The Viral Script System

Use it when: you found a tool, feature or prompt that produces a visible result on screen. Best stage: Solution-aware (C) or Problem-aware (B). Sells: the comment keyword. Why it works: Schwartz's intensification: the product in action, the viewer in the scene. The screen does the convincing; you only narrate.

Speaking pace used for timing: about 2.5–3 words per second (a rough rule of thumb; time your own voice once).

---

The timing

Time	Block	What you say	What's on screen
0–2s	Hook	The result they want, or the problem they have. Not the tool name. ≤12 words.	The finished result already on screen + hook text (3–6 words).
2–5s	This is you	"If you [DO THE TASK] by hand, this is for you."	Your face or the "old way" (messy doc, 14 tabs).
5–8s	Name it fast	"It's called [TOOL]. [PRICE: free / $X]."	Logo or homepage, 1 second max.
8–20s	Show it	3–4 short beats: "Paste [INPUT]. Click [X]. That's it."	Screen recording, zoomed in. Cut every 2–3 seconds.
20–25s	Proof	One specific number: time, price, stars, result.	The number on screen, big.
25–30s	One CTA	"Comment [KEYWORD] and I'll send you [ASSET + LINK]."	Keyword in huge text.

Word budget: ~75–90 words total.

---

Fill-in script

[0–2s] [RESULT] in [TIME]. (or: Still [DOING TASK] by hand?) [2–5s] If you [DO TASK] every [PERIOD], this is for you. [5–8s] It's called [TOOL]. It's [FREE / $PRICE]. [8–20s] You [STEP 1]. You [STEP 2]. And [RESULT APPEARS]. Watch — [ONE DETAIL THAT SURPRISES]. [20–25s] [PROOF: what used to take X now takes Y / it costs Z / N people use it]. [25–30s] Comment [KEYWORD] and I'll send you the link plus [BONUS].

---

Fully written example

Topic: using a Claude Project as a "brand brain" so every draft sounds like you. (Claude Projects let you add files as project knowledge; check current plan limits in the app.)

Time	Voice-over	On-screen text	Visual
0–2s	"Your AI drafts sound like a robot. Fix it once."	AI that sounds like you	A finished LinkedIn post in your voice, highlighted
2–5s	"If you rewrite every AI draft by hand, this is for you."	Rewriting every draft?	You, scrolling a red-marked draft
5–8s	"It's a Claude Project. Ten minutes to set up, once."	Claude → Projects	Projects sidebar, 1 second
8–20s	"Drop in five posts you're proud of. Paste this one-line instruction. Now ask for a post about anything — it pulls your phrases, your length, your rhythm."	5 posts → 1 instruction → done	Upload files → paste instruction → type request → output streams in
20–25s	"My edits went from rewriting everything to fixing two lines."	Edits: 2 lines	Side-by-side: old draft vs new draft
25–30s	"Comment VOICE and I'll send you the exact instruction I paste."	Comment VOICE	Keyword huge, finger pointing down

Word count: 79 · Length: ~30s

The proof line is my experience; replace it with what you actually measure.

---

Mistakes that kill this format

• Opening on the tool's homepage. The first frame must be the result.
• Explaining features ("it has a knowledge base, a context window…"). Show one result instead.
• Two CTAs ("follow and comment"). Pick one.
• A 10-second login/setup scene. Cut straight to the moment it works.

Variations

• "Free vs paid": beat 5–8s becomes "It does what [PAID TOOL] does. Free."
• "Hidden feature": hook = "You use [TOOL]. You're not using [FEATURE]." (Product-aware viewers only.)

---

Made by Dan Shipped — @danshipped

Use it when: your niche repeats advice that's wrong, half-true or outdated. Best stage: Unaware (A) or Problem-aware (B). Sells: the follow (or a keyword for the "right way" asset). Why it works: it opens with a belief the viewer already holds, then builds a bridge from their side (Schwartz: never flatly contradict what they believe; agree first, then redirect). It also answers a saturated market: you're not adding a claim, you're correcting one.

---

The timing

Time	Block	What you say	What's on screen
0–2s	Hook: state the myth	"[COMMON BELIEF]." + "Mostly wrong."	Myth in quotes, crossed out on beat
2–6s	Agree first	"I believed it too. It makes sense, because [WHY IT FEELS TRUE]."	You, nodding / old result
6–14s	The turn	"But [WHAT ACTUALLY HAPPENS]. Here's what I saw: [OBSERVATION]."	Evidence: screen, data, side-by-side
14–26s	The right way	"Do this instead: [STEP 1], [STEP 2], [STEP 3]."	Each step as a 2–3s shot
26–31s	Proof	"[ONE SPECIFIC RESULT]."	Number big
31–35s	One CTA	Follow line or "Comment [KEYWORD] for [ASSET]."	CTA text

Word budget: ~85–100 words.

---

Fill-in script

[0–2s] "[MYTH]." Mostly wrong. [2–6s] I believed it too. It feels true, because [REASON]. [6–14s] But [WHAT REALLY HAPPENS]. When I [TESTED IT], [OBSERVATION]. [14–26s] Do this instead. One: [STEP]. Two: [STEP]. Three: [STEP]. [26–31s] [RESULT WITH A NUMBER]. [31–35s] Follow — I break one [NICHE] myth every [DAY]. (or keyword CTA)

---

Fully written example

Myth: "Longer prompts get better answers."

Time	Voice-over	On-screen text	Visual
0–2s	"'Longer prompts get better answers.' Mostly wrong."	Longer prompt = better	A 600-word prompt, scrolling
2–6s	"I believed it too. More detail should mean more control, right?"	More detail = more control?	You, shrugging
6–14s	"But long prompts bury the one thing that matters. The model tries to obey all of it, and the main ask gets diluted."	The main ask gets buried	Long prompt with the real task highlighted at line 38
14–26s	"Do this instead. One: put the task in the first line. Two: add one example of the output you want. Three: list only the rules you'd actually check."	1. Task first 2. One example 3. Rules you'd check	Rewritten 6-line prompt, built line by line
26–31s	"Six lines. Same model. I stopped rewriting the output."	6 lines. 0 rewrites.	Clean output next to messy output
31–35s	"Comment SHORT and I'll send you my 6-line prompt template."	Comment SHORT	Keyword

Word count: 89 · Length: ~33–35s

Honest note: long prompts are fine when you're giving reference material (docs, data). The myth is about instructions, not context. Say that in the caption; it earns trust.

---

Mistakes that kill this format

• Calling the viewer stupid for believing it. Agree first.
• Busting a myth nobody believes (no tension).
• Showing no evidence. "Trust me" isn't a turn.
• Busting a myth that's actually true sometimes, without saying when. Admit the exception.

---

Made by Dan Shipped — @danshipped

Use it when: you can show the same thing twice: ugly then good, slow then fast, expensive then free. Best stage: Problem-aware (B). Sells: the comment keyword. Why it works: the viewer does the math themselves. Two images side by side are proof that needs no adjectives. It's the shortest format here, so it's the best one to test hooks with.

---

The timing

Time	Block	What you say	What's on screen
0–2s	Hook: the after, first	"[ASSET] before. [ASSET] after [ONE ACTION]."	Split screen or fast flip: after → before
2–5s	This is you	"If your [ASSET] looks like the left, you're not alone."	The "before", zoomed on its worst part
5–17s	The one change	"The only thing I changed: [ACTION]. [WHY IT WORKS in one line]."	The change happening, 3 cuts
17–21s	Proof	"[MEASURED DIFFERENCE]."	Before/after numbers
21–25s	One CTA	"Comment [KEYWORD] and I'll send you [THE EXACT THING I USED]."	Keyword

Word budget: ~60–70 words.

---

Fill-in script

[0–2s] My [ASSET] before. My [ASSET] after [ONE ACTION]. [2–5s] If yours looks like the left one, you're not alone. [5–17s] The only thing I changed: [ACTION]. It works because [ONE-LINE REASON]. [SHOW STEP] — [SHOW STEP]. [17–21s] [OLD NUMBER] → [NEW NUMBER]. [21–25s] Comment [KEYWORD] and I'll send you the exact [PROMPT/TEMPLATE].

---

Fully written example

Topic: rewriting a landing-page headline with one Claude prompt.

Time	Voice-over	On-screen text	Visual
0–2s	"My landing page before. And after one prompt."	Before → after 1 prompt	Flip between two hero sections
2–5s	"If your headline says what your product is, you're not alone."	"The all-in-one platform for…"	Old headline, circled
5–17s	"I asked Claude one thing: rewrite this as the result my customer wants, in their words, under ten words. Then I pasted three real customer reviews so it could steal their phrases."	Result + their words + <10 words	Prompt typed → reviews pasted → 5 options appear
17–21s	"Old headline: 14 words, about me. New one: 8 words, about them."	14 words → 8 words	Word counts side by side
21–25s	"Comment HEADLINE and I'll send you the prompt."	Comment HEADLINE	Keyword

Word count: 71 · Length: ~25–27s

Notice the proof is word count, not conversion rate. Only claim a conversion lift if you measured one.

---

Mistakes that kill this format

• Showing "before" first for 5 seconds. Lead with the after (the desire), then flash back.
• Changing 6 things. One change, or the viewer can't copy it.
• A "before" that's a strawman nobody would make. It must look like their work.

---

Made by Dan Shipped — @danshipped

Use it when: there are several options and your viewer is tired of hype reviews. Best stage: Product-aware (D), or Solution-aware (C) if you lead with the task. Sells: the follow ("I test so you don't have to") or a keyword for the full results. Why it works: effort proof (you did the boring part) + deadly sincerity (you name the losers). In the breakout data, "I spent 100 hours…" did 8.4× and "I tested … on 15 real use cases" did 2.9×. The number of tests is the hook.

---

The timing

Time	Block	What you say	What's on screen
0–3s	Hook: the test	"I tested [NUMBER] [TOOLS] on [ONE REAL TASK]. Only [N] survived."	Grid of logos, most getting crossed out
3–7s	Why you care	"If you're about to pay for [CATEGORY], watch first."	Pricing pages flashing
7–12s	The rules	"Same input. Same [TIME LIMIT]. I judged on [CRITERIA 1] and [CRITERIA 2]."	The test input on screen
12–28s	Results	Fast verdicts, losers first: "[A]: [FLAW]. [B]: [FLAW]. [C]: [STRENGTH], but [CATCH]."	3–4s per tool, output shown
28–35s	The winner + proof	"[WINNER] — [SPECIFIC REASON]. And it's [PRICE]."	Winner's output, number big
35–40s	One CTA	"Comment [KEYWORD] for the full scorecard." (or follow line)	Keyword

Word budget: ~100–115 words.

---

Fill-in script

[0–3s] I tested [NUMBER] [TOOLS] on [TASK]. Only [N] survived. [3–7s] If you're about to pay for one, watch this first. [7–12s] Same [INPUT], same [LIMIT]. I judged [CRITERIA]. [12–28s] [TOOL A]: [ONE-LINE VERDICT]. [TOOL B]: [VERDICT]. [TOOL C]: [VERDICT]. [28–35s] The winner: [TOOL]. [REASON]. It costs [PRICE]. [35–40s] Comment [KEYWORD] and I'll send you the full scorecard.

---

Fully written example

Topic: AI meeting note-takers on one real sales call. The verdicts below are placeholders that show the shape. Run your own test and say what you actually saw — that's the whole point of this format.

Time	Voice-over	On-screen text	Visual
0–3s	"I tested 5 AI note-takers on the same sales call. Two survived."	5 tested. 2 survived.	5 logos, 3 get a red line
3–7s	"If you're about to pay for one, watch this first."	Before you pay	Checkout pages
7–12s	"Same 32-minute call. I judged two things: did it catch the next steps, and did it get the price right?"	Next steps? Price right?	Call recording waveform
12–28s	"[TOOL A] missed the follow-up date. [TOOL B] invented a discount we never discussed. [TOOL C] was perfect — but only in English."	❌ missed date ❌ invented discount ⚠️ English only	Each tool's summary, error highlighted
28–35s	"The winner caught all four next steps and the right price. Free plan covered it."	4/4 next steps · $0	Winner's summary, checkmarks
35–40s	"Comment NOTES and I'll send you the full scorecard and my test call script."	Comment NOTES	Keyword

Word count: 93 · Length: ~35–40s

Never fake a test result. If a tool won, say so even if it's a competitor of something you like. That's what makes people follow a tester.

---

Mistakes that kill this format

• Testing on a toy task. Use a real one your viewer does.
• No criteria. "I liked this one" isn't a test.
• Burying the winner. Verdicts are fast; the winner gets the most time.
• Hiding the price. Say it.

---

Made by Dan Shipped — @danshipped

Use it when: something happened to you (a mistake, a win, a weird moment) that carries one lesson. Best stage: Unaware (A). The best format for a cold feed. Sells: the follow. Why it works: a story doesn't look like an ad (Schwartz's camouflage), so the viewer's guard is down. The lesson does the selling. Follows come from who you are, and story is how you show it.

---

The timing

Time	Block	What you say	What's on screen
0–3s	Hook: start in the middle	The most tense moment, not the beginning. "[TIME] ago I [LOW POINT / SURPRISE]."	You, close, natural light. Or the artifact (the email, the dashboard).
3–10s	Context	"I was [SITUATION]. Like most [AUDIENCE], I [WHAT EVERYONE DOES]."	B-roll of the situation
10–22s	The turn	"Then [EVENT]. [WHAT I REALIZED]."	The moment itself: screenshot, message, number
22–34s	What I do now	"Now I [NEW BEHAVIOR]. [HOW, in one or two steps]."	Screen of the new way
34–40s	Proof / where it led	"[RESULT SINCE THEN]."	Number
40–45s	One CTA (follow)	"I share what [I'M BUILDING / LEARNING] every [DAY]. Follow if you're [IDENTITY]."	Your face

Word budget: ~110–125 words.

---

Fill-in script

[0–3s] [TIME] ago, [MOMENT OF TENSION]. [3–10s] I was [SITUATION]. Like most [AUDIENCE], I [COMMON BEHAVIOR]. [10–22s] Then [EVENT]. That's when I realized [LESSON IN ONE SENTENCE]. [22–34s] So now I [NEW BEHAVIOR]: [STEP], then [STEP]. [34–40s] Since then, [RESULT]. [40–45s] I'm [WHAT YOU DO, IN PUBLIC]. Follow if you're [IDENTITY].

---

Fully written example

Illustrative example in a build-in-public style (the details are made up to show the shape, they are not a real account). Use your own real story and numbers: a story with a fake detail collapses the moment one viewer checks.

Time	Voice-over	On-screen text	Visual
0–3s	"My first SaaS launch got four signups. Three were my friends."	4 signups. 3 were friends.	Dashboard showing "4"
3–10s	"I'd spent two months building features. Like most builders, I thought the product would sell itself."	2 months of features	Fast scroll through a feature list
10–22s	"Then a stranger emailed me one question: 'Does it work with my checkout?' I'd never asked anyone what they needed. I'd just built what I thought was cool."	"Does it work with my checkout?"	The email, highlighted
22–34s	"Now I do it backwards. I post the problem first, read the comments, and only build what people ask for twice."	Problem → comments → build	Comments being tagged in a spreadsheet
34–40s	"That's how my checkout product got its first paying merchants."	First paying merchants	Real payout or merchant list, blurred
40–45s	"I build my SaaS alone with AI, in public. Follow if you'd rather ship than plan."	Follow @yourhandle	Your face

Word count: 103 · Length: ~40–45s

---

Mistakes that kill this format

• Starting with "So, a few years ago…". Start at the tense moment.
• Three lessons. A story carries one.
• A hero story with no low point. The low point is what makes it believable.
• Ending on a moral ("and that's why you should never give up"). End on what you do now.

---

Made by Dan Shipped — @danshipped

Use it when: you want the pull of "3 things…" without becoming one more "5 AI tools" video. Best stage: Solution-aware (C). Sells: the comment keyword. Why it works: plain lists are the most saturated format in the AI niche (sophistication stage 4–5: viewers discount them). This format looks like a list in the hook, but each item is a rung of the same ladder: one system, three levels. It borrows the formula behind the biggest breakout in our data ("20 Real Examples From Useful to Almost Impossible", 12.8×): a range that makes people stay for the top rung.

---

The timing

Time	Block	What you say	What's on screen
0–2s	Hook: the ladder	"[NUMBER] levels of [SKILL]. Most people stop at [N]."	"LEVEL 1 / 2 / 3" stacked, 3 locked
2–5s	This is you	"If you [DO THE BASIC VERSION], you're at level 1."	Basic version on screen
5–12s	Level 1	"Level 1: [BASIC]. Fine, but [LIMIT]."	Fast demo
12–21s	Level 2	"Level 2: [ADD ONE THING]. Now [BETTER RESULT]."	Demo, the added piece highlighted
21–30s	Level 3	"Level 3: [THE MECHANISM]. This is where [BIG RESULT]." + proof	Demo + number
30–35s	One CTA	"Comment [KEYWORD] and I'll send you the level-3 [ASSET]."	Keyword

Rule: the last level gets the most time and the proof. The viewer stayed for it.

Word budget: ~85–100 words.

---

Fill-in script

[0–2s] [3] levels of [SKILL]. Most people stop at level [1]. [2–5s] If you [BASIC BEHAVIOR], that's level one. [5–12s] Level one: [BASIC]. It works, but [LIMIT]. [12–21s] Level two: [ADD ONE THING]. Now [IMPROVEMENT]. [21–30s] Level three: [MECHANISM]. [RESULT + PROOF NUMBER]. [30–35s] Comment [KEYWORD] and I'll send you the level-three [TEMPLATE/PROMPT].

---

Fully written example

Topic: three levels of using AI for short-video scripts. (This one sells this very lead magnet.)

Time	Voice-over	On-screen text	Visual
0–2s	"Three levels of writing scripts with AI. Most stop at one."	Level 1 · 2 · 3	Three locked boxes
2–5s	"If you type 'write me a viral script,' that's level one."	"write me a viral script"	That prompt in a chat
5–12s	"Level one: you get a script. It sounds like everyone else's, because it's built from everyone else's."	Sounds like everyone	Generic output, cliché phrases highlighted
12–21s	"Level two: you give it your audience and one desire. Now it's about someone."	+ audience + 1 desire	Same prompt with two lines added; better output
21–30s	"Level three: you make it diagnose first. What does the viewer already know? What have they seen a hundred times? Then it writes the hook from there. My hooks now score 80-plus before I record."	Diagnose → then write	Script doctor output: awareness stage, 5 hooks, scores
30–35s	"Comment VIRAL and I'll send you the level-three prompt, free."	Comment VIRAL	Keyword

Word count: 99 · Length: ~35–37s

"80-plus" refers to the hook scorer in this kit; use your own number once you've used it.

---

Mistakes that kill this format

• Three unrelated tips. If the levels don't build on each other, it's just a list.
• Equal time per level. Level 3 gets the most.
• Revealing level 3 in the hook. Name the ladder, not the top.

---

Made by Dan Shipped — @danshipped

Part of The Viral Script System · by Dan Shipped (@danshipped) · Last verified: October 2026

A hook has one job: make the viewer hear the next sentence. It doesn't sell. It doesn't explain. It opens a door.

Every hook below has two versions:

• a fill-in template with [BRACKETS], and
• a filled example for the AI / business niche.

Honest note on the numbers in the examples. They show the shape of a good hook. Replace every number with your own real one. A made-up number is the fastest way to lose a viewer's trust (and it's a lie). If you don't have a number yet, use a time ("in 4 minutes") or a count you can show on screen.

---

How to use this file (2 minutes)

1. Pick the stage. Who will see this video? Strangers on the For You feed → start in Stage A, B or C. Your followers → D or E. (Not sure? Read chapter 3 of the README.)
2. Pick 3 hooks of different types from that stage. Different types, not 3 versions of the same one.
3. Fill the brackets with your task, your number, your audience.
4. Score them in tools/hook-scorer.html. Keep anything 75+.
5. Say it out loud. If it takes more than ~4 seconds to say, cut words.
6. Put the same promise on screen in 3–6 words.

Placeholders used: [TASK] a job they do (gerund: "writing proposals") · [RESULT] what they want · [TIME] a duration · [NUMBER] a real count · [AUDIENCE] who it's for · [TOOL] a product name · [OLD WAY] how they do it now · [PRICE] a real price · [ASSET] the thing they make (video, email, page).

---

Stage A · Unaware (cold feed: they don't know they have the problem)

Where the hook starts: from them — their identity, their day, a story, a surprising observation. Never from the tool. Never from your name.

A1 · Call-out (name who it's for)

#	Template	Example (AI niche)
1	If you [DAILY TASK] every day, this is for you.	If you answer client emails every morning, this is for you.
2	[AUDIENCE] who still [OLD WAY], give me 20 seconds.	Freelancers who still write proposals from scratch, give me 20 seconds.
3	This is for the [AUDIENCE] who [SECRET STRUGGLE].	This is for the founder who rewrites the landing page every Sunday.
4	You're a [ROLE] with [NUMBER] tabs open. Watch this.	You're a solo founder with 14 tabs open. Watch this.
5	If your [ASSET] gets [BAD RESULT], it's not the [EXCUSE].	If your videos get 200 views, it's not the algorithm.

A2 · Identity (who they are, or want to be)

#	Template	Example
6	Most people [WEAK ACTION]. [GROUP] [STRONG ACTION].	Most people ask AI questions. Builders give it jobs.
7	The [AUDIENCE] who [WIN] all do one boring thing.	The freelancers who never run out of clients do one boring thing.
8	You're not bad at [SKILL]. You're [REAL CAUSE].	You're not bad at writing. You're writing for nobody.
9	Every [ROLE] I know who [RESULT] does this before 9am.	Every solo founder I know who ships weekly does this before 9am.
10	You don't need [WHAT THEY THINK]. You need [MECHANISM].	You don't need a team. You need three agents and a checklist.

A3 · Story open (start in the middle)

#	Template	Example
11	[TIME] ago I [LOW POINT]. Today [SPECIFIC WIN].	A year ago I had zero customers. Today merchants pay me monthly.
12	My [PERSON] asked me [QUESTION]. So I [ACTION].	My dad asked what I do all day. So I recorded it.
13	I let [TOOL/AI] [UNUSUAL ACTION]. It found [SURPRISE].	I let Claude read my 40 worst emails. It found the pattern.
14	The day I stopped [OLD HABIT], my [AREA] changed.	The day I stopped writing captions by hand, my Mondays changed.
15	Nobody tells you this about [GOAL] until [LATE MOMENT].	Nobody tells you this about your first SaaS until month three.

A4 · Paradox (two things that shouldn't fit)

#	Template	Example
16	The best [THING] I found this [PERIOD] is [FLAW].	The best AI tool I found this week is ugly. And beta.
17	I [DID LESS] and got [MORE].	I posted half as often and got more followers.
18	The more [GOOD THING], the less [RESULT].	The more AI tools you pay for, the less you ship.
19	[COMMON THING] is free. That's the problem.	ChatGPT is free. That's why your copy sounds like everyone's.

A5 · Belief challenge (flip what they assume)

#	Template	Example
20	You don't need [REQUIREMENT] to [GOAL] anymore.	You don't need to code to ship a SaaS anymore.
21	"[POPULAR ADVICE]" is why you're [STUCK].	"Post every day" is why your account isn't growing.
22	Stop [COMMON ACTION]. It's [HIDDEN COST].	Stop asking AI for ideas. It gives everyone the same ones.
23	Most of what you learned about [TOPIC] stopped working.	Most of what you learned about prompting stopped working.
24	[THING] isn't a [ASSUMED CATEGORY]. It's [REFRAME].	AI isn't a writing tool. It's an intern that never sleeps.

A6 · Observation (you noticed something they didn't)

#	Template	Example
25	I studied [NUMBER] [THINGS]. [NUMBER] had one thing in common.	I studied 77 breakout AI videos. 75 used 12 words or fewer.
26	Look at [FAMILIAR THING] again. Notice [DETAIL]?	Look at this week's top AI reels. Notice the first word?
27	Here's what [WINNERS] do in the first [TIME].	Here's what 3× creators do in the first second.

A7 · Future pace (show them the day after)

#	Template	Example
28	My bet: in [TIME], [AUDIENCE] who [ACTION] will [CONSEQUENCE].	My bet: in a year, freelancers without agents compete on price.
29	Imagine [DESIRED STATE] by [SPECIFIC TIME].	Imagine your week's content done by Monday at 10am.

A8 · Confession (deadly sincerity)

#	Template	Example
30	I'm embarrassed how long I [MISTAKE].	I'm embarrassed how long I wrote every caption by hand.
31	I wasted [AMOUNT] on [THING] so you don't have to.	I wasted $340 on AI subscriptions so you don't have to.
32	I was wrong about [TOPIC].	I was wrong about AI video editors.

A9 · Disqualify (push the wrong people away)

#	Template	Example
33	If you love [TEDIOUS TASK], skip this one.	If you love editing subtitles by hand, skip this one.
34	This isn't for [NON-TARGET]. It's for [TARGET].	This isn't for AI hobbyists. It's for people who bill clients.

A10 · Pattern interrupt & curiosity

#	Template	Example
35	Don't [EXPECTED ACTION] yet.	Don't post your next video yet.
36	This [SMALL THING] got [BIG RESULT].	This 9-word sentence got 400 people to comment.
37	There's a reason [EFFECT]. Almost nobody knows it.	There's a reason some 20-second videos get 50× more comments.
38	[NUMBER] words decide if anyone [RESULT].	Seven words decide if anyone watches your video.
39	Why do [UNEXPECTED WINNER] beat [EXPECTED WINNER]?	Why do ugly screen recordings beat polished AI videos?
40	What would you do with [TIME/MONEY] back every week?	What would you do with 6 hours back every week?

---

Stage B · Problem-aware (they feel the pain, don't know the fix)

Where the hook starts: from the problem, said the way they'd say it. Then promise there's a way out.

B1 · Name the problem

#	Template	Example
41	You have [PROBLEM NAME]. Here's the fix.	You have subscription bloat. Here's the cure.
42	It's called [NAME], and it's why [SYMPTOM].	It's called the blank-page tax. It's why posts take 2 hours.
43	[SYMPTOM]? That's not [WRONG CAUSE]. It's [REAL CAUSE].	Views but no followers? That's not luck. It's your last 4 seconds.
44	If [SYMPTOM], your [PART] is broken.	If people watch but never comment, your CTA is broken.

B2 · Pain (time, effort, frustration)

#	Template	Example
45	Spending [TIME] on [TASK] every [PERIOD]? Stop.	Spending 3 hours on slides every week? Stop.
46	[TASK] shouldn't take [CURRENT TIME].	A client proposal shouldn't take your whole afternoon.
47	Still [MANUAL TASK] in [YEAR]?	Still transcribing your own videos in 2026?
48	You [EFFORT]. You get [BAD RESULT]. Here's why.	You post every day. You get 300 views. Here's why.
49	You know that moment when [PAINFUL MOMENT]?	You know that moment when the client says "quick call?"
50	Ever [PAINFUL MOMENT]?	Ever spent an hour on a one-minute reply?
51	[BIG NUMBER] [UNITS] a year on [TASK]. I counted.	312 hours a year on email. I counted mine.

B3 · Mistake

#	Template	Example
52	The #1 mistake [AUDIENCE] make with [TOPIC].	The #1 mistake founders make with AI support bots.
53	[NUMBER] mistakes that make your [ASSET] look [BAD].	3 mistakes that make your AI images look AI.
54	You're using [TOOL] like [WRONG THING]. Here's the fix.	You're using ChatGPT like Google. Here's the fix.
55	If you start with "[BAD OPENER]," you've lost them.	If you start your video with "hey guys," you've lost them.
56	You're [DOING STEPS] in the wrong order.	You're writing the script before the hook. Flip it.
57	Your [ASSET] isn't bad. It's [SPECIFIC FLAW].	Your hook isn't bad. It's 23 words long.

B4 · Warning

#	Template	Example
58	Don't [ACTION] until you've seen this.	Don't pay for another AI tool until you've seen this.
59	Before you [BIG DECISION], watch this.	Before you hire a VA, watch this.
60	[COMMON ACTION] can quietly [HARM].	Pasting client data into free AI tools can quietly break your NDA.
61	Stop [WRONG FIX]. [RIGHT FIX] instead.	Stop buying better mics. Rewrite your first sentence instead.

B5 · Cost (show the bill)

#	Template	Example
62	[TASK] costs you about [AMOUNT] a [PERIOD]. Here's the math.	Manual invoicing costs you about 4 hours a month. Here's the math.
63	You're paying [PRICE] for something [ALTERNATIVE] does free.	You're paying $30 a month for something a free tool does.
64	Every [UNIT] you [ACTION] by hand is [LOSS].	Every reply you type by hand is a lead you answer late.
65	[COMPETITOR] [FAST]. You [SLOW].	Your competitors answer leads in 2 minutes. You take 2 days.

B6 · Question (one guaranteed "yes")

#	Template	Example
66	Who else is tired of [PAIN]?	Who else is tired of paying for 12 AI subscriptions?
67	Why does your [ASSET] [BAD RESULT] when theirs doesn't?	Why does your AI copy sound robotic when theirs doesn't?
68	Would you [SEND/POST] this? Most [AUDIENCE] do.	Would you send this cold email? Most freelancers do.

B7 · Before / after

#	Template	Example
69	[TASK] used to take me [OLD]. Now it takes [NEW].	Mondays used to be 3 hours of slides. Now: 3 minutes.
70	Left: [OLD]. Right: [NEW]. Same [INPUT].	Left: my prompt. Right: same prompt, one line added.
71	This is my [ASSET] before. This is after [ACTION].	This is my landing page before. This is after one prompt.
72	[PROBLEM]. Fixed in [TIME].	Messy client notes. Fixed in 40 seconds.

B8 · Hidden cause & myth

#	Template	Example
73	The real reason [PROBLEM] isn't [OBVIOUS CAUSE].	The real reason your reels flop isn't the edit.
74	"[COMMON BELIEF]." Mostly wrong. Watch.	"Longer prompts get better answers." Mostly wrong. Watch.
75	This is why your [ASSET] gets [A] but no [B].	This is why your reels get views but no follows.
76	[OLD WAY] was built for [OLD WORLD]. You're [NEW REALITY].	Content calendars were built for agencies. You're one person.
77	[GROUP] makes money when you think [BELIEF].	Course sellers make money when you think AI is hard.

B9 · Self-test

#	Template	Example
78	Try this: [QUICK TEST]. If [RESULT], you have [PROBLEM].	Read your hook out loud. Over 4 seconds? Rewrite it.
79	If you've ever [RELATABLE FAIL], keep watching.	If you've ever ghosted your own content calendar, keep watching.
80	Count your [THINGS]. More than [NUMBER]? This is for you.	Count your AI subscriptions. More than three? This is for you.

---

Stage C · Solution-aware (they want the result, don't know your way exists)

Where the hook starts: from the desire, plus "there's a way" — and as fast as possible, a hint of how (the mechanism).

C1 · There's a way / measured speed

#	Template	Example
81	There's a way to [RESULT] in [TIME].	There's a way to turn any link into slides in 30 seconds.
82	This turns [INPUT] into [OUTPUT] in [TIME].	This turns a 40-minute call into a proposal in 2 minutes.
83	[RESULT] in [NUMBER] clicks. No [OBSTACLE].	A full brand kit in 3 clicks. No designer.
84	[NUMBER] minutes. That's how long [TASK] takes now.	Four minutes. That's how long my weekly newsletter takes now.

C2 · Remove a limit / ease

#	Template	Example
85	[RESULT] — without [HATED REQUIREMENT].	Edit videos with AI — without learning a single editing app.
86	How to [RESULT] even if you [LIMITATION].	How to build an app even if you've never written code.
87	If you can [EASY THING], you can [RESULT].	If you can paste a link, you can build this.
88	One prompt. [RESULT].	One prompt. A month of LinkedIn posts in your voice.
89	Want [RESULT] without [COST]?	Want a site that converts without paying an agency?
90	Turn [COMMON INPUT] into [VALUABLE OUTPUT].	Turn your voice notes into finished blog posts.

C3 · Mechanism (the "how" — your best weapon in a crowded niche)

#	Template	Example
91	I built [SYSTEM] that [JOB] while I [OTHER].	I built an AI that reads 2,000 AI sources while I sleep.
92	The [NAMED METHOD]: [RESULT] in [NUMBER] steps.	The 3-tab method: research any market in 10 minutes.
93	Here's the system behind [RESULT].	Here's the system behind 30 videos a month, as one person.
94	[RESULT] = [A] + [B] + [C]. That's it.	A viral script = one desire + one proof + one ask.
95	I automated [TASK]. It runs every [PERIOD] without me.	I automated my client reports. They send every Friday without me.

C4 · How I / steal my

#	Template	Example
96	How I [RESULT] in [SMALL INPUT].	How I write a week of scripts in one 45-minute session.
97	The exact [ASSET] I use to [RESULT].	The exact prompt I use to turn comments into video ideas.
98	Steal my [ASSET] for [RESULT].	Steal my 4-line prompt for cold emails that get replies.
99	Copy this [ASSET]. Change [ONE THING]. Post it.	Copy this script template. Change the tool. Post it.
100	[RESULT]. Here's what I did differently.	47 replies from 100 cold emails. Here's what changed.

C5 · Shortcut / tutorial

#	Template	Example
101	The lazy way to [RESULT].	The lazy way to turn one video into ten posts.
102	Skip [PAINFUL STEP]. Do this instead.	Skip the content calendar. Do this every Sunday instead.
103	[RESULT] in [TIME]. Follow along.	A working landing page in 15 minutes. Follow along.
104	Watch me [RESULT] before this video ends.	Watch me build a client portal before this video ends.
105	This replaced my [EXPENSIVE THING].	This free tool replaced my $200-a-month editor.

C6 · Borrowed authority, levels & ranges (strongest families in the breakout data)

#	Template	Example
106	How [EXPERT ROLE] [DO RESULT] (just copy this).	How senior engineers review AI code (just copy this).
107	A [YEARS]-year [EXPERT]'s [PROCESS] in [TIME].	A 20-year copywriter's hook process in 30 seconds.
108	[GROUP] [RESULT] with one move you can copy.	Agencies write 50 ad variations with one move you can copy.
109	[SKILL]: level 1 vs level 10.	Prompting: level 1 vs level 10.
110	[NUMBER] ways to [RESULT], from easy to almost impossible.	7 ways to use AI agents, from easy to almost impossible.
111	[BIG NUMBER] hours of [SKILL] in [SMALL NUMBER] seconds.	200 hours of testing AI editors in 40 seconds.
112	Bet you can't tell which [ONE] is [AI/REAL].	Bet you can't tell which of these two decks AI made.

C7 · Proof-first

#	Template	Example
113	This [SHORT ASSET] got [RESULT]. Here's the structure.	This 27-second video got 1,200 comments. Here's the structure.
114	[SPECIFIC RESULT] from [SMALL INPUT]. Here's how.	9 booked calls from one carousel. Here's how.
115	Use this [FORMAT] and you'll never [PAIN] again.	Use this 5-line structure. Never face a blank script again.

---

Stage D · Product-aware (they know the tool or category, not convinced)

Where the hook starts: from the difference — superiority, comparison, a test, a verdict. Here, and only here (or when the tool is today's news), the tool name can go first.

D1 · Superiority & comparison

#	Template	Example
116	Everyone uses [TOOL A] for [TASK]. [TOOL B] does it [BETTER WAY].	Everyone uses ChatGPT for this. Claude does it with your files open.
117	[A] vs [B] for [TASK]. One clear winner.	ChatGPT vs Claude for client proposals. One clear winner.
118	It does what a [PRICE]/month tool does. For free.	It does what a $30/month tool does. For free.
119	[A] or [B]: which should you use in [YEAR]?	ChatGPT or Claude: which should you use in 2026?
120	Same prompt, [NUMBER] models. Look at number [N].	Same prompt, four models. Look at number three.
121	[TOOL] vs me doing it by hand. Timed.	AI vs me writing 10 captions by hand. Timed.

D2 · I tested it

#	Template	Example
122	I tested [NUMBER] [TOOLS] for [TASK]. Only [N] survived.	I tested 23 AI note-takers. Only 2 survived.
123	I spent [TIME] with [TOOL]. Honest verdict.	I spent 30 days with Cursor. Honest verdict.
124	I gave [NUMBER] [TOOLS] the same [TASK].	I gave three AI models the same messy spreadsheet.
125	Shipped or skipped: [TOOL] after [NUMBER] tests.	Shipped or skipped: this new AI browser after 10 tasks.
126	Is [TOOL/PLAN] worth [PRICE]? I paid so you don't have to.	Is the $20 AI plan worth it? I paid for six.

D3 · Hidden feature / better use

#	Template	Example
127	You use [TOOL]. You're not using [FEATURE].	You use Claude. You're not using Projects.
128	[NUMBER] [TOOL] settings I change before anything else.	3 ChatGPT settings I change before anything else.
129	[TOOL] for [UNEXPECTED USE]. Almost nobody does this.	Claude for prepping sales calls. Almost nobody does this.
130	[TOOL A] + [TOOL B] is the [ROLE] you can't afford to hire.	Claude + a spreadsheet is the analyst you can't afford to hire.

D4 · Honest verdict / switch

#	Template	Example
131	[TOOL] is overrated for [USE]. Great for [OTHER USE].	Midjourney is overrated for logos. Great for moodboards.
132	Why I switched from [A] to [B] after [TIME].	Why I switched from Notion to plain text after 4 years.
133	Don't upgrade to [PLAN] until you've tried this.	Don't upgrade to the $200 plan until you've tried this.
134	The free alternative to [PAID THING] nobody mentions.	The free alternative to paid transcription that runs offline.
135	I cancelled [NUMBER] subscriptions and replaced them with one.	I cancelled 5 subscriptions and replaced them with one.
136	[CATEGORY] isn't [BAD REP]. Your [INPUT] is.	AI writing isn't bland. Your prompt is.
137	Every [CATEGORY] tool, ranked for [AUDIENCE].	Every AI video editor, ranked for solo creators.
138	[TOOL] just changed [THING]. Here's what it means for you.	Your AI tool just changed its pricing. Here's what it means.
139	[TOOL] isn't built for [JOB]. This is.	Your notes app isn't built for 200-page PDFs. This is.
140	If you use [CATEGORY], try this before your next [ASSET].	If you edit in CapCut, try this before your next video.

---

Stage E · Most aware (followers, fans, warm viewers)

Where the hook starts: from the offer or the series. They know you. Be direct.

#	Type	Template	Example
141	Direct offer	Comment [KEYWORD] and I'll send you [ASSET].	Comment VIRAL and I'll send you all 155 hooks.
142	Offer + reason	[ASSET] is free. Here's why — and how to get it.	My script system is free. Here's why, and how to get it.
143	Drop	Drop #[N]: [RESULT].	Drop #014: the AI that writes SOPs from a screen recording.
144	Series	[SERIES], day [N]: [TODAY'S PROGRESS].	Shipped in 48h, day 2: payments are live.
145	Exclusivity	My [SYSTEM] found this at [TIME]. You're seeing it first.	My radar found this at 3am. You're seeing it first.
146	Request	You asked for [THING]. Here it is.	You asked for the hook template. Here it is.
147	Reply to comment	Replying to "[COMMENT]": [SHORT ANSWER].	Replying to "does this work for coaches?": yes. Watch.
148	Behind the scenes	Here's what [RESULT] looks like behind the scenes.	Here's what my paying merchants see behind the scenes.
149	Milestone	[NUMBER] of you [DID ACTION]. Here's what I learned.	800 of you commented SHIP. Here's what you asked most.
150	Teaser	Tomorrow: [NEXT PROMISE]. Today: [THIS PROMISE].	Tomorrow: the tool that replaced my editor. Today: the setup.
151	Experiment update	Update on [EXPERIMENT]: [PROGRESS].	Update on my AI-only content month: week 3 numbers.
152	Missing piece	If you saw [PREVIOUS VIDEO], this is the missing piece.	If you saw yesterday's video, this is the missing piece.
153	Participation	Drop your [THING] below and I'll [ACTION] tomorrow.	Drop your hook below. I'll score 10 of them tomorrow.
154	v2	Last time [NUMBER] of you asked for [MORE]. Here's v2.	Last time 400 of you asked for more hooks. Here's v2.
155	Advanced	If you already [USE ASSET], here's the advanced version.	If you already use my prompts, here's the advanced version.

---

Quick picker: which type for which goal?

You want…	Best stages	Best types
Reach strangers (growth)	A, B, C	Call-out, identity, observation, pain, before/after, mechanism
Comments (keyword)	B, C, E	Steal my, "there's a way", cost, direct offer
Follows	A, C	Identity, mechanism, series, teaser
Trust with skeptics	D	I tested it, honest verdict, confession
A crowded topic ("5 AI tools…")	C, A	Mechanism, identity, borrowed authority, ranges

The 5 rules every hook here follows

1. ≤12 words. 75 of the 77 breakout titles in our data were 12 words or fewer (see BREAKOUT_PATTERNS_77.md).
2. One thought. If you need "and" twice, it's two hooks.
3. Start from their desire or problem, not your tool or your name (unless the tool is the news).
4. A specific number or picture beats any adjective.
5. Open a gap, don't close it. The answer comes after the hook, never inside it.

---

Made by Dan Shipped — @danshipped

Part of The Viral Script System · by Dan Shipped (@danshipped) · Last verified: October 2026

My radar (Shipyard OS) watches the channels of 27 AI and business creators. Every time a video does far more views than that creator usually gets, it gets flagged as a breakout. Between July 5 and October 1, 2026 it flagged 77. Together they pulled about 33.5 million views.

The smallest breakout did 2.5× its creator's usual views. The biggest did 12.8×. The median breakout did 3.3×. Half of them (50) cleared 3×, and 16 cleared 5×.

I sorted every title into a formula and asked one question: which formulas show up most at the top?

Read this first (what this data can and can't tell you)

Deadly sincerity, before the patterns:

• These are YouTube titles, not Reels or TikToks. A title does the same job as a spoken hook (stop the scroll, open a gap), so the patterns transfer. The exact words don't. I translate each formula into a short-form hook below.
• There's no control group. I only have the winners, not the flops. So I can say "this formula shows up a lot among breakouts and has a high median multiplier". I can't say "this formula causes 4×". Treat it as strong hints, not laws.
• Small groups. Some formulas have 6–8 examples. One huge video can move a group's average, so I use the median (the middle video), which one outlier can't drag up.
• Channel effect. Creators with authority (big names, a known brand) get away with titles a newcomer can't. Keep that in mind on formula 3.
• Model and product names in the titles (GPT-6 Astra, Jev, Fable 5.1, etc.) are quoted from the titles as published. I'm not vouching for any claim inside them.

---

The 6 findings that matter

1. Short wins. Almost always.

75 of 77 titles are 12 words or fewer. The median is 9 words. The 16 biggest breakouts (5×+) have the same median: 9.

→ Rule: your spoken hook gets ≤12 words. Your on-screen text gets 3–6.

2. Numbers are everywhere.

47 of 77 (61%) contain a number: hours, dollars, a count, a version, a year.

→ Rule: one real number per hook. "100 hours", "$18K/month", "15 real use cases", "24 minutes".

3. "Copy the expert" and "who you become" have the highest medians.

Formula	Videos	Median ×	Hit 5×+	Best example
Borrowed authority ("X reveals his workflow", "just copy him")	6	4.85×	3 of 6	Matt Pocock's Agentic Engineering Workflow (just copy him) — 7.9×
Identity / who you become	6	4.65×	3 of 6	How To Become Dangerously Self-Educated (with AI) — 6.0×
It's here (arrival)	7	4.3×	2 of 7	This New AI Could Be the Biggest Breakthrough Since ChatGPT — 5.1×
Short verdict on the new thing	10	4.0×	2 of 10	GPT 6 Astra, so good even OpenAI are worried — 8.6×
Ranged list / levels	6	3.05×	2 of 6	GPT-6 Astra: 20 Real Examples From Useful to Almost Impossible — 12.8×
Compressed learning (time-box)	6	3.2×	1 of 6	1000+ hours of Learning Claude in 15 Minutes — 6.5×
Hands-on effort proof	8	2.95×	1 of 8	I Spent 100 Hours Using GPT-6 Astra — 8.4×
Money result	7	3.1×	1 of 7	This Insanely Simple App Makes $50K/Month — 5.0×
Stakes / warning	8	3.25×	0 of 8	GPT-6 Escaped. This Is Worse Than You Think — 4.2×
Comparison	3	2.9×	1 of 3	ChatGPT vs. Claude: Which Should You Use in 2026? — 5.2×
Contrarian	2	3.15×	0 of 2	FORGET Loop Engineering. Agentic Engineering is about THIS — 3.6×
Plain how-to / guide	8	2.75× (lowest)	0 of 8	Build Anything with Jev, Here's How — 4.2×

Why this fits Schwartz: in a crowded market (and "AI tools" is very crowded), a plain promise ("how to use X") is stage 1–2 sophistication. It's been said a thousand times. Authority gives a new reason to believe (someone credible does it), and identity speaks to who the viewer wants to be. Both are stage 4–5 answers.

4. "How to" alone is the weakest formula.

The 8 plain guides/tutorials have the lowest median (2.75×) and zero reached 5×. They still broke out (they're in this list), but they never broke out big.

→ Rule: never let "how to [use tool]" be the whole hook. Add a who (authority), a who-you-become (identity), a number (effort/time), or a range.

5. Fear gets a click, not a breakout.

The 8 stakes/warning titles ("…is getting ugly", "The End Is Near…", "Worse Than You Think") have a decent median (3.25×) but none reached 5×.

→ Rule: a warning hook works best as a door to a solution you show right away (Schwartz: attack only if you show the alternative). Pure doom doesn't build follows.

6. Ranges and effort numbers produce the biggest single spikes.

The #1 breakout (12.8×) is a ranged list: "20 Real Examples From Useful to Almost Impossible". The #3 (8.4×) is effort proof: "I Spent 100 Hours Using…". A range promises a ladder (the viewer wants to see the top rung). An effort number says "I did the boring part for you".

→ Rule: if you make a list, make it a ladder ("from easy to almost impossible", "level 1 to level 10"). If you tested something, say how much.

---

Two places the data argues with my rules (and what I do)

1. Intensity words. 16 titles use words like insane, freak, ridiculous, incredible, so good, and 6 of them are in the 5×+ group. On YouTube that works because it's a verdict from a creator the viewer already trusts, about news the viewer already cares about. On a short video to a cold stranger, the same word reads like an ad (Schwartz: adjectives the reader notices lower belief). What I do: replace the adjective with the evidence. Not "this AI is insane" → "it wrote 2,000 lines and none of them broke." If you must use one, use one, and prove it in the next 2 seconds.

2. Tool name first. 36 titles open with a product or model name. That breaks my "start from the desire" rule, except it doesn't: when a model launches, the whole market becomes product-aware overnight. People already know the name and are looking for a take. What I do: tool-first hooks only on news days (first 72 hours of a big launch) or for followers. Every other day, desire or problem first.

---

The formulas, translated to short-form hooks

Each one: the YouTube pattern → a template you can say in under 4 seconds → an example. (Numbers in examples are placeholders. Use your real ones.)

#	Formula	Short-form template	Example
1	Borrowed authority	How [EXPERT ROLE] [DOES RESULT]. Just copy this.	How senior engineers review AI code. Just copy this.
2	Borrowed authority	[EXPERT] showed me [PROCESS]. It's [SURPRISING TRAIT].	A 20-year copywriter showed me her hook process. It's 3 questions.
3	Identity	How to become the [IDENTITY] who [RESULT].	How to become the freelancer who never chases invoices.
4	Identity	You're not behind. Yet. Here's [FIRST STEP].	You're not behind on AI. Yet. Here's your first agent.
5	Identity	Once you [MILESTONE], [ACTION] these [NUMBER] things.	Once you land your first client, automate these 3 things.
6	Ranged list	[NUMBER] [THINGS], from [EASY] to almost impossible.	7 AI agent uses, from easy to almost impossible.
7	Levels	[NUMBER] levels of [SKILL]. Most people stop at [N].	5 levels of prompting. Most people stop at 2.
8	Compressed learning	[BIG NUMBER] hours of [SKILL] in [SMALL] seconds.	1,000 hours of using Claude, in 40 seconds.
9	Effort proof	I spent [TIME] [TESTING/USING] [THING]. Here's the verdict.	I spent 100 hours testing AI editors. Here's the verdict.
10	Effort proof	My [WORKFLOW] after [BIG NUMBER] [SESSIONS].	My content workflow after 400 videos.
11	Money result	This simple [THING] makes [REAL $]/month. Here's how.	This simple Notion template makes $3K a month. Here's how.
12	Price swap	[EXPENSIVE THING] for $0. Here's the catch.	My $20-a-month AI setup, for $0. Here's the catch.
13	Comparison	[A] vs [B]: which should you use in [YEAR]?	ChatGPT vs Claude: which should you use in 2026?
14	Verdict (news days)	[NEW THING] just [DID WHAT]. Here's what it means for you.	That new model just shipped. Here's what it means for you.
15	Contrarian	Forget [POPULAR THING]. [REAL THING] is about this.	Forget prompt tricks. Good AI output is about this.

Use the hook scorer (tools/hook-scorer.html) on your version. Anything under 70, rewrite.

---

How to rebuild this list for your niche (15 minutes a week)

1. Pick 10–20 creators in your niche (YouTube is the easiest place to get numbers).
2. For each one, look at their last ~15 videos and write down the middle view count (sort the numbers, take the middle). That's their "usual".
3. Any video at 3× their usual or more goes in your swipe file with its multiplier.
4. Every Friday, tag each title with a formula from the table above. After 4 weeks, you'll see which formulas your niche rewards.

The full manual method (YouTube, TikTok and Instagram) is in the README, chapter 10: "The outlier method".

---

Appendix: all 77 titles, by formula

Multiplier = the video's views ÷ the creator's usual (median) views, as computed by the Shipyard OS radar. Views are rounded as captured.

1 · Borrowed authority (6)

×	Title	Channel	Views
7.9×	Matt Pocock’s Agentic Engineering Workflow (just copy him)	David Ondrej	450,000
5.8×	Ex-NASA dev reveals his Agentic Engineering Workflow	David Ondrej	333,000
5.6×	Google Just Dropped a Masterclass on Agentic Engineering (It's SO Good)	Cole Medin	163,000
4.1×	Pi Agent dev reveals his Agentic Engineering Workflow	David Ondrej	236,000
2.8×	L8 Principal's Agentic Engineering Setup (just copy him)	David Ondrej	159,000
2.6×	Anthropic Engineer Explains: What to Build Instead of AI Agents	Nate Herk / AI Automation	349,000

2 · Identity / who you become (6)

×	Title	Channel	Views
7.7×	Install DLSS 5 EARLY & your Games will Look DISTURBINGLY Real.	MattVidPro	120,000
6.0×	How To Become Dangerously Self-Educated (with AI)	Dan Martell	1,200,000
5.7×	GPT-6 Astra will Make you Think BIGGER.	MattVidPro	88,000
3.6×	You’re Not Behind (Yet): How to Build Your First AI Agent (Full Guide)	Dan Martell	709,000
2.8×	If you use AI, switch to Omarchy immediately	Alex Finn	243,000
2.5×	Use Claude Better Than 99% of People (Beginner to Pro)	Futurepedia	253,000

3 · Short verdict on the new thing (10)

×	Title	Channel	Views
8.6×	GPT 6 Astra, so good even OpenAI are worried	AI Explained	853,000
6.5×	GPT 6 Astra is a freak	AI Search	1,500,000
4.8×	GPT-6 Astra Is Finally Here (And It’s REALLY Good)	Matt Wolfe	500,000
4.4×	Googles New Gemini 4 Argon is Now The Worlds Smartest AI	TheAIGRID	109,000
4.2×	Jev + Treg is a crazy combo for automation...	AI Jason	146,000
3.8×	Claude Opus 5.5 is ridiculous	AI Search	886,000
3.7×	Meta Muse is an INCREDIBLE AI agent	Alex Finn	324,000
3.4×	Deepseek just did the impossible	AI Search	787,000
3.0×	Claude Opus 5 is a freak	AI Search	691,000
2.9×	I Didn't Expect YuE2 to Sound THIS Good	MattVidPro	45,000

4 · It's here (arrival) (7)

×	Title	Channel	Views
5.1×	This New AI Could Be the Biggest Breakthrough Since ChatGPT - Jev	TheAIGRID	128,000
5.0×	Jev is HERE. How to use it	Greg Isenberg	744,000
4.3×	Claude Can Do So Much More For You Now...	The AI Advantage	132,000
4.3×	The Biggest ChatGPT Update Yet: Meet dots	Skill Leap AI	166,000
3.7×	Finally, an Open Standard for the Karpathy LLM Wiki is HERE	Cole Medin	108,000
3.3×	ASTRA IS HERE (GPT-6 RELEASED)	Matthew Berman	351,000
2.8×	OpenAI Just Revealed Dots… This Changes ChatGPT Forever	TheAIGRID	69,000

5 · Ranged list / levels (6)

×	Title	Channel	Views
12.8×	GPT-6 Astra: 20 Real Examples From Useful to Almost Impossible	The AI Advantage	389,000
7.0×	Once You Make Money, Upgrade These 7 Things ASAP	Dan Martell	1,400,000
3.2×	10 Levels of Jev For Agentic Engineers	IndyDevDan	114,000
2.9×	28 Insane Things Astra Can Do (2 Hour Course)	Riley Brown	259,000
2.6×	11 Tiny Coding Agent Fixes With A Stupid Amount Of Payoff	Cole Medin	76,000
2.5×	14 Insane Things ChatGPT Work Can Do (Automate Anything)	Riley Brown	225,000

6 · Compressed learning (time-box) (6)

×	Title	Channel	Views
6.5×	1000+ hours of Learning Claude in 15 Minutes (Beginner to Pro)	Dan Martell	1,300,000
3.5×	Obsidian in 24 Minutes	Tina Huang	706,000
3.3×	Every Size Local AI In 24 Minutes	Tina Huang	679,000
3.1×	How to Use New ChatGPT Work in 12 Minutes	The AI Advantage	94,000
3.0×	Hermes Agent Fundamentals In 29 Minutes	Tina Huang	604,000
2.6×	Learn 10X Faster In 10 Minutes	Tina Huang	532,000

7 · Hands-on effort proof (8)

×	Title	Channel	Views
8.4×	I Spent 100 Hours Using GPT-6 Astra (This Feels Like AGI)	Riley Brown	745,000
3.8×	I Turned GPT-6 Astra Into a 24/7 Stock Trader (tutorial)	Nate Herk / AI Automation	519,000
3.3×	GPT-6 Astra Made This Entire Video	Nate Herk / AI Automation	456,000
3.0×	Jev (Fully Tested) + Browser Use: FASTEST AI Agent I'VE TRIED YET!	AICodeKing	40,000
2.9×	I Tested GPT-6 Astra vs Fable 5.1 on 15 Real Use Cases	Nate Herk / AI Automation	395,000
2.8×	My Agentic Engineering Workflow (after 6,775 sessions)	David Ondrej	161,000
2.7×	ChatGPT Dots: How I Spent My First 24 Hours	The AI Advantage	81,000
2.6×	I Tested Jev on 12 Real Use Cases. My Honest Thoughts.	Nate Herk / AI Automation	354,000

8 · Money result (7)

×	Title	Channel	Views
5.0×	This Insanely Simple App Makes $50K/Month	Starter Story	385,000
3.6×	How This SaaS Hit $69K/Month In Just 2 Months	Starter Story	272,000
3.6×	Claude Design Now Builds Beautiful $10,000 Websites (NO AI Slop)	Jack Roberts	243,000
3.1×	Start a $10,000/mo Solo AI Creative Agency (Higgsfield + Claude)	Liam Ottley	129,000
3.0×	DeepSeek Harness = Claude Code for $0	Jack Roberts	204,000
2.9×	Build a $5,000 AI Datacenter at Home, Here’s How	David Ondrej	164,000
2.9×	This Simple App Makes $18K/Month	Starter Story	220,000

9 · Stakes / warning (8)

×	Title	Channel	Views
4.2×	GPT-6 Escaped. This Is Worse Than You Think	TheAIGRID	106,000
3.7×	We need to talk about Jev...	Matthew Berman	397,000
3.5×	Did OpenAI actually build AGI? GPT-6 Astra first look	Fireship	4,200,000
3.4×	The most important video you'll ever watch in the age of AI	Nick Saraev	213,000
3.1×	The End Is Near for Claude Code AI Operating Systems...	Liam Ottley	127,000
2.9×	Opus 5.5: How Close Are We to Automated AI Research?	AI Explained	286,000
2.8×	OpenAI's biggest math breakthrough is getting ugly...	Fireship	3,400,000
2.6×	Claude JUST found hidden DNA...	Wes Roth	331,000

10 · Comparison (3)

×	Title	Channel	Views
5.2×	ChatGPT vs. Claude: Which Should You Use in 2026?	The AI Advantage	160,000
2.9×	GPT-6 Astra (Fully Tested & Side by Side comparison with Fable 5.1): ONE is a CLEAR WINNER!	AICodeKing	39,000
2.9×	Hands on: Fable 5 makes GPT 5.5 feel like a "toy"	MattVidPro	45,000

11 · Contrarian (2)

×	Title	Channel	Views
3.6×	FORGET Loop Engineering. Agentic Engineering is about THIS	IndyDevDan	130,000
2.7×	I'm Obsessed With Local AI. Here's Why	Greg Isenberg	401,000

12 · Plain how-to / guide (8)

×	Title	Channel	Views
4.2×	Build Anything with Jev, Here’s How	David Ondrej	240,000
3.8×	Building a GPT-6 Kalshi AI Trading Bot From Scratch (full guide)	All About AI	23,000
3.1×	Ultimate Guide To Claude Skills	Skill Leap AI	118,000
2.8×	Ultimate Guide To ChatGPT Codex for Everyday People	Skill Leap AI	106,000
2.7×	How to Use ChatGPT Sites to Build Anything (Insane)	Riley Brown	243,000
2.7×	Build Anything with Kimi K3, Here’s How	David Ondrej	152,000
2.7×	Agentic AI Trading For Beginners: A New Money Making Era Is Here	All About AI	16,000
2.5×	CLAUDE CODE MARKETING FULL COURSE (6 HOURS)	Nick Saraev	158,000

---

Made by Dan Shipped — @danshipped

Copy everything inside the box into a new chat. Fill the part at the bottom. Send. Works best on a strong model. If the answer gets cut off, type "continue".

---

You are a direct-response script editor for short-form video (Reels, TikTok, YouTube Shorts). You use Eugene Schwartz's principles: one desire per video, the viewer's awareness stage decides where the hook starts, and market sophistication decides whether you lead with a claim, a mechanism (the "how") or the viewer's identity. You are blunt, specific and kind. No padding.

HARD RULES
1. Never invent proof. If I didn't give a number, result, price or credential, write a placeholder like [PROOF: your real minutes saved]. No fake stats, testimonials or claims about real companies.
2. One desire, one idea, one CTA per video. If my draft has two ideas, keep the stronger one and list the other under "Next video".
3. Hooks: 12 words max, one thought, starting from the viewer's desire or problem, not the tool name or my name (exception: product-aware viewers, or the tool is today's big news).
4. Banned words: insane, crazy, mind-blowing, game-changer, revolutionary, unlock, supercharge, next-level, incredible, unbelievable, ultimate, delve. Replace with a number or a picture.
5. Show, don't explain: the middle of the script describes the result happening on screen.
6. Plain US English, short spoken sentences.
7. Script length 25–45 seconds at about 2.5–3 spoken words per second (65–130 words).
If something is missing from my input, don't ask: assume, state the assumption in one line, and continue.

STEP 1 — DIAGNOSIS (short)
- The one desire, in the viewer's words.
- Awareness stage of who'll see it, and where the hook must start:
  Unaware (cold feed) → identity, story or observation · Problem-aware → the problem · Solution-aware → the desire + "there's a way" + a hint of how · Product-aware → the difference (test, comparison, verdict) · Most aware (followers) → the offer or the series.
- Market sophistication: stage 1–2 (fresh) → direct claim with a number · stage 3–4 (crowded) → lead with the mechanism, named and specific · stage 5 (exhausted, e.g. "5 AI tools") → lead with identity.
- One-line verdict on my draft's angle.

STEP 2 — SCORE MY HOOK (0–100) with this rubric. Show a table: criterion, points, one-line reason. Then total and band (0–49 Rewrite, 50–69 Needs work, 70–84 Good, 85–100 Ship it).
1. Length (15): 4–12 words = 15 · 1–3 words = 7 · 13–16 = 6 · 17+ = 0.
2. Specificity (15): a real number/price/%/number word = 15 · a named thing but no number = 10 · neither = 5 · −5 for vague words (stuff, a lot, so much).
3. Speaks to the viewer (10): "you/your", a named audience, or a direct instruction = 10 · only "I/my" = 5 · no person = 0.
4. Curiosity gap (15): 6 points per type of signal, max 15. Types: pointer ("this", "here's", "look at"), question (why/how/what or "?"), contrast ("but", "not", "until", "nobody", "stop", "before/after", "from X to Y", "vs"), open ending or formula (":", "=", "…").
5. No hype (15): start at 15, −8 per hype word, −3 for 2+ ALL-CAPS words (acronyms excluded) or "!!".
6. Starts from desire/problem (15): first 4 words carry the viewer's outcome or pain = 15 · neutral = 10 · tool talk first = 7 · filler opener ("Hey guys", "So", "In this video") = 0.
7. One clear thought (15): start at 15, −5 for 2+ "and", −5 for 3+ commas, −5 if average word length is over 6.5 letters.

STEP 3 — FIVE NEW HOOKS of five different types (e.g. call-out, before/after, mechanism, I-tested, belief challenge), at the right awareness stage. For each: hook · type · word count · estimated score. Mark your top pick with ★ and say why in one line.

STEP 4 — TIGHTENED SCRIPT using the ★ hook, as a table: Time | Voice-over | On-screen text | Visual.
Structure: 0–2s Hook (make them hear the next line) · 2–6s "This is you" (a guaranteed "that's me") · 6–20s Show it (result happening on screen, 2–3s per beat) · 20–26s Proof or how (one specific number or one line of mechanism, where they'd doubt) · last 4–6s One CTA.
Then: word count, estimated length, and which format it is (tool demo, myth-buster, before/after, I tested X, story, list-that-isn't-a-list).

STEP 5 — ON-SCREEN TEXT: hook overlay (3–6 words, same promise), 3–5 beat captions, CTA overlay, and a post caption (≤150 characters, first line restates the promise, ends with the CTA).

STEP 6 — THE CTA, written in full. Keyword CTA must have: the asset with a specific number, extreme ease ("just comment X"), a reason it's free, one keyword only (one word, caps, easy to spell). Follow CTA: promise of the next video + identity ("Follow — tomorrow: [NEXT]. If you'd rather ship than scroll.").

STEP 7 — PRE-PUBLISH CHECK. ✅/❌ for each, with a short reason for any ❌:
1. One desire, sayable in one sentence? 2. Hook starts where this viewer is? 3. Crowded topic → mechanism or identity, not the same claim? 4. Hook only makes them want the next sentence, ≤12 words, answer not given away? 5. "That's me" by second 6? 6. Result shown, not explained? 7. One true, specific proof where they'd doubt? 8. No ad adjectives? 9. Exactly one CTA, with a reason it's free? 10. Only I could post this (my test, numbers, story, system)?

End with "Fix this first:" (the single highest-impact change, one sentence) and "Next video:" (any idea you cut).

Use these headings: Assumptions · 1. Diagnosis · 2. Hook score · 3. Five new hooks · 4. Tightened script · 5. On-screen text · 6. CTA · 7. Pre-publish check · Fix this first · Next video.

MY INPUT
Niche / audience: [e.g. freelancers who use AI]
Platform: [Reels / TikTok / Shorts]
Goal: [follow OR comment keyword: KEYWORD]
Proof I actually have: [numbers, results, screenshots — or "none yet"]
Idea or draft script:
[PASTE HERE]

---

Example input (to see it work)

Niche / audience: freelance designers
Platform: Reels
Goal: comment keyword: BRIEF
Proof I actually have: my briefs used to take ~45 min, now ~5 min
Idea or draft script:
Hey guys! Today I want to show you an amazing ChatGPT trick that will totally change the way you write client briefs. So basically you open ChatGPT and you paste your call notes and then you ask it to make a brief and it's so good. Follow me for more tips and comment BRIEF for the prompt!

What you should get back: a hook score around 40 (filler opener, hype, tool-first, two CTAs), five rewrites like "Client briefs shouldn't take 45 minutes. Mine take 5.", a 30-second script, and "Fix this first: cut the follow ask — one CTA (BRIEF)."

---

Made by Dan Shipped — @danshipped

The same rubric used by tools/hook-scorer.html, so scores from Claude and from the tool are comparable. It's a set of heuristics, not a prediction of views. A 90 hook on a weak video still loses.

#	Criterion	Points	How it's scored
1	Length	15	4–12 words = 15 (sayable in ~4 seconds) · 1–3 words = 7 · 13–16 words = 6 · 17+ words = 0
2	Specificity	15	Contains a real number, price, %, or number word ("seven", "hundred") = 15 · a named thing (a product, a role) but no number = 10 · neither = 5 · −5 if it leans on vague words ("stuff", "things", "a lot", "so much")
3	Speaks to the viewer	10	"You / your", a named audience ("freelancers", "founders", "people"), or a direct instruction ("Stop…", "Copy this…") = 10 · only "I / my" = 5 · no person at all = 0
4	Curiosity gap	15	6 points per type of gap signal, max 15. Signal types: a pointer ("this", "here's", "there's a", "the exact", "look at", "watch"), a question or why/how/what, a contrast ("but", "not", "instead", "until", "nobody", "stop", "before/after", "used to… now", "from X to Y", "vs"), an open ending or formula (":", "=", "…")
5	No hype	15	Start at 15 · −8 per hype word (insane, crazy, amazing, mind-blowing, game-changer, revolutionary, unlock, supercharge, incredible, unbelievable, ultimate, next-level, epic, magic, hack, literally, "really cool", "you won't believe") · −3 for 2+ ALL-CAPS words (acronyms like AI, SaaS, API excluded) or "!!"
6	Starts from desire/problem	15	First 4 words carry the viewer's outcome or pain (you, if, stop, still, tired, hours, money, clients, views, without, behind, mistake…) = 15 · neutral start = 10 · first words are tool talk (a product name, "this AI tool", "new app") = 7 · filler opener ("Hey guys", "So", "In this video", "Today I'm going to") = 0
7	One clear thought	15	Start at 15 · −5 if "and" appears 2+ times · −5 for 3+ commas · −5 if average word length is over 6.5 letters (jargon)

Bands: 0–49 Rewrite · 50–69 Needs work · 70–84 Good · 85–100 Ship it.

Exceptions you're allowed to make (say so when you do)

• Tool-first is fine for product-aware viewers or in the first ~72 hours of a big launch, when the tool name is the news. Then criterion 6 scores on whether the second half names the viewer's stake.
• A one-word intensity verdict ("This is ridiculous.") can work for a creator the audience already trusts, if the proof appears within 2 seconds. Still costs points here, on purpose.

Worked example

Hook: "This insane new AI tool will completely change how you work!!" (11 words)

Criterion	Score	Why
Length	15	11 words
Specificity	5	No number, no named result
Viewer	10	"you"
Curiosity	12	Two signals ("This", "how"), but the promise is vague
No hype	4	"insane" (−8), "!!" (−3)
Desire first	7	Opens with tool talk ("This insane new AI tool")
One thought	15	One idea
Total	68	Needs work

Rewrite: "Your client proposals take 2 hours. This makes them 2 minutes." → 90+.

Print it, or keep it next to your editor. Two or more "no" → fix before you post.

#	Question	Yes / No	If no, do this
1	One desire. Can I say the single desire this video channels in one sentence, in the viewer's words?	☐	Cut every line that serves a second desire. Save it for the next video.
2	Right starting point. Does the hook start where this viewer is? (Cold → identity/story · Problem-aware → the problem · Solution-aware → desire + "there's a way" · Product-aware → the difference · Followers → the offer)	☐	Pick the stage, then pick a hook from that section of hooks/150_HOOKS.md.
3	Crowded topic? If they've seen 100 videos like this, am I giving the how (mechanism) or speaking to who they are (identity)?	☐	Add one line of mechanism ("it reads the transcript, then…") or an identity line ("for people who bill clients").
4	Hook does one job. ≤12 words, one thought, makes them want the next sentence, doesn't give the answer away?	☐	Run it through tools/hook-scorer.html. Below 70 → rewrite.
5	"That's me" by second 6. Will the right viewer recognize their task, their moment, their words?	☐	Add: "If you [their exact task] every [period], this is for you."
6	Shown, not explained. Is the result happening on screen?	☐	Replace the explanation with a screen recording of the result.
7	One true proof, right place. A number, price, time or screenshot exactly where they'd think "is this real?"	☐	Add one real number around second 20. If you don't have one, measure before posting.
8	No ad words. Zero insane / game-changer / unlock / mind-blowing?	☐	Replace each with the evidence.
9	One CTA. Follow OR comment the keyword, never both, with a reason it's free?	☐	Delete the weaker CTA. Add "free because…".
10	Only I could post this. If I swapped my name for another creator's, would something be lost?	☐	Add your test, your number, your story or your system.

Bonus: the 4 production checks

• The first frame shows the result (not your logo, not a title card, not you saying hi).
• On-screen hook text = same promise as the spoken hook, 3–6 words, readable on a phone.
• Captions on (many people watch on mute).
• The keyword is spelled the same in the video, the caption and your DM automation.

---

Made by Dan Shipped — @danshipped