התמלול המלא של #119 Speechwriting: The Expert Guide by Simon Lancaster

#119 Speechwriting: The Expert Guide by Simon Lancaster
===
[00:00:00] Podcast Welcome Setup
---
Welcome to the InnoGuide podcast, where we share the works of authors and mentors to guide innovation. I'm your host, Bob Bouthillier. This podcast, my website, and my newsletter are focused on helping you discover the secrets of innovation in the age of AI. Since I don't rely on paid ads, I wanna thank customers who bring me onto their programs as a medical systems expert to work alongside their teams, helping them develop products smarter and faster without adding risk, while bringing human-centered AI skills into their organizations. Learn more at innoguide.net.
In this episode, number 119, I'll be sharing Speechwriting: The Expert Guide by Simon Lancaster.
[00:00:45] Why Speechwriting Matters
---
So you might be wondering, what does a book about speechwriting have to do with a podcast about innovation? Well, I'm glad you asked. Innovation is moving at a speed never seen before in anyone's lifetime. And because of what we shared in the last three podcasts, we now understand what's holding companies back from successful AI implementations from Podcast 116. we understand how to motivate and engage your teams in leveraging AI to achieve ROI in your organization from Podcast 117. And we've learned how to do this in a way that includes appropriate guardrails so you maintain quality and avoid brand debt from Podcast 118.
The bottom line for all of these initiatives is that we need to understand how to get our message across to our teams. And there are a few who know how to do this better than a seasoned speechwriter like Simon Lancaster. So that's why I'm sharing this important piece of the communications puzzle today, and understand that my goal is not to turn us into speechwriters, but rather to give us the highlights on how to reach our audiences in a reliable and repeatable way.
And at the end of the day, we just need to understand the tricks of the trade to grab attention and to hold a room. I hope you enjoy the podcast as much as I enjoyed pulling it all together from this angle. So without further ado, let's get started.
[00:02:45] Ratner Joke Disaster
---
We start with a story that shows that there's more than one way to get the attention of the people in a room. However, this particular approach wound up costing the speaker five hundred million pounds. It was April twenty-third, nineteen ninety-one. It was London at the annual dinner of the Institute of Directors.
This room was full of chief executives and politicians, dinner jackets, good wine, kind of audience that decides between the soup and the speeches, who is going places. At the podium is a man named Gerald Ratner. He runs a jewelry empire with shops on nearly every high street in Britain. And he's been given some advice, the kind of advice that sounds perfectly sensible in a planning meeting.
Put some humor in it. These are powerful people. Loosen them up. So Ratner tells a joke. He talks about one of his bestsellers, a cut glass sherry decanter set, which sells for four pounds ninety-five. People ask him how he can sell it so cheaply, and he says, "Because it's total crap." The room laughs, so he keeps going.

He talks about a pair of earrings. "They cost less," he says, "than a prawn sandwich from Marks & Spencer. But I have to say, the sandwich will probably last longer than the earrings." Bigger laugh. Ratner sits down. Good night's work. The next morning, the newspapers ran it. And here's the thing about that. The audience Gerald Ratner was performing for was a few hundred people in dinner jackets.
The audience that mattered most was not them. It was everyone reading the paper over breakfast, his customers. Within weeks, they stopped coming. His company lost around five hundred million pounds in value. Within two years, Ratner was out. The company eventually changed its name to get away from his name.
The joke has its own entry in the language now. When somebody torpedoes their own business with a careless remark, Britain says, "They've done a Ratner." Think about the arithmetic of that. A few minutes at a mic, a few dozen words, half a billion pounds. So why am I opening an episode for product teams and tech leaders with the story of a jeweler?
Because Ratner didn't fail at the business. He failed at the audience. He told a joke to a room he could see and forgot about the room he couldn't. Welcome to this episode of The InnoGuide Podcast.
[00:05:42] First Follower Question
---
Last episode ended with an assignment. Go find your first follower. Remember the dancing man on the hillside and the second person who stood up and joined him?
That was the whole story and the whole point. Nobody changes an organization by themselves. Change spreads one follower at a time. But I left a question hanging, and it's the one that matters in the 90 seconds after you find that person. When you walk over to them, what do you actually say? How do you get their attention?
How do you keep it? And how do you say it so they repeat it to someone else on Wednesday, on Thursday, on Friday? That's the point of today's show. And I'm leaning on a book I've been enjoying a great deal, Speechwriting, The Expert Guide by Simon Lancaster. If the name sounds familiar, it should. He's the former speechwriter for British cabinet ministers I borrowed from heavily in the last few podcasts.
[00:06:46] Episode Roadmap
---
Now, let me be clear about what this episode is not. I'm not going to teach us all to write speeches for a senator. We're not going to leave here as ghostwriters. What I want to do is smaller and I think far more useful.
Apologies to the politicians out there. I want to take the handful of techniques that professional speechwriters use to reach an audience and keep it. And I want to give those to the people who need the most. The engineer with a better way to work. The product manager trying to get a team to try something new.
The director who can see the problem but doesn't own the building. Because if you're trying to help a team learn to use AI well, the research from our first episode made one thing very clear. The technology is not the hard part. The hard part is the conversation. We'll take it in the order a speech actually happens.

What you do before you open your mouth. Then the first 20 seconds. How to make your case. How to make it stick. How to cut it down. How to deliver it. And finally, how to run it like a campaign because one good talk rarely changes anything on its own. And we'll finish where we started, with the question underneath all of it. What do you owe the people in the room?
[00:08:14] Speeches Still Matter
---
So this is nothing new. You've been giving speeches for a long time. We all have. So let's deal with the obvious objection first. You're thinking, "Nobody gives speeches anymore. We have Slack. We have email. We have a 30-slide deck that nobody will read. They'll all be on their phones during that anyway."
Lancaster opens his book by dismantling exactly that claim, and he's got a great target. Before the 2010 election in the UK, commentators declared it was the first social media election. Blogs and tweets would replace speeches. Orators, they were history. Lancaster points out that this complaint is really old. A Roman historian named Tacitus made the same argument in the year ninety-seven. Oratory is dying, he said, and he was also wrong. And the twenty-ten campaign proved it. Over five weeks, the three main candidates delivered fifteen hundred speeches, about a million words.
Britain held its first televised leaders' debate, and thirty million people watched, twelve times the audience of a popular reality show. The tweets everyone talked about, six hundred thousand were sent during the first debate. They were tweets about speeches. Careers rise and fall on this. In two thousand and five, David Cameron was a twenty-five to one outsider for the leadership of the Conservative Party.
He stood up at the party conference and spoke without notes. By the end of that day, he was the favorite. The frontrunner, who gave a limp performance at the same event, was out of the race. And the pattern isn't limited to politics. Richard Branson, Steve Jobs, and Anita Roddick all used speeches to pour their personalities into their companies.
Lancaster cites a two thousand and three study finding that half of a company's reputation flows from the reputation of its CEO. Half. Which brings us back to Mr. Ratner. Here's what I'd like you to take from this. You are already giving speeches. The Monday stand-up, the sprint review, the design critique, the two minutes you get in the all-hands, the moment after somebody says, "Does anyone have any questions?"
And the room goes quiet because they've stopped listening. In each of those moments, a person with an idea is trying to move a room. You may not call it a speech, but your audience certainly does. And just like Ratner's dinner, the audience is bigger than the people you see.
The skeptic in the corner repeats what you said to a colleague. The recording of your demo is forwarded to a VP. The comment you made in the meeting becomes the story people tell about you. So the question isn't whether you speak. The question is whether you speak on purpose.
[00:11:24] Aristotle Triangle Basics
---
Lancaster begins with Aristotle, who wrote the foundational text on rhetoric around 350 BC. And I find one thing about Aristotle comforting. He didn't invent any of this. He watched naturally eloquent people in ancient Greece, noticed what worked, and wrote it down. His idea was a triangle. Every persuasive speech needs three things working together: ethos, which is the speaker's credibility; pathos, which is the audience's emotion; and logos, which is the argument.
Lancaster has a neat way to remember it. An audience decides with its gut, its heart, and its head. Do I believe this person? Do I feel something? Does the story add up? We'll spend time on all three, but first, the part of the job that happens before any of them.
[00:12:23] Homework Know Audience
---
Lancaster's first golden rule is the one that opened the last episode. The audience matters more than the speaker. If you remember the Women's Institute and the slow hand clap, you know the cost of forgetting it. What I want to add is how a speechwriter actually does this homework, because it's more specific than know your audience. At the end of his research, Lancaster says the writer should be able to picture one imaginary person in the audience, a single human being.
How old are they? What car do they drive? What newspaper, what social media feed do they look at? What do they watch on television? That person sits on your shoulder through the whole writing process, and you keep asking, "Would they still get this? Would they roll their eyes?" If you build products, that probably sounds familiar.
It's a persona. But try applying it to the person you're about to talk to rather than the customer. Say you're pitching a new AI-assisted workflow to the principal engineer. She has watched three silver bullets come and go. She's seen a rewrite that took two years. She's not hostile. She's just been burned.
Give her a name in your head. Now write the first minute for her. And Lancaster tells a story about how hard this research is to do in real life. He once needed to interview a government minister who was just impossible to pin down. Sounds like most of the executives I worked for. So he lurked near the lift lobby.
When the minister appeared, he rode down with him and bombarded him with questions for the 22 seconds it takes to reach the ground floor. 22 seconds. I love that. It's a reminder that the most important work in communications happens before anyone gets on a stage. Walk to your skeptic's desk, ask one honest question, and listen to the answer.
[00:14:21] Set Aim Big Idea
---
The second piece of homework is the aim. Every speech, Lancaster says, needs one. Who exactly do you want to influence? And what change do you want in them? Do you want them to think something new, feel something different, or do something specific? Last episode, I called that a thin slice. calls it salami tactics.
Realistic aims are small. Nobody goes from no thanks to yes, transform everything in one speech. His best example is Geoffrey Howe's resignation speech in 1990, the one that helped end Margaret Thatcher's time in office. Howe's speechwriter said that every word in the speech was sharpened to inflict the deepest possible cut.
When the aim is that clear, the writing gets clear along with it. One more note, a single talk can have more than one audience. Look at your next design review. The engineer who wants to know it won't break the build. The program manager who wants to know whether the date moves. The executive who wants to know whether to worry.
One conversation, three audiences. Know which one you're talking to. The third piece of homework is the big idea. Every great speech has one Earl Spencer's eulogy for his sister, Diana was hunted. President Kennedy landed a man on the moon. One idea you could say to a stranger in a single sentence. And Lancaster has a lovely observation about where big ideas come from.
They're usually collisions between things that already exist. The iPhone combined technologies that were already on the shelf. Star Wars, he points out, was cowboys and Indians in space. Even Tony Blair's famous line, I have no reverse gear, was a twist on Margaret Thatcher's The Lady's Not for Turning, which was itself a twist on the title of a play.
He calls this steal and twist. And every product person has done it. Every pitch you've ever heard began with, it's like Uber but for, is steal and twist. Doing exactly what it's supposed to do. It borrows an idea the audience already holds and bends it to a new purpose. So before you say a word, you need three things.
A picture of one real person, an aim small enough to achieve, and an idea you can state in one breath. Now you're ready to open your mouth.
[00:17:05] First 20 Seconds Hook
---
So how do we get started? And what do we do in the first 20 seconds? Here's a fact about audiences that every speaker should be tattooing on the back of their hand. Your audience can leave at any moment. I don't mean they can walk out.
They're in the chair, camera's on, but their mind can be gone before you finish your first sentence, and there's nothing you can do to stop it. Lancaster puts it this way: "A speech is a series of signals. Some of those signals say, 'Wake up!' A dramatic pause, a change in your voice, a surprising statement, a story, a joke.
Others say, 'Go to sleep.' a flat monotone, slouched posture, platitudes, and above all, jargon." If you've ever heard someone say they plan to leverage synergies across the value chain, you know the signal he means. It's a lullaby. So the job is to send a clear wake-up signal at the start, and then another one every few minutes.
So let's talk about the start. Lancaster gives a handful of examples that I find really helpful because none of them are fancy. When Steve Jobs launched the iPhone, his first words were, "We're going to make some history together today." When Barack Obama gave the speech that made his career, he opened with, "They said this day would never come."
Edward Kennedy opened his eulogy for John F. Kennedy Jr. with a funny story about how the younger Kennedy had joked that if he were ever elected president, he would call his uncle to gloat. And Jeff Bezos, telling the origin story of Amazon, starts with a number. In 1994, he discovered that web usage was growing at 2,300% a year.
He asked himself what he should sell first, and he picked books. Look at the pattern. A promise, a challenge, a story, a number that makes you lean forward, and not one of them started their speech with, "Thank you so much for having me. A little background on the project." Don't do that. Lancaster also loves the power of silence.
In the middle of a Cash for Honors scandal, a fundraiser named Lloyd Levy had to give a speech to a room that very much knew about his troubles. He walked slowly to the lectern. Then he stood there. He paused for a full ten seconds. Ten seconds is an eternity at a microphone. The whole room was leaning in.
Then quietly, he acknowledged the scandal and moved on to stories about the people his work was meant to help. Try counting ten seconds in your head the next time you present. It will feel like a month, but it will work. Then there are stunts. Richard Branson launched a product by hologram. The chief executive, Carlos Ghosn, drove a car onto the stage.
You don't need a hologram, But remember John Stegner's glove shrine from a few podcasts back? The pile of four hundred and twenty-four work gloves on a boardroom table? That was a stunt. It was just a quiet one, and it ended a debate before it started. Physical object on the table is a wake-up signal. So is a printout of the form with nine signatures. So is a stopwatch.
[00:20:56] Keep Attention Rhythm
---
Now, keeping them awake. Remember the numbers from last time? Audiences can take in about seven hundred words a minute, and you speak at about a hundred and twenty.
All that spare brain power wanders off to check email or social media unless you give it something to do. Lancaster's advice is to pull them back every two or three minutes with something, a strong statistic, a story, a quote, a joke, a physical action. He compares it to the structure of the Beatles song Help.
The verses are quieter. The chorus is spectacular, and the chorus only feels spectacular because the verses came first. You're not trying to be at full volume for thirty minutes. You're trying to build a rhythm of valleys and peaks, and the biggest peak should be the ending. Churchill described it as a rapid succession of waves of sound and vivid pictures Building until the last words fall amid a thunder of assent. Most presentations in most companies end with the words, "So yeah, that's it. Any questions?" That's the opposite of a thunder assent. That's the sound of a balloon leaking air. Save your best line for last.
[00:22:18] Ethos Credibility Signals
---
Write it first if you have to So now let's get into the case itself, the gut, the heart, and the head. This is Aristotle's triangle at work, and I'd like to take the three corners the audience experiences them, the gut, the heart, and finally the head. Let's start with the gut, ethos, or believability. Lancaster makes the point that a speech is a window into the speaker's soul.
His example is almost funny. When the politician Robin Cook resigned from the British government, he opened his statement by remarking on how much better the view was from the back benches. He meant it as a pleasantry. What did the audience hear? They heard a man who enjoyed looking down on people.
That's ethos. You're never only saying what you say. You're always saying who you are. He catalogs the way speakers build credibility, and a handful of them are worth knowing. Association, connecting yourself to people your audience admires. And analogy, which reveals your thinking. Lancaster's example is great.
If you say, this is like the first meeting between Lenin and McCarthy, the audience hears someone with warmth and a sense of fun. If you say, this is like the first meeting between Watson and Crick, you sound like a person who values rigor. Same idea, two different people. Knowledge, and here's a surprising one. Gordon Brown, the former prime minister, was famous for blasting audiences with statistics. Lancaster's observation is that nobody remembered any of the numbers, but everybody remembered that Brown was a man of knowledge. The statistics weren't there to be remembered.
They were there to create an impression. So you don't need to share every figure. You need to leave a sense that you've mastered the material. Which brings us back to Mr. Ratner. And I want you to notice the asymmetry. Ethos takes years to build and can be destroyed in one sentence. And you can't take the sentence back.
Remember the Chicago study from last time where students judged an identical statement by who they thought said it? They listened to the messenger, not the message. That's why a team's best-liked skeptic, once converted, is worth ten of you.
[00:24:53] Pathos Emotion Priming
---
Now to the heart, pathos. Lancaster quotes the Roman orator Cicero, who said, "People make decisions far more by hate or love, or lust or rage, or sorrow or joy, or fear or illusion than by reality."
I'm not gonna repeat the lesson about matching the dominant emotion in the room. You'll remember the council that showed bar charts to a grieving town. What I want to add is the idea of priming, setting the mood before the speaker walks out.
Think about it. Why is the music loud before a keynote? Why do the lights go down? Lancaster says it's unchanged since Caesar. John Wesley used incense and candles. Steve Jobs used lasers and logos. But priming can go wrong. In nineteen ninety-two, Labour leader Neil Kinnock held a rockstar rally in Sheffield.
In the hall, it was electric. But to the people watching at home, it looked uncomfortably like the rallies of nineteen thirties Germany. Labour lost that election. Two audiences, one moment, the room and the recording. That's Ratner again. And we all live with it now. Every video call has a person in a living room on the other end.
Behavior that fills a conference hall can feel enormous on a laptop screen. Lancaster identifies six emotional appeals: hope, fear, pity, anger, pride, and shame. Fear is the one that deserves a warning He quotes a character in a film who described advertising as throwing a brick through someone's window, then selling them a burglar alarm.
Create a problem, inflate it, offer the solution. It works. It's also what every adopt AI or die keynote does. And as we said the last time, fear gets people moving, but it also freezes them. A workforce afraid for its jobs won't experiment. Pride is the underrated one. Lancaster notes that most people think they're above average.
In one study, eighty-three percent of Harvard students believed their performance was above average. Everybody thinks they're a better than average driver. Everybody thinks they're a better than average engineer. And that's an opening. Appeal to the pride in their craft. You're the person who spots the bug nobody else sees. Here's something that gives you more time to do that.
[00:27:34] Logos Premise Framing
---
Finally, the head, logos. And here Lancaster says something that sounds shocking at first. In a speech, logic is optional. Not because logic doesn't matter, because speeches move too fast for an audience to check for flaws. What audiences need is the illusion of logic, a clean structure, and a confident delivery.
I wanna be careful with that because I don't want you to hear that as permission to be sloppy or to mislead people. Lancaster spends a whole section cataloging fallacies like smears and false connections. He does that so you can recognize them in other people's speeches. He also warns that if an audience catches you using a trick, they stop trusting your logic and your character at the same time.
So treat this as a defensive skill first. With that said, a few things about argument are worth borrowing. First, the premise. Every argument starts from something the audience accepts. Johnnie Cochran, defending OJ Simpson, built his entire closing around if it doesn't fit, you must acquit. It's catchy.
It's also based on a premise that the glove didn't fit. That was contested. Premises can be powerful even when shaky. The strongest ones are premises everybody already agrees with. Lancaster points out that great American speeches follow the same pattern.
Start with, all men are created equal. Show the gap between that promise and reality. Then call for the change. Premise, gap, remedy. Gettysburg Address. I have a dream. You could use it for a pitch. We all want customers to get fixes sooner. Today it takes 11 weeks. Here's one way to close that gap. Same shape, much smaller stakes.
Second, framing. In 1941, Franklin Roosevelt wanted Americans to see a distant war as a threat to them. He began a radio address with, this is not a fireside chat on war. It's a talk on national security. One sentence moved the whole conversation. And labels frame before any evidence arrives. The British government once called an NHS review, the next stage review.
And the title itself implied the previous stage had failed. A different program was called Every Child Matters, which makes criticism sound heartless. So name your initiative carefully. Remember Friday's back from last time?
[00:30:20] Repetition That Lands
---
The name is the first argument you make. Third, repetition. Lancaster quotes a famous political communicator, Alastair Campbell, who said that a message only begins to work when you're sick of saying it. The first time, they barely hear it. The fifth time, they start to. That's a hard lesson for people who think their job is done after one email.
[00:30:44] Gut Heart Head Test
---
Gut, heart, head. Do I believe you? Do I feel it? Does it add up? Win all three and you hold the room. So now we're getting to the part that people usually think of as speech writing. The story, the metaphor, and the line people repeat afterwards.
[00:31:10] Make The Audience Hero
---
Let's start with stories. In the last episode, I gave you Robert McKee's definition: a protagonist overcoming forces of antagonism to reach something they desire. Hero, conflict, goal. Let's build on that with a few lessons from Lancaster that I found surprising.
First, who's the hero? In two thousand and eight, a British politician named David Davis resigned his seat in Parliament to force a by-election over a civil liberties issue. His speech was five hundred and fifty-eight words long, and Lancaster notices something clever. Davis didn't make himself the protagonist.
He made his constituency the protagonist, a place with a name that comes from a medieval phrase meaning noble endeavor. That sidestepped the charge of ego. And within those five hundred and fifty-eight words, Lancaster counted twelve separate conflicts, one every twenty seconds. Freedom against intrusion, duty against sacrifice.
For product people, the translation is immediate. The hero of your story is not your team, and it's not your tool. The hero is the customer or the patient or the colleague who used to leave at seven. You're the guide.
[00:32:39] Villains Without People
---
Second, choose your villain carefully. This is the part of the book that gave me pause.
Conflict gets attention. A common enemy unites an audience. But Lancaster is blunt about the damage when the enemy is a group of people. Margaret Thatcher called the trade unions the enemy within, and union membership fell by half during her time in office.
When a senior minister publicly criticized women who wear veils, Lancaster said he saw shortly thereafter a woman abuse a veiled stranger in a chip shop, something he had never witnessed there before. Words don't stay on the stage, so here's a rule: Make the villain the problem, not the people. Last time, our villain was the belts and pulleys.
Nobody had to be in the dock. The five signature approval isn't a person. Aim at the process. Nobody will defend it, and no one ends up humiliated.
[00:33:45] Show Dont Tell Stories
---
Third, show don't tell. There's a famous six-word story often attributed to Ernest Hemingway. "For sale: baby shoes, never worn." It says nothing about grief. It doesn't have to.
The audience writes the rest, and that's why it hurts. Remember Blair's maternity ward and the two babies from last time? Same technique. And Lancaster adds a warning I'd like every leader to hear: "Stories have to be real." Blair once told a story about a man he called Sierra Man, a self-employed electrician who'd switched his vote.
It worked beautifully until a newspaper went looking for the man and couldn't find him. The suggestion that Blair had invented him spread much farther than the story ever did. Remember Dana and claims from last time? Dana was a sketch, a placeholder to show the shape. Your Dana has to be a real person with her permission and her real Monday mornings.
Fourth, stories can carry a complex idea. Franklin Roosevelt wanted Congress to support aid to Britain. He didn't give a speech on the economics. He said it was like lending your neighbor a garden hose when his house is on fire. Complicated legislation one image. You know this technique. Someone in your industry once called it technical debt, and overnight, every non-engineer understood why you can't just keep shipping.
[00:35:23] Metaphors That Guide
---
Metaphors. Lancaster calls them the most powerful communication tool the speaker has. We use about six every minute without noticing. Politicians use one every twenty-three words. "And they work," he says, "like guided cruise missiles. They slip past the part of the mind that checks facts and land in the part that decides."
The most powerful business metaphor in history, Lancaster suggests, may be Adam Smith's invisible hand. It took a harsh-sounding idea that markets run on self-interest and made it feel gentler and almost benevolent. Here's the practical lesson for us. People reveal how they think through the metaphors they use.
Lancaster lists some of the common ones. Arguments are wars. Ideas are containers. People are machines. And he notes that a good doctor listens for them. If a patient says a migraine feels like a vice around their head, the doctor says, "Let's see if we can loosen that vice." So listen to the metaphors in your team's mouths when they talk about AI.
Is it going to eat my job? That's a predator. Is it an intern I have to supervise? That's, that's an apprentice. Is it a co-pilot? That's a partner. Autocomplete. That's a toy. Each one is a theory of the relationship. Mirror the metaphor first, and then gently offer a better one. It's pacing and leading again, now at the level of a single word One warning from Lancaster, metaphors can end up controlling you.
Margaret Thatcher became the Iron Lady and in the end said, "We will go on and on and on." George W. Bush's cowboy phrases like, "Bring it on," made him, by his own later admission, sound like not a man of peace. Be careful which image you adopt. You'll be living inside it. And keep it fresh, not paradigm shift, not low-hanging fruit. George Orwell's advice was never use a figure of speech that you're used to seeing in print.
[00:37:44] Soundbites And Rhetoric
---
And then there's the soundbite, the line that gets repeated. Lancaster is frank that the lines that seem effortless are anything but. John F. Kennedy's, "Ask not what your country can do for you," took him two years and dozens of versions.
The toolbox is ancient. The Greeks and Romans had names for all of it. A few that you can use this Monday, the rule of three. Example, education, education, education. Three feels complete.
Contrast, which we've discussed in its many forms. There's correctio, where you say what something is not before saying what it is. Churchill did this. "This is not the end. It's not even the beginning of the end. But it is perhaps the end of the beginning." There's reversal. "Government of the business is not the business of government."
The puzzle solution, which Churchill used for wit when the French predicted that England would have her neck wrung like a chicken. His reply, "Some chicken, some neck." There are four rules for using these. Work at them. They don't arrive by accident. Pick devices that sound natural coming from you. President Bush favored contrast, "Dead or alive," "With us or against us."
Gordon Brown favored alternative pairs, "Boom and bust." Make sure they serve the argument and use them sparingly. Overuse starts to sound like spin. Let me show you how this works, because it's easier to hear than to explain. Say you want your team to stop starting test plans from a blank page.
The standard version is, "We should leverage AI capabilities to optimize our test planning workflow efficiencies." That's the go to sleep signal. Now, try this for a contrast. "We don't need faster typists. We need fewer blank pages." Or a three, "Draft it, check it, ship it." I wrote those. They're allusions, and they're probably imperfect, but you can hear the difference. The first one gets forgotten. The others get repeated. And repetition, as we said, Is how a message begins to take on a life of its own.
[00:40:15] Drafting Without Friction
---
So how does anyone actually produce these things? Lancaster is refreshingly practical about process, and I think it's the part people will recognize most. Rule one, don't start writing too soon. Plan the shape first. Rule two, separate the creative mind from the critical mind. They use different parts of the brain, and when you mix them, you'll kill the thing you're making.
Lancaster's example is from John Lennon and Paul McCartney, who wrote A Hard Day's Night in about a half hour. Imagine if they had paused to debate whether it sounded better in the key of G or the key of A. The song probably would not exist. So write the first draft in isolation with no email, no internet, and don't edit as you go.
Lancaster likes to dictate his first drafts using voice recognition software because it forces the language to sound the way people actually speak. If you've ever had trouble writing a clear proposal, try talking it into your phone while you walk around the block. You'll be surprised how much clearer you are out loud.
I use WhisperFlow for this. And when you get stuck, he has a trick I love. Ask the questions the audience would ask. So what? Why is that? Who says? What's next? Answer them one after another, and you've got an argument. He also says messy is good. A messy first draft means you made something. Blank page means you didn't.
If the structure doesn't come naturally, Lancaster offers a few. The storyteller's setup, conflict, and resolution. Cicero's six-part structure. Behind about eighty percent of party conference speeches. Or the countdown built around a numbered list like three things. And a warning about the countdown that made me laugh.
Lancaster once drafted a statement for a minister with seven points. But during the editing, it was cut down to three. The minister got to the lectern and announced, "I will tell you seven things." There were only three. Only the intervention of the people who transcribed Parliament's official record spared him lasting embarrassment. So if you promise a number, count. Make sure it matches.
[00:42:46] Editing To Find Truth
---
Now the editing. This, Lancaster says, is where speeches are really made. His rule of thumb is that a five thousand-word draft, which is close to an hour, will shrink to about fifteen hundred after four or five edits.
Good copy, he says, has about seven distinct points per fifty words. Dull copy has three or four He keeps a file called The Cutting Room Floor, where everything he deletes goes. Because, he says, almost nothing he removes is ever missed. Give yourself time between writing and editing. Cut the repetition. Cut hesitations like, I think, and it seems.
Cut the opening throat clearing. Cut anything you can't pronounce. And sometimes writing exposes the truth. In 1968, Lyndon Johnson sat down to write a speech announcing an increase in troops to Vietnam. As he drafted it version after version, he realized he couldn't justify the policy. In the end, he announced instead that he was pulling back and that he would not seek re-election.
I think about that often when I work with teams on proposals. Writing the one-pager is not a formality. It's the test. If the sentence won't come, the idea may not be ready, and you're better off finding that out at your desk than in front of a vice president. Don't borrow other people's words without credit.
In 1988, Joe Biden's presidential campaign collapsed after he used, almost word for word, a passage from the British politician Neil Kinnock. Give credit away generously. As I said last time, it's safer and it's kinder. And be careful about committees. Lancaster quotes Theodore Sorensen, Kennedy's speechwriter.
Boldness decreases in proportion to the number of people who must clear a draft. Eisenhower said that 10 reviewers kill a speech's character. If you need feedback, and you do, gather the key people in a room at the same time and read it together rather than collecting 10 separate sets of comments.
Lancaster closes his chapter on this topic with George Orwell's rules, and the short version is this. Use short words. Cut every word you can. Prefer everyday language to jargon. Tape that above the monitor where you write your presetations.
[00:45:25] Deliver Like A Pro
---
So you've done the homework, you've built the case, you've cut it down to what matters. Now you have to stand up, and Lancaster's first point about performance is a comforting one. Great orators are made, not born. Demosthenes, the greatest orator of ancient Greece, began as a stuttering, stumbling speaker.
So he went away and trained. According to the story, he shaved his head to resist the temptation to leave, put pebbles in his mouth to work his jaw, and shouted at the ocean to build his volume. When he came back, the crowds were thunderous, and they were thunderous because of the time he'd spent working on his craft.
Reagan learned in Hollywood. Thatcher took voice lessons. "The nervous speaker who blinks too much and rushes through the words," Lancaster notes, "is often someone who got the job without ever serving an apprenticeship." Second principle, your body has to agree with your words. Audiences are astonishingly good at detecting the mismatch.
Lancaster's example is Gordon Brown again. He would speak of being a servant to the people while thumping the lectern. He'd talk of fairness and compassion while gesturing angrily, and his bitten nails and buried chin betrayed self-doubt as he claimed steadfast purpose. By contrast, Thatcher saying, "No, no, no," to European federalism looked completely sincere.
When what we see and what we hear don't match, we believe what we see and conclude the speaker is either being evasive or lying. Lancaster also cites a famous bit of research suggesting that when we judge someone's feelings, fifty-five percent, more than half, come from what we see, thirty-eight from the sound of their voice, so that's more than a third, and just seven from the words.
A caution from me, that research was about feelings and attitudes, not all communication, and it gets misquoted constantly as words don't matter. Words matter enormously. That's why this whole episode exists. But when the words and the face disagree, the face wins. Now, a practical question. Do you memorize your speech?
Lancaster's answer is that memorization shows conviction better than anything else. In ancient Rome, a speech read from a page wasn't considered worth listening to. He recommends a technique. Break the talk into chunks and attach a ridiculous image to each one. His example is a doctor with an enormous stethoscope standing on a crumbling Victorian school which collapses to reveal fields of yellow rapeseed with tanks rolling across them.
Health, education, agriculture, and defense were the topics. You'll never forget that image, and you'll never forget the order. If you can't memorize it all, memorize three things, your opening lines, your central argument, and your ending. Always read it aloud beforehand so you can catch the words you can't pronounce.
And then the physical side. Pick three friendly faces in the audience, one on the left, one in the middle, one on the right. Pick them out and rotate, then the whole room feels addressed. Pace. Fast speech signals nerves. Slow speech signals confidence. The last episode I mentioned that great orators speak at about a hundred words a minute.
Here's a vivid comparison from Lancaster. During the two thousand and eight Democratic nomination fight, Hillary Clinton was speaking at around hundred and sixty words a minute. Barack Obama, on the other hand, stayed at around ninety. One sounded like a person making a case. The other sounded like a person making history.
A lower pitch conveys authority, too. Thatcher deliberately lowered her voice and was ridiculed for it, but it worked. And clothing sends signals, too. Jobs' turtleneck. Gandhi's hand-spun cloth. When a peasant asked Gandhi what his message was, he answered, "I am the message." I told you that story last time, but here's why it keeps coming back.
The audience reads you before it hears you. Lancaster also assures us on one thing: Don't be afraid to go over the top. Audiences understand that drama is for their benefit. The caution, again, is the camera. What fills a hall can look excessive to those in a remote site on a screen.
Finally, logistics, because they ruin more talks than bad ideas do. Print the script in a large font. Never let a sentence break across pages. And never put stage directions in the text itself because speakers read them aloud. At the nineteen eighty-nine Brit Awards, two presenters' microphones were mixed up, one for a man who was six foot two and one for a woman who was five foot one, and it spoiled the whole evening.
Bill Clinton once found the wrong speech on the teleprompter during a State of the Union address, and he improvised for fifteen minutes. Not everyone recovers so well. Check your slides, your microphone, your last-minute edits. And speaking of slides, Lancaster is not a fan. He suggests avoiding PowerPoint where possible because it pulls the audience's eyes away from you.
He even notes that one investigation into the Columbia Space Shuttle disaster blamed, in part, a culture of PowerPoint thinking at NASA. Last time, I gave you a talk with no slides in ninety seconds. If you must use a slide, use one picture. And have a plan for a hostile room. Stop, address a heckler directly, or go to questions. As I said last time, people heckle because they don't feel heard. Let them be heard.
[00:51:47] Speech Strategy And Channels
---
There's a reason I saved this part for last. Before the ending, a single good talk almost never changes anything. The best speeches Lancaster describes were part of something much larger. His chapter on strategy begins with a claim I found liberating. Poor speeches are usually the result of poor planning, not poor writing.
When a speechwriter is asked to produce six speeches a week, he becomes a sausage machine, and quality collapses. The fix is to control the diary, build the speech strategy. It's a four-step plan, and I think it translates remarkably well to anyone running an initiative. Step one: look at the outside world honestly in the blunt tone of a tabloid newspaper.
No spin. Step two: set your priorities across messages, audiences, timing, and channels. Keep the messages few. Disney is making people happy. 3M's is to solve unsolved problems. Bill Clinton's nineteen ninety-two campaign had three headline messages.
Few messages repeated often. On channels, Lancaster says something very useful. A speech is not always the right tool. It's perfect for new arguments, for leadership, for setting out a vision, for floating an idea, and for reaching opinion formers informally. It's a poor way to announce a consultation.
That's better done with a webpage and a discussion board. Not everything is a presentation. Some things are a document, a Slack thread, or a recorded walkthrough. Step three: build a program. A simple grid of dates, events, and fifty-word sketches of each talk built around two or three big themes. Match the audience to the message and choose venues that reinforce it.
Blair launched his two thousand and one manifesto at a school to signal that education was his priority. Hold your pilot readout where the work is done Step four, put it into action and say no. Lancaster describes a pair of speechwriters at Toyota who built a four-question screening test for speaking invitations and accepted only the ones that scored well.
His own version is simpler. Accept an invitation only if it passes at least two of three tests. Does it advance the policy? Does it strengthen a relationship? Does it get the story to someone who matters? Run your calendar through that. If the lunch and learn does none of the three, send a colleague or send a recording.
[00:54:33] Media Timing And Clips
---
Which brings me to getting heard. Lancaster has a chapter on the media, and while he's writing about politicians in the press, the principles transfer to any organization where attention is scarce, which is pretty much all of them. He says the first thing that attracts attention is novelty. Announce something new.
And his joke about scale is that you may not be the biggest employer in the world, but you could be the biggest IT employer in the borough. Find the category where your result is the biggest, the fastest, the first, or the best. Then there's timing. Lancaster recommends what he calls the drip-drip approach.
Brief the key people in stages before and after the main event. A well-managed speech can generate days of coverage. In our world, in our world, pre-wire. Walk the three most influential skeptics through your idea before the all-hands. Nobody likes being surprised in public. One caution Lancaster gives, if you tell people ahead of time that you're going to deliver a stern rebuke, the audience arrives hostile.
Brief for curiosity, not for combat. And finally, he talks about what he calls word by mouth. In the ancient world, a speech spread by word of mouth. Today, it spreads by recording. Obama's podcast routinely drew seven million viewers. I Have a Dream has been watched nine million times on one video site by the time this book was written.
That cuts both ways. A typical speech by Gordon Brown drew a few hundred views. A video of him picking his nose in Parliament a million. So ask yourself, what is the equivalent in your organization?
What's the three-minute clip of your bright spot, the analyst showing how she gets her Mondays back in her own words, that can travel to the people you'll never meet in a hallway? That one clip may reach more people than you will in a year. Make sure it's good, and make sure she wants to be in it.
[00:56:44] Ethics And Final Takeaways
---
I want to finish with the part of Lancaster's book that I appreciate most. He doesn't look away from the dark side. In his introduction, he acknowledges that the same force that built movements against apartheid also enabled the Holocaust and the genocide in Rwanda. Oratory has no moral compass of its own.
The techniques we've covered today, the framing and the priming, the metaphors and the storytelling, are all tools. A hammer can build a house or break a window. In his epilogue, Lancaster steps back and makes a broader point. Language developed some eighty thousand years ago. Before that, people were isolated wanderers, but after it, they could coordinate, plan together, and form civilizations.
Every major advance since has depended on someone convincing others to try something new. Progress has always been, in some sense, a speech. And then he ends with a warning from Plato: Don't use the power of rhetoric against your friends. Think about who is on the other side of your conversations.
It's not a crowd of strangers, it's the people you'll sit next to tomorrow who will remember a year from now whether what you said turned out to be true. If you win the room by tricking people, you've lost the next room.
Credibility is the only currency you've got, which is where we came in. Gerald Ratner stood before an audience of a few hundred and lost half a billion pounds of money because he forgot who else was listening. He told a joke to the room, and he forgot about his customers. So let's bring this back to the hillside.
On that Washington State lawn, the first dancer who got up didn't win anyone over with a perfect speech. He didn't say a word. But look closely at what happened in the ninety seconds after the second person joined. The newcomer was welcomed as an equal. A third person saw it was safe, and then a crowd formed because each step made the next step easier.
Isn't that what everything in this episode is for? You're not trying to dazzle. You're trying to make it easier for the next person to stand up and join. So before you speak, picture one real person and what they need to hear. Open with something that wakes them up. Win their gut, their heart, and their head in that order.
Make it concrete enough to repeat, and cut it and cut it again till it's short. Say it like you mean it and run it like a campaign, not a one-off. None of it's magic. Demosthenes was a stutterer. Cameron was a twenty-five to one outsider. Every one of those skills is learnable, which is the best news in the book.
And if you're working with a team to learn how to use AI well, remember what the research said about where the real bottleneck is. It's not the technology, it's the people. And people change one conversation at a time. Make the next conversation a good one That's the show.
I'm Bob Bouthillier. This has been The InnoGuide Podcast on the book Speechwriting: The Expert Guide by Simon Lancaster.
I hope the insights from this book have been helpful, and I invite you to find more podcast episodes, articles, resources, and information about how I help companies develop products faster while learning to put AI into practice at innoguide.net. Again, I'm Bob Bouthillier, and this has been the InnoGuide Podcast.
Until next time, keep learning, keep questioning, and stay curious. I wanna thank Steve McIntire for producing this episode, and I especially wanna thank you for joining us. I look forward to seeing you next time. That's a wrap.