TED Talks: The Official TED Guide to Public Speaking

by Chris J. Anderson

Presentation literacy isn’t an optional extra for the few. It’s a core skill for the twenty-first century.

Sophie’s core idea about laughter—that its evolutionary purpose is to convert social stress into pleasurable alignment—had

The central thesis of this book is that anyone who has an idea worth sharing is capable of giving a powerful talk.

People love stories, and everyone can learn to tell a good story. Even if the lesson you might draw from the story is familiar, that’s OK—we’re humans! We need reminding! There’s a reason religions have weekly sermons that tell us the same things over and over, packaged different ways. An important idea, wrapped up in a fresh story, can make a great talk, if it’s told the right way.

In fact, the same questions you ask as you do your research can help provide the blueprint for your talk. What are the issues that matter most? How are they related? How can they be easily explained? What are the riddles that people don’t yet have good answers for? What are the key controversies? You can use your own journey of discovery to suggest your talk’s key moments of revelation.

The key principle is to remember that the speaker’s job is to give to the audience, not take from them.

“When people sit in a room to listen to a speaker, they are offering her something extremely precious, something that isn’t recoverable once given: a few minutes of their time and of their attention. Her task is to use that time as well as possible.”

An organization is fascinating to those who work for it—and deeply boring to almost everyone else.

There’s a drastic consequence when you rush through multiple topics in summary form. They don’t land with any force.

To say something interesting you have to take the time to do at least two things: Show why it matters . . . what’s the question you’re trying to answer, the problem you’re trying to solve, the experience you’re trying to share? Flesh out each point you make with real examples, stories, facts.

Whether your time limit is 2 minutes, 18 minutes, or an hour, let’s agree to this as a starting point: You will only cover as much ground as you can dive into in sufficient depth to be compelling. And this is where the concept of a throughline really helps. By choosing a throughline you will automatically filter out much of what you might otherwise say.

So a throughline requires you first to identify an idea that can be properly unpacked in the time you have available. You should then build a structure so that every element in your talk is somehow linked to this idea.

Ken Robinson. He told me that most of his talks follow this simple structure: A. Introduction—getting settled, what will be covered B. Context—why this issue matters C. Main Concepts D. Practical Implications E. Conclusion

An issue-based talk leads with morality. An idea-based talk leads with curiosity. An issue exposes a problem. An idea proposes a solution. An issue says, “Isn’t this terrible?” An idea says, “Isn’t this interesting?” It’s much easier to pull in an audience by framing the talk as an attempt to solve an intriguing riddle rather than as a plea for them to care. The first feels like a gift being offered. The second feels like an ask.

Persuasion is the act of replacing someone’s worldview with something better.

Paradigm and dialectic are not technical terms like DNA that specialists can’t avoid. They’re metaconcepts—concepts about other concepts, rather than concepts about things in the world. Academese, bizspeak, corporate boilerplate, and art-critic bafflegab are tedious and incomprehensible because they are filled with metaconcepts like approach, assumption, concept, condition, context, framework, issue, level, model, perspective, process, range, role, strategy, tendency, and variable.

How can we switch the tone from “look what we’ve achieved” to “look how intriguing this is”?

Slides move at least a little bit of attention away from the speaker and onto the screen. If the whole power of a talk is in the personal connection between speaker and audience, slides may actually get in the way of that.

REVEAL! The most obvious case for visuals is simply to show something that’s hard to describe.

Other speakers still seem to believe that you enhance the explanatory power of your slides by filling them with words, often the same words that they plan to utter. Nothing could be farther from the truth. Those classic PowerPoint slide decks with a headline followed by multiple bullet points of long phrases are the surest single way to lose an audience’s attention altogether. The reason is that the audience reads ahead of the speaker, and by the time the speaker covers a specific point, it feels old hat. When we see speakers come to TED with slide decks like this, we pour them a drink, go and sit with them at a computer monitor, and gently ask their permission to delete, delete, delete.

But otherwise, words on the screen are fighting your presentation, not enhancing it.

But it is important to distinguish unscripted from unprepared. In an important talk, there’s no excuse for the latter. Many unscripted talks, alas, result in half-baked explanations, non sequiturs, key elements missed, and rambling overruns.

One temptation many speakers fall prey to is to use their slides as crutches. In the worst form, this means a series of dismal slides covered with text and bullet points that the speaker works through laboriously. Most people by now understand that this is a truly terrible way to give a talk. Every word you speak that someone has already seen on a slide is a word that carries zero punch. It’s not news anymore.

Believing what you are saying in real time has a much larger impact than saying the exact right words. I personally tend to list out bullet points of what I want to talk about and then try communicating those ideas in my natural language as if I’m talking to friends at a dinner table. The key is to keep your mind focused on the ideas and let the words fall out.

There are, to be sure, occasions when you can start with a thank-you or two, especially when you’re speaking at an event where there’s a strong sense of community.

It’s amazing how many talks simply fizzle out. And how many more go through a series of false endings, as if the speaker can’t bear to leave the stage. Unless you plan your ending carefully, you may well find yourself adding paragraph after paragraph. Finally, the key point, as I said . . . So, in conclusion . . . And just to emphasize again, the reason this matters . . . And of course it’s important to still bear in mind . . . Oh, and one last thing . . . It’s exhausting. And it will damage the talk’s impact.

Many speakers use their slides as memory nudges. We discussed this briefly earlier in the book. What you mustn’t do, of course, is to use PowerPoint as a full outline of your talk and deliver a series of text-crammed slides. That’s awful. But if you have elegant images to accompany each key step of your talk, this approach can work very well, provided that you’ve thought about each transition.

Sometimes speakers look at the wrong monitor, confuse the next and current slide screens, and panic that the wrong slide is showing. But much worse is the tendency to become too dependent on the notes on these screens and to be constantly referring to them. This is actually more off-putting than a speaker looking down at notes.

An 18-minute talk contains maybe 2,500 words. Many people can read 2,500 words in less than 9 minutes and retain good comprehension. So why not do that instead? Save the auditorium cost. Save everyone’s travel. Save the chance that you might flub your lines and look foolish. And get your talk across in less than half the time it takes to speak it.

There are two ways of losing an audience: going too fast is by far the rarer of the two. Going too slowly is actually the bigger problem, since it allows time for people’s minds to wander off.

Martin Luther King’s “I have a dream” speech was delivered at around 100 words per minute. It was perfectly crafted and delivered for its purpose. But it’s unlikely that your task today is to address a crowd of 200,000 people at the heart of a major social movement.

There are numerous debate formats that offer exciting ways for this to happen. One of the best is an Oxford Union format, two against two. The speakers alternate with, say, 7-minute presentations for and against a controversial proposition. After moderator or audience engagement, they each have a 2-minute wrap-up, followed by an audience vote. (You can see this in action on the excellent website IntelligenceSquaredUS.org

Self-proclaimed “geek events,” the Ignite talk series has a similar format, though in this case speakers’ time is reduced to 15 seconds per slide. Both methods make for terrific, fast-moving events.

Most conferences serve a single industry or knowledge specialty. There, everyone has a common language and starting point, and it makes sense to allow speakers time to go really deep and describe some specific new learning. But when the content and audience are wide-ranging, a speaker’s goal isn’t to exhaustively cover a niche topic. Instead, it’s to make her work accessible to others. To show why it’s interesting. To show why it matters. That can usually be done in less than 20 minutes.

As of late 2015, TED Talks are viewed some 100 million times every month—1.2 billion times a year.