this you cannot help but deliver a good talk.
Curiosity and passion show, no matter how smooth or awkward a speaker you are, and they’re endearing qualities. If you’re really into the subject matter your audience will pick up on this and be much more likely to pay attention, follow along, and trust you. You’re also much more likely to write a better, more compelling, more interesting talk if given the opportunity to talk about something you’re really interested in.
And imagine the opposite: listening to someone give a talk that they’ve given many times before on a topic they’re just not that interested in. A common example of this is when someone on the marketing team of a technical project that they don’t fully understand or appreciate gives a “sales pitch” talk—a talk they didn’t write and just aren’t very invested in. Nothing is worse than giving a talk on a topic that you find dull. You’ll cut corners, you’ll lack energy and pizzazz, and all but the most sociopathic, pathological liars are simply unable to muster the energy and fake passion for a topic that bores them.
You can simply tell when someone is absorbed by a topic and when someone just isn’t that into it. You can see it on their face and hear it in their words. When a compelling, charismatic founder pitches you on their vision, or a passionate builder describes something they’ve poured themselves into building, you can clearly tell the topic means everything to the speaker. That sort of energy and enthusiasm is infectious; we’re hard wired to react this way to infectious enthusiasm and it’s very hard to resist drinking in someone else’s energy and passion.
This sort of begs the question, how do you find a topic that interests you? I don’t think it’s very hard. I keep running lists of interesting topics as they occur to me (those lists also form the basis for the things I write about here). So I suggest starting such a list if you don’t already have one and just jotting things down when they occur to you. When you ask yourself a question or wonder why a thing is such and such a way, write that down. When you notice a curious headline, listen to an interesting podcast, or someone tells you about an interesting book they read, write that down (taking notes, even brief ones, after reading or listening to books, articles, magazines, and podcasts is a great exercise). The next time you give a talk might be the perfect opportunity to explore one of those topics or answer one of those questions.
Finally but equally importantly, I also recommend thinking outside the box and keeping your topics spicy and a little controversial. I’m not saying you should shit on everyone and make enemies, but consider taking a controversial stance on a popular topic or playing devil’s advocate. No one will blame you for this, especially if you say upfront that you intend to play devil’s advocate, explain why, and keep it constructive and positive. There are literally hundreds of talks at an Ethereum conference called “How to build an xyz app”; but talks called Bitcoin is for Dictators aren’t as common. A good rule of thumb is that you want to piss someone off with your talk title and its contents, but ideally not too many people (and if you haven’t pissed anyone off you’re not thinking outside the box and keeping it spicy enough).
To summarize, a good strategy is: keep a running list of interesting topics and questions, pick one that you’re deeply curious about, and take an unpopular or outside the box stance when you talk about it.
Thing #2: Throw Away the Script
I’m all about authenticity. In any conversation or dialog one of the most important things is that you come across as honest and authentic, i.e., as your full and true self. This applies equally to giving a talk. But for some reason I think a lot of people don’t realize this. They think of a talk as a performance and feel that, like a performance, a talk should be scripted and memorized. Unfortunately if you actually script and memorize a talk, it will sound robotic and inauthentic and your audience will notice, I promise. This is just how communication works.
Alright, so if you’re not supposed to script a talk, what should you do instead? One obvious alternative is to wing it! Prepare some slides, review them a few times, have some rough idea what you want to say to each slide, and then deliver your talk off the cuff. I’ve done this a few times (typically when I simply didn’t have much time to prepare) and I can say from experience that it almost never goes well. It checks the authenticity box but it fails in other ways. For one thing, when I do this my delivery isn’t clean and I find myself tripping over my words. I say more than I should. I repeat myself. I sometimes slip into word salad. And sometimes I blank out and forget what I wanted to say in the first place. It can also result in a poorly structured presentation where information isn’t presented in the optimal order.
I think it’s important to strike a balance between being structured and prepared on the one hand and being authentic on the other. In my experience the best way to do this is to prepare the story that you want to tell without scripting it too much. First of all, make sure the story has a beginning, middle, and end (more on that next) and that it’s told in the right order and flows well.
The only way to really do this is to speak the story out loud to yourself in order. I start by creating a draft deck and I aim for 30-60 seconds per slide. I try to keep my slides extremely visual (with bright colors, simple ideas, and clear photos) with minimal information (numbers, text, graphs, etc.) so that the audience pays attention to me and not to the slides. Each slide should set the tone of the moment but you should communicate the information, not your slides.
Once the draft is done I begin walking through it from the beginning until I reach a point where the story or the presentation feels unnatural. I stop, fix the problem, then start again from the beginning. It typically takes 10-20 passes over the deck from the beginning until I’m comfortable with the story: that all the required information is there, is in the right order, and that anything spurious or unnecessary has been removed or, at least, moved to an appendix at the end.
Once I’m comfortable with the deck I’ll walk through it a few more times to make sure I’m comfortable with the verbal delivery and timing. I try to always do this out loud, with a timer, but I’ll admit that I sometimes cheat and cut corners on these points. If the talk is especially important and/or if I have extra time I’ll practice it in front of some friends or colleagues for additional feedback. This is also extraordinarily helpful because almost by definition your story will come across differently to other people than it does to you, but given how last minute I prepare my talks this is a rare luxury!
I’ve been using this method for years and it works well for me every time. Give it a shot and let me know if it works for you too.
Thing #3: Structure
Giving a talk without a script works much better if the talk has structure. Too little structure and you’ll get lost and the talk might fall apart. But, too much structure and the talk will feel scripted and inauthentic, and you won’t be able to adapt to the circumstances. You want some room for improvisation. The best example of too much structure is to structure each and every slide, each paragraph and sentence—in other words, scripting everything.
In my experience the “goldilocks spot” involves an overall narrative arc with checkpoints and bookmarks (I’ll explain the difference in a moment). There are many narrative formats, and the perfect format for any given talk will depend on your style, the audience, the subject matter and the story you want to tell. But for the vast majority of talks I find that a Hegelian dialectic format works quite well (and you can always tweak the format to your needs).
The gist is that you break your talk into three sections or acts. Act I is the “thesis”: it sets the stage and introduces a dilemma and/or sets up some sort of tension. Act II is the “antithesis”: you refute or contradict what you presented in Act I, but in an ultimately unsatisfactory way. Act III is the conclusion or “synthesis”, in which you bring together the thesis and antithesis in a satisfying way, resolve the tension, and tie off any loose ends.
These ideas are quite abstract and sometimes a concrete example is helpful. The structure I used for my Bitcoin is for Dictators talk used this format. Act I introduced some basic background information on El Salvador, Bitcoin, and the Lightning network, and introduced the cast of characters. It centered on Nayib Bukele and his antidemocratic tendencies, and suggested (without saying so explicitly) that he could become the world’s first “Bitcoin dictator.” Act II presented a very different perspective, suggesting that in fact Bukele may be an enlightened leader playing the long game and truly acting in the best interest of his people and that, while the country’s investments in Bitcoin and its infrastructure may be unpopular in the short term, they’ll ultimately pay dividends for the country. Act III connected the two perspectives by urging skeptical optimism (or optimistic skepticism!). It concluded the story by saying that we should remain vigilant but give Bukele and El Salvador a chance, and continue to hold him and the country to a high standard that includes the freedoms that Bitcoin stands for, because only then can both dreams be realized.
If you use this or a similar three act structure you’ll have no more than five sections in your talk: a brief introduction (explaining who you are and where you’re coming from, why this topic is interesting and important, why the audience should listen to you, etc.), acts I, II, and III, and a brief conclusion or Q&A. Each section represents a natural checkpoint and an opportunity to structure the use of time. For instance, if you have 25 minutes for a talk, you could simply use 5 minutes per section, or else 2-3 minutes each for the intro and conclusion and slightly longer for the main acts. Then you don’t need to time yourself on each slide or each point; you just need to check your time at the end of each section and speed up or slow down accordingly. I find this sort of pacing incredibly helpful.
Bookmarks are like checkpoints but they’re not connected to acts/section breaks. They’re individual points that you want to emphasize or really drive home. A talk should have no more than 2-3 of these, maximum. You might decide to conclude each act with one such point. While you shouldn’t memorize the talk itself, you should commit the list of bookmarks to memory and make absolutely certain that you touch upon each one of them during the talk no matter what. Write them on an index card or on your wrist or something if you must!
If you adopt a structure along these lines I think you’ll find that your talks are better paced, easier to follow, and more well received overall. If you do decide to use this structure, let me know. I’d love to know how it goes and to compare notes.