The iPhone, 2007
Say: "About four and a half minutes. Watch what he says first, and what he saves for last." Play the clip with the sound on. It stops by itself at 4:35.Cue: after it ends, wait a few seconds. Then: "Which part did you notice first?" Take two or three answers in chat.Book: Gallo uses this reveal as his road-map example (his Scene 5), and treats the last line as the model headline (Scene 4). For you only; the room does not need the book.Into next: "He did three things. Here they are."
Three Things He Did
He gave his Big Idea in one sentence. "Today, Apple is going to reinvent the phone." If they remember one line, it is that one. He gave his outline : three things. "A widescreen iPod, a revolutionary phone, a breakthrough internet communicator." Then the three became one: iPhone. The room knows the shape of what is coming. He named the problem . Smartphones are "not so smart and not so easy to use." Then a matrix, smart vs. easy to use, with the iPhone alone in the top corner. Now they want the fix.
Where in the clip (from the captions, give or take a couple of seconds). On screen the order is Big Idea, outline, problem; in the clip the outline comes first: outline: count 1:21, names 1:29 to 2:06, repeated 2:12 to 2:39, "Are you getting it?" 2:39, three become one 2:43 to 2:48, name 2:54 · Big Idea 2:59 to 3:07 · problem from 3:26, matrix right after it, before the clip stops at 4:35. Before the count he spends 1:17 on Mac 1984 and iPod 2001.Inside the outline, if you want to point them out: the count before the names, a power adjective on each product, the repeat, the question "Are you getting it?", three into one, the name said last.His words, for you: "a widescreen iPod with touch controls", "a revolutionary mobile phone", "a breakthrough internet communications device". The problem: smartphones are "not so smart and not so easy to use". The matrix: he calls it a "business school 101 graph".Say: "Three things. A Big Idea. An outline. A problem."Cue: in chat: "Which of the three is missing from your last presentation?" Read two out loud by first name.Into next: "The third one is where we start. The problem."
The Problem
A good idea can die unheard because the person with it can't show people what it's worth to them.
Chat
Say: "Think of an idea you had that deserved a yes and didn't get one." Read the problem line once. "The idea was fine. People walked past it, because nobody showed them what it was worth to them."Cue: ask them to type it in chat. Read two or three out loud by first name.Into next: "So if the idea is not the problem, what is?"
The Big Idea
What stands between a good idea and a yes is whether people care, and a talk is how you make them.
Say: the Big Idea, slowly, once. "People don't say yes to good ideas. They say yes to ideas they care about. The way you get people to care is a talk." Point at the picture: one person cares, then the next.Cue: pause after the line. Don't explain it yet.Into next: "Here is the best example I know."
Example: The iPod, 2001
Every music player was sold on its specs.
"A thousand songs" iPod introduction, 23 October 2001, 2:45 to 3:35. Open on YouTube
The benefit test: add "which means ___ for you" to any claim.
The spec 5 GB hard drive. Which means 1,000 songs in your pocket.
Say: "October 2001. Every MP3 player was sold on its specs. Watch what he says first." Play both clips.After: "He did not lead with 5 gigabytes. He led with 1,000 songs, in your pocket." Then the benefit test: "Add 'which means ___ for you' to any claim. If you can't fill the blank, it's a feature, not a reason."
Cue: in chat, "What is your 5 GB? The spec you lead with."Into next: "You might be thinking: fine for Steve Jobs. Here are the three doubts I hear."
3 Doubts
The doubt The answer Where
"My idea is good, so it should speak for itself." People say yes to what it is worth to them. Start from the listener. Stage 1 "I'm not a natural presenter." A talk is a set of decisions. Each stage is one decision anyone can make. Stages 2 to 5 "AI will make it sound generic." AI drafts only after you decide. You check every draft. Every stage
Say: name each doubt and its answer, one at a time, as each door opens. "The next hour proves each one."Cue: "Which of these three is yours? Type 1, 2, or 3."Into next: "All three answers come from one method. It has 5 stages."
5 Stages
1. Decide
You decide Who it is for, what should change, what it is worth to them
AI does Writes eight candidate lines to choose from
2. Shape
You decide Which three doubts the talk answers, in what order
AI does Plays a skeptical listener and lists the doubts
3. Prove
You decide What proof you have where they doubt you most
AI does Says what proof each doubt needs and what to ask
4. Make It Felt
You decide The headline and one moment
AI does Drafts plainer versions and three moments to pick from
5. Deliver
You decide How it opens, where chat is used, your one ask
AI does Marks up your script and writes recovery lines
The prompt for each stage is on the Workflow Card you take home.
Say: read the five names in order. "In every stage you make one decision, and AI does the heavy lifting after you decide. Not before. That is why it doesn't sound generic."Optional: "This talk was built with these five stages."
Into next: "Stage 1 is the one most people skip."
Stage 1: Decide
Write one line about the listener before slides, before a script.
Answer five questions first
1 Who is it for, and what do they believe now?2 What should change for them?3 What is it worth to them?4 What makes you credible?5 What can you remove?
The reaction test
Say your line to someone. Listen to the first thing they say.
What stands between a good idea and a yes is whether people care.
"What do you mean?" It works.
"That's nice." Rewrite it.
You decide Your answers to the five questions. What you keep is your Big Idea, the one line that is your headline.
AI does Writes eight candidate lines from your answers.
Check it Give ten votes across the eight. Then run the reaction test on the winner.
Say: "Before slides, before a script, write one line." Walk the five questions."Give AI your answers and ask for eight lines. You vote: ten points across the eight. Then say the winner to someone. 'What do you mean?' means you have it. 'That's nice' means rewrite."
Tie back: "That answers doubt one. You started from them, not from the idea."
Cue: optional live demo on one idea from the chat.Into next: "You have one line. Next, find what they doubt about it."
Stage 2: Shape
The three parts of your talk come from what they doubt.
Then put the whole talk in this order
1 Get their interest2 Name the problem3 Give the fix4 Say what it means for them5 Ask for one thing
You decide Which three doubts the talk answers, and in what order. What you keep is the outline of your talk: your Big Idea and three parts.
AI does Plays your most skeptical listener, lists every doubt, and groups them into three.
Check it Each part answers a real doubt. Add "which means ___ for you" to each one.
Say: "Ask AI to play your most skeptical listener and list every doubt. You sort them into three. Those three are the three parts of your talk." Order them by asking "what happens next?"The five steps for the whole talk. "This talk follows them. You saw the problem at minute one."
Cue: point back at the 3 Doubts. "I did this for today."Into next: "Each of the three needs proof. That is Stage 3."
Stage 3: Prove
A claim becomes real with a specific and something to compare it to.
The MacBook Air envelope Macworld, 15 January 2008, 0:40 to 4:45. Open on YouTube
The claim "It's thin." The proof Its thickest part is thinner than the other one's thinnest part.
Put proof where they doubt you most
Not after every claim. Let the person who vouches for you start with the doubt.
Weak "Max is great to work with."
Strong "I thought it would take weeks. It took one afternoon."
Pick the proof by the doubt
If they doubt Use Is that number big or small? A number beside something to compare it to Will it work for someone like me? A story from someone like them, opening with the doubt Does it really work? A short demo they can see Why should I trust you? Someone else vouches for you, on stage or side by side Is there a catch? Raise the objection yourself, then answer it with an example
You decide What proof you have where they doubt you most.
AI does Says what proof each doubt needs and the questions to ask. It never invents proof.
Check it Each doubt has one proof story from someone who lived it, and it opens with the doubt.
Say: before the clip, "Watch how he proves 'thin'." After: "He never says thin. He says the thickest part of his is thinner than the thinnest part of theirs. A specific, and something to compare it to."Point at the coffee: "$1,460 a year means nothing. $4 a day, one coffee, means something."
Proof: "Put it where they doubt you most. Let the testimonial open with the doubt." The two bubbles are examples, not real quotes.
Menu: read the doubt, then the proof that settles it. "Big or small" gets a comparison. "Like me" gets a story from someone like them. "Does it work" gets a demo. "Why you" gets a second voice. "A catch" gets the objection raised and answered.Cue: "Type one number you use. What could you compare it to?"Into next: "He also gave them a moment they still remember. That is Stage 4."
Stage 4: Make It Felt
Plan one moment, and say your line so a stranger can repeat it.
How to build a moment, like the envelope
1 Pick the one feeling you want in the room2 Make it something they can see: an object, a demo, a before and after3 Pause after it. Let it land4 Tie it back to your one line
Say your line three times, in the same words: at the start, in the middle, at the end.
The retell test
Can someone repeat it to a stranger in one sentence?
Tell it to your friend. One sentence.
"A talk is how you make people care about an idea." It works.
"It was about… presentations?" Cut it back.
You decide Your headline and one moment.
AI does Rewrites your draft as you would say it to a friend, gives three headlines and three moments with how each works.
Check it Someone can retell your line to a stranger and get it right.
Say: "The envelope is a moment. You can build one on purpose." Walk the four steps.Then the line: "Say it at the start, the middle, and the end, in the same words."
Retell test: "Ask someone to tell your line to a friend. If it comes back wrong, cut it back."
Cue: "What did I say at minute four?" Let them type the Big Idea back. That is the retell test on this talk.Into next: "Now you have to deliver it, and most of you will do that on Zoom."
Stage 5: Deliver
On a video call, use the chat early and often.
In the first two minutes, ask a question with no wrong answer.
Before you share your screen, say what you are about to show.
Write your line for when the screen share fails, before the call.
Put every link in the chat.
End with the room typing your three things back.
You decide How it opens, where you use chat, and your one ask.
AI does Marks up your script with chat questions and screen moments, and writes two lines for when something breaks.
Check it The room types your three things back. The close asks for one thing.
Say: the five moves. On the first: "I did that to you at minute one."Cue: if there is time, ask AI to mark up a short script live.Into next: "Let me do that last one now."
Recap
Chat
Show the three
People say yes to what it is worth to them.
A talk is a set of decisions.
AI does the heavy lifting after you decide.
This week: run Stage 1 on one real idea. Write one line. Say it to one person.
Open the Workflow Card
Say: the Big Idea once more, in the same words as minute four. Then "Type the three." Read back what they type, then click "Show the three."Cue: put the Workflow Card link in chat.Close: one next step. "This week, run Stage 1 on one real idea. One line. Say it to one person."
Persuasion Techniques
Seven ideas that make people care.
Start with the human result Say what changes in their life, not what the thing is. The power of 3 Three points are easy to follow and easy to remember. Make the cost of the old way visible Let them recognize the problem from their own experience. The five-step plan The order the whole talk follows. The benefit test Add "which means ___ for you" to any claim. Show your enthusiasm Say what you really feel, in everyday words. Story guide Find a story, give it one twist, test it.
Say: "Everything so far was the method. These seven are the ideas underneath it. You can use any of them on its own." Read the seven names, one line each.Cue: "Which one do you use already?" Take two or three answers in chat.Into next: "The first one is the one Jobs used on the iPod."
Start With The Human Result
People care about what changes in their life or work.
The spec A 5 GB hard drive. The human result 1,000 songs in your pocket. Your version What will they do, feel, or stop worrying about?
The question to ask
What am I really selling?
Not the thing. What it does to their day.
Say: "You saw this with the iPod. Every player listed its storage. He said a thousand songs in your pocket. Same drive, different sentence."Then the question: "What am I really selling? Not the thing. What it does to their day."
Cue: chat: "Finish this: my idea means ___ for them."Into next: "Once you know the result, keep the points to three."
The Power Of 3
Three points are easy to follow and easy to remember.
1 Goldilocks Three bears, not four. Sean D'Souza builds a whole talk the same way: what happens first, what happens next, how it ends.2 Steve Ballmer Opening the 2009 Consumer Electronics Show, he used groups of three at least five times. One was three words: simplicity, reliability, and speed.3 Miller's law In 1956 George Miller reported that people hold about seven things in short-term memory, give or take two. Later work is said to put it nearer three or four.
Cut to Three. Say them first. Tick them off as you go.
Say: "People remember about three things. So pick three." Walk the three examples quickly.Be straight about it: Miller's 1956 paper is about short-term memory, about seven items. The "three or four" figure is quoted without a named study. Three is a strong habit, not a proven limit. Do not say "science shows three."This talk is built on three: three doubts, three parts.
Cue: "How many points are in your talk now? Type the number."Into next: "Now the second idea, how to make the problem real."
The Cost Of The Old Way
A new idea matters more once people remember what the old way cost them.
The problem with the old phones Same video as the opening, about 5:03 to 6:02. Open on YouTube
Without it "Our phone has a better interface." With it Buttons that cannot change after the phone ships.
Say: before the clip, "Watch what he does before he shows anything of his." After: "He lists four phones people already know. He points at what is wrong with all of them. Buttons that are the same for every app, and you can't add one after it ships. Now you want the fix."The clip runs from about 5:03 to 6:02 of the same video. The times come from the captions, give or take a couple of seconds.
Cue: chat: "What did the old way cost your listener? Type it in their words."Into next: "Next, the order the whole talk follows."
Aristotle The Five-Step Plan
The order the whole talk follows.
Step This talk The iPhone opening 1. Get their interest The iPhone clip, then the question in chat The line of earlier products, then three become one 2. Name the problem The Problem Phones are not smart and not easy to use 3. Give the fix The Big Idea and the 5 Stages One leapfrog product: the iPhone 4. Say what it means for them The iPod example, the benefit test Fingers instead of a stylus 5. Ask for one thing The Recap's next step Not in the opening
Say: "Five steps. Interest, problem, fix, what it means for them, one ask." Then point down the middle column: "This talk followed them. You saw the problem at minute one."Then the right column: "He did it in the first few minutes of the iPhone."
Origin: the bust is Aristotle because Gallo credits this plan to him. I have not verified that, and it may be closer to Monroe's motivated sequence from the 1930s. If you say it, say "credited to Aristotle."Cue: "Which step is weakest in your talk?"Into next: "Step four is where this test lives."
The Benefit Test
Add "which means ___ for you" to any claim.
The feature A 5 GB hard drive. Which means 1,000 songs in your pocket. The feature A proximity sensor. Which means The screen turns off at your ear, so your cheek cannot press anything.
If you cannot fill the blank in their words, it is a feature, not a benefit.
Say: "Add 'which means ___ for you.' If you cannot fill the blank, you are describing the thing, not what it does for them."The second pair is Jobs's own explanation of the sensor in the long iPhone video: it turns the display off at your ear "so you don't get spurious inputs from your face."
Cue: chat: "Type one feature of yours. Add 'which means.'"Into next: "The words you use to say it matter too."
Show Your Enthusiasm
If you are excited about it, say so, in everyday words.
1 Use words you already use Jobs's favorites are watercooler words: amazing, incredible, gorgeous, awesome. He sounded the same on stage as off.2 Mean it Pick words that fit what you really think. "Amazing" about something you find ordinary sounds false. "I'm especially excited about this one" is honest.3 Let it carry Joe Vitale, in Hypnotic Writing, says excitement you feel while writing reaches the reader. Sean D'Souza calls selling a transfer of enthusiasm.
Be straight about it: Vitale and D'Souza are practitioners' reports. Neither is backed by a study.Say: "You are allowed to be excited. Jobs said 'amazing' and 'incredible' because he meant it. If you do not mean it, pick a different word."Vitale's picture: the person who tells a great story at a party freezes when handed a pen, and the written version turns flat. His advice, in short: write to share a feeling, not to impress.
Sean's own words are everyday ones too: "super-yummy," "gobsmacked," "rocket fuel," "boof." Say yours in your own words, not his.
Sean also says to match the amount of detail to how much you care, and gives no source for the "transfer of enthusiasm" line. Teach both as experience, not as research.
Jobs's own lines in the long iPhone video: "it works like magic," "it's gorgeous."
Cue: chat: "Type the word you really use when you love something."Into next: "Last one: how to find a story."
Story Guide
Find a story, give it one twist, test it.
1 Stock a bank One ordinary event a day, in one line. Mark where it went against what you expected.2 Fix one word Cut the idea to one word. Then match it to an event from your day.3 Pick common ground About 90 percent already known to them, 10 percent yours.4 Build it One small twist near the end. Check the sequence, the feeling, and the point.5 Test it Have someone repeat it back. Notice what drops out.
Say: read the five steps. Then one example: Superman on the floor in pain. Everyone expects kryptonite. The twist is chronic back pain. They keep the twist, and they can retell it.Credit: the five steps are a short version built from Sean D'Souza's story and analogy material. His 90/10 split is a rule of thumb, not data.
The worksheet and an AI prompt are on the Frameworks tab of the working doc.
Cue: chat: "What is one ordinary thing that happened to you this week?"
N hides notes. P opens notes in their own window. Arrows move.
