15-Second Hook Tests — Scripts

Character and shot rules: Production Spec. Hook IDs: Hook Library. Numbers: Verified Data Sheet.

Every episode ends on the catchphrase. “This is why I’m high.” The line is the domain, so the sign-off does the funnel work with no “link in bio.” Delivered flat and sincere, never explained, never in the first three seconds. See The Funnel.

What This Pack Is For

Every other pack tests a show. This pack tests a hook. Fifteen seconds is too short to carry a show format — there is no room for a setup, a turn, and a payoff. What it is perfect for is isolating the opening, because at 15 seconds the hook is most of the video. If someone watches to the end, the hook did it. So this pack is built as a controlled experiment. Two payloads. Four hook families each. The payoff, runtime, environment, and CTA are identical within each pair — the only thing that changes is the first two seconds.

Payload	Hook Families Tested	Script IDs
A — Nearly half of all cannabis is myrcene-led	Bold Claim, Data Drop, Question, Comedy	`HT-001` to `HT-004`
B — Almost no strain actually hits 30% THC	Bold Claim, Data Drop, Problem Agitation, Contrarian	`HT-005` to `HT-008`

Running two payloads matters. If Data Drop wins on payload A and payload B, that is a real finding about the audience. If it wins on one and loses on the other, the result belongs to the payload, not the hook — and one payload alone would have told you the wrong thing.

Post these in a randomized order across the week, not back to back. Four near-identical videos in a row trains the algorithm to treat them as one piece of content and suppresses the later ones. Interleave them with other packs.

Why 15 Seconds

Completion rate ceiling

A 15-second video is the easiest thing on the platform to finish. Completion rate is the strongest distribution driver, so this length has the highest floor for reach.

Rewatch behavior

Short enough to loop before a viewer decides to scroll. Rewatch rate above 1.0 is realistic here and almost impossible at 60 seconds.

Cheap to produce

One or two shots. A full batch of eight is a single production session.

Hook is the variable

At this length there is nothing else to blame. That is the entire point of the pack.

Word budget: 35-40 spoken words. That is the whole script. Anything longer will not fit at natural pace.

PAYLOAD A — The Myrcene Monoculture

All four scripts land the same payoff: nearly half of all cannabis is led by the same terpene, so “variety” on the shelf is mostly an illusion. Only the hook changes.

HT-001 — Bold Claim


Runtime	15 seconds
Hook family	Bold Claim (`BC-03`)
Register	Data
CTA type	Comment
Test variable	Hook family, payload A held constant

Script

Time	Voiceover	On Screen	Shot
0:00-0:02	”Half of all cannabis is the same terpene.”	HALF. THE SAME TERPENE.	deadpan, direct to lens + static chest-up + no graphic yet
0:02-0:09	”Myrcene leads nearly half of every strain out there. Different names, different prices, same molecule steering the experience.”	MYRCENE — NEARLY HALF	neutral, sliding chart in + slow push-in + terpene distribution bar
0:09-0:13	”So when your shelf looks varied and your nights all feel the same — that is why.”	THAT’S WHY IT FEELS THE SAME	neutral, direct to lens + static chest-up + statement card
0:13-0:15	”What is in your jar? Comments.”	WHAT’S IN YOUR JAR?	neutral + static chest-up + closing card

Caption

Myrcene leads nearly half of all cannabis. Different names, different prices, same molecule steering the night. That’s why a varied shelf can still feel like one strain. This is why I’m high → thisiswhyimhigh.com

Hashtags #myrcene #terpenes #cannabiseducation #cannabisscience #strainreview Cantina Prompt

Professor High — 15S HOOK TESTS — HT-001
Runtime: 15 seconds | 9:16 vertical
CHARACTERS
- professor-high-v1 — LOCKED. Already loaded with his image, voice and
  personality. Do NOT re-describe him — re-describing a saved character is the
  main cause of drift.
- No supporting cast — Professor High alone.
- Any supporting role may be recast freely between packs, never inside one — a
  recast mid-pack turns a format test into a casting test. Briefs and swap
  rules: /social/cast
TONE: Deadpan and certain. He is stating a fact he finds mildly annoying that
nobody else has noticed. Sincere throughout, never winking at the camera.

ENVIRONMENT: Working research lab. Dark pine-green backdrop, out-of-focus
shelving, glassware and papers behind him. Warm key light from camera left.

SHOTS + DIALOGUE:
1. (0:00-0:02) Static chest-up, deadpan, direct to lens. No graphic.
   SAYS: "Half of all cannabis is the same terpene."
   ON SCREEN: HALF. THE SAME TERPENE.
2. (0:02-0:09) Slow push-in. He slides a terpene distribution bar chart into
   the lower third.
   SAYS: "Myrcene leads nearly half of every strain out there. Different
   names, different prices, same molecule steering the experience."
   ON SCREEN: MYRCENE — NEARLY HALF
3. (0:09-0:13) Static chest-up, direct to lens, chart gone.
   SAYS: "So when your shelf looks varied and your nights all feel the same
   — that is why."
   ON SCREEN: THAT'S WHY IT FEELS THE SAME
4. (0:13-0:15) Static chest-up, closing card.
   SAYS: "What is in your jar? Comments."
   ON SCREEN: WHAT'S IN YOUR JAR?

ON-SCREEN TEXT STYLE: Bold sans-serif, white with heavy stroke, max 6 words
per line, lower third. Key terms in pineapple yellow #D98E04. Word-synced.
Keep 250px clear at the bottom and 120px at the top.

AUDIO: Lo-fi lab bed, low under voice. Voice normalized loudest.

DO NOT: No smoking, no lit product, no cannabis paraphernalia in frame. No
medical claims. No real dispensary, producer, or lab names. No logos.

HT-002 — Data Drop


Runtime	15 seconds
Hook family	Data Drop (`DD-01`)
Register	Data
CTA type	Comment
Test variable	Hook family, payload A held constant

Script

Time	Voiceover	On Screen	Shot
0:00-0:02	”Forty-five point eight percent of all cannabis is myrcene-dominant.”	45.8% MYRCENE-DOMINANT	neutral, direct to lens + static chest-up + big number card
0:02-0:09	”That is not a category. That is one molecule leading almost half the shelf, under a thousand different names.”	ONE MOLECULE. HALF THE SHELF.	neutral, sliding chart in + slow push-in + terpene distribution bar
0:09-0:13	”Which is why buying a new strain so often gets you the same night.”	SAME NIGHT, NEW NAME	neutral, direct to lens + static chest-up + statement card
0:13-0:15	”What is in your jar? Comments.”	WHAT’S IN YOUR JAR?	neutral + static chest-up + closing card

Caption

45.8% of all cannabis is myrcene-dominant. One molecule leading almost half the shelf under a thousand different names. That’s why a new strain so often gets you the same night. This is why I’m high → thisiswhyimhigh.com

Hashtags #myrcene #terpenes #cannabisdata #cannabisscience #strainreview Cantina Prompt

Professor High — 15S HOOK TESTS — HT-002
Runtime: 15 seconds | 9:16 vertical
CHARACTERS
- professor-high-v1 — LOCKED. Already loaded with his image, voice and
  personality. Do NOT re-describe him — re-describing a saved character is the
  main cause of drift.
- No supporting cast — Professor High alone.
- Any supporting role may be recast freely between packs, never inside one — a
  recast mid-pack turns a format test into a casting test. Briefs and swap
  rules: /social/cast
TONE: Precise and matter-of-fact. A researcher reading out a result he double-
checked. Sincere throughout, never winking at the camera.

ENVIRONMENT: Working research lab. Dark pine-green backdrop, out-of-focus
shelving, glassware and papers behind him. Warm key light from camera left.

SHOTS + DIALOGUE:
1. (0:00-0:02) Static chest-up, direct to lens. Large number card on screen.
   SAYS: "Forty-five point eight percent of all cannabis is myrcene-dominant."
   ON SCREEN: 45.8% MYRCENE-DOMINANT
2. (0:02-0:09) Slow push-in. He slides a terpene distribution bar chart into
   the lower third.
   SAYS: "That is not a category. That is one molecule leading almost half
   the shelf, under a thousand different names."
   ON SCREEN: ONE MOLECULE. HALF THE SHELF.
3. (0:09-0:13) Static chest-up, direct to lens, chart gone.
   SAYS: "Which is why buying a new strain so often gets you the same night."
   ON SCREEN: SAME NIGHT, NEW NAME
4. (0:13-0:15) Static chest-up, closing card.
   SAYS: "What is in your jar? Comments."
   ON SCREEN: WHAT'S IN YOUR JAR?

ON-SCREEN TEXT STYLE: Bold sans-serif, white with heavy stroke, max 6 words
per line, lower third. Key terms in pineapple yellow #D98E04. Word-synced.
Keep 250px clear at the bottom and 120px at the top.

AUDIO: Lo-fi lab bed, low under voice. Voice normalized loudest.

DO NOT: No smoking, no lit product, no cannabis paraphernalia in frame. No
medical claims. No real dispensary, producer, or lab names. No logos.

HT-003 — Question


Runtime	15 seconds
Hook family	Question (`QN-01`)
Register	Data
CTA type	Comment
Test variable	Hook family, payload A held constant

Script

Time	Voiceover	On Screen	Shot
0:00-0:02	”Can you name one terpene? Most people cannot.”	CAN YOU NAME ONE?	neutral, direct to lens, slight challenge + static chest-up + no graphic
0:02-0:09	”Here is the one you have definitely had. Myrcene. It leads nearly half of all cannabis, and almost nobody buying it knows the name.”	MYRCENE — NEARLY HALF	excited, crown perked, sliding chart in + slow push-in + terpene distribution bar
0:09-0:13	”You have been choosing it by accident your entire life.”	YOU CHOSE IT BY ACCIDENT	neutral, direct to lens + static chest-up + statement card
0:13-0:15	”Did you know it? Be honest.”	BE HONEST	neutral + static chest-up + closing card

Caption

Can you name one terpene? Here’s the one you’ve definitely had — myrcene leads nearly half of all cannabis and almost nobody buying it knows the name. Did you know it? Be honest. This is why I’m high → thisiswhyimhigh.com

Hashtags #myrcene #terpenes #cannabisquiz #cannabiseducation #cannabisscience Cantina Prompt

Professor High — 15S HOOK TESTS — HT-003
Runtime: 15 seconds | 9:16 vertical
CHARACTERS
- professor-high-v1 — LOCKED. Already loaded with his image, voice and
  personality. Do NOT re-describe him — re-describing a saved character is the
  main cause of drift.
- No supporting cast — Professor High alone.
- Any supporting role may be recast freely between packs, never inside one — a
  recast mid-pack turns a format test into a casting test. Briefs and swap
  rules: /social/cast
TONE: Curious and lightly challenging, then genuinely delighted to explain. He is
not smug about the audience not knowing. Sincere throughout, never winking
at the camera.

ENVIRONMENT: Working research lab. Dark pine-green backdrop, out-of-focus
shelving, glassware and papers behind him. Warm key light from camera left.

SHOTS + DIALOGUE:
1. (0:00-0:02) Static chest-up, direct to lens, slight challenge in the look.
   No graphic.
   SAYS: "Can you name one terpene? Most people cannot."
   ON SCREEN: CAN YOU NAME ONE?
2. (0:02-0:09) Slow push-in. Crown perks up, visibly pleased. He slides a
   terpene distribution bar chart into the lower third.
   SAYS: "Here is the one you have definitely had. Myrcene. It leads nearly
   half of all cannabis, and almost nobody buying it knows the name."
   ON SCREEN: MYRCENE — NEARLY HALF
3. (0:09-0:13) Static chest-up, direct to lens, chart gone.
   SAYS: "You have been choosing it by accident your entire life."
   ON SCREEN: YOU CHOSE IT BY ACCIDENT
4. (0:13-0:15) Static chest-up, closing card.
   SAYS: "Did you know it? Be honest."
   ON SCREEN: BE HONEST

ON-SCREEN TEXT STYLE: Bold sans-serif, white with heavy stroke, max 6 words
per line, lower third. Key terms in pineapple yellow #D98E04. Word-synced.
Keep 250px clear at the bottom and 120px at the top.

AUDIO: Lo-fi lab bed, low under voice. Voice normalized loudest.

DO NOT: No smoking, no lit product, no cannabis paraphernalia in frame. No
medical claims. No real dispensary, producer, or lab names. No logos.

HT-004 — Comedy


Runtime	15 seconds
Hook family	Comedy (`CM-07`, new)
Register	Comedy
CTA type	Comment
Test variable	Hook family, payload A held constant

Script

Time	Voiceover	On Screen	Shot
0:00-0:02	”Your four favorite strains are the same strain. Sorry.”	THEY’RE THE SAME STRAIN	deadpan, direct to lens + static chest-up + no graphic
0:02-0:09	”Not legally. Chemically. Myrcene leads nearly half of all cannabis, so four random jars landing on the same molecule is the base case, not bad luck.”	THE BASE CASE, NOT BAD LUCK	deadpan, sliding chart in + slow push-in + terpene distribution bar
0:09-0:13	”You did not build a collection. You bought the same night four times.”	SAME NIGHT. FOUR TIMES.	deadpan, direct to lens + static chest-up + statement card
0:13-0:15	”Prove me wrong in the comments.”	PROVE ME WRONG	neutral, faint smile + static chest-up + closing card

Caption

Your four favorite strains are probably the same strain. Not legally — chemically. Myrcene leads nearly half of all cannabis, so four jars landing on one molecule is the base case. Prove me wrong. This is why I’m high → thisiswhyimhigh.com

Hashtags #myrcene #terpenes #plantmedicine #cannabiseducation #strainreview Cantina Prompt

Professor High — 15S HOOK TESTS — HT-004
Runtime: 15 seconds | 9:16 vertical
CHARACTERS
- professor-high-v1 — LOCKED. Already loaded with his image, voice and
  personality. Do NOT re-describe him — re-describing a saved character is the
  main cause of drift.
- No supporting cast — Professor High alone.
- Any supporting role may be recast freely between packs, never inside one — a
  recast mid-pack turns a format test into a casting test. Briefs and swap
  rules: /social/cast
TONE: Flat, dry, faintly apologetic — delivering bad news he considers obvious.
The comedy comes entirely from how seriously he takes it. He never winks at
the camera or signals that it is a joke.

ENVIRONMENT: Working research lab. Dark pine-green backdrop, out-of-focus
shelving, glassware and papers behind him. Warm key light from camera left.

SHOTS + DIALOGUE:
1. (0:00-0:02) Static chest-up, deadpan, direct to lens. No graphic.
   SAYS: "Your four favorite strains are the same strain. Sorry."
   ON SCREEN: THEY'RE THE SAME STRAIN
2. (0:02-0:09) Slow push-in, still deadpan. He slides a terpene distribution
   bar chart into the lower third.
   SAYS: "Not legally. Chemically. Myrcene leads nearly half of all cannabis,
   so four random jars landing on the same molecule is the base case, not bad
   luck."
   ON SCREEN: THE BASE CASE, NOT BAD LUCK
3. (0:09-0:13) Static chest-up, deadpan, direct to lens, chart gone.
   SAYS: "You did not build a collection. You bought the same night four
   times."
   ON SCREEN: SAME NIGHT. FOUR TIMES.
4. (0:13-0:15) Static chest-up, the faintest smile, closing card.
   SAYS: "Prove me wrong in the comments."
   ON SCREEN: PROVE ME WRONG

ON-SCREEN TEXT STYLE: Bold sans-serif, white with heavy stroke, max 6 words
per line, lower third. Key terms in pineapple yellow #D98E04. Word-synced.
Keep 250px clear at the bottom and 120px at the top.

AUDIO: Lo-fi lab bed, low under voice. Voice normalized loudest.

DO NOT: No smoking, no lit product, no cannabis paraphernalia in frame. No
medical claims. No real dispensary, producer, or lab names. No logos.

PAYLOAD B — The 30% THC Lie

All four scripts land the same payoff: almost no strain actually reaches 30% THC, so a menu full of them is a menu problem. Only the hook changes.

HT-005 — Bold Claim


Runtime	15 seconds
Hook family	Bold Claim (`BC-01`)
Register	Data
CTA type	Comment
Test variable	Hook family, payload B held constant

Script

Time	Voiceover	On Screen	Shot
0:00-0:02	”THC percentage is the biggest scam in cannabis.”	THE BIGGEST SCAM	deadpan, direct to lens + static chest-up + no graphic
0:02-0:09	”Fewer than one in two hundred strains actually reaches thirty percent. The average is around twenty. So why is half of every menu claiming thirty-plus?”	UNDER 1 IN 200 REACH 30%	skeptical, one brow raised, sliding chart in + slow push-in + THC distribution chart
0:09-0:13	”Either you found the rarest flower in the state, or somebody rounded.”	OR SOMEBODY ROUNDED	deadpan, direct to lens + static chest-up + statement card
0:13-0:15	”Highest number you have seen? Comments.”	HIGHEST YOU’VE SEEN?	neutral + static chest-up + closing card

Caption

Fewer than 1 in 200 strains actually reaches 30% THC. The average is around 20%. So why is half of every menu claiming 30-plus? What’s the highest number you’ve seen? This is why I’m high → thisiswhyimhigh.com

Hashtags #thcpercentage #cannabismyths #dispensary #cannabiseducation #cannabisscience Cantina Prompt

Professor High — 15S HOOK TESTS — HT-005
Runtime: 15 seconds | 9:16 vertical
CHARACTERS
- professor-high-v1 — LOCKED. Already loaded with his image, voice and
  personality. Do NOT re-describe him — re-describing a saved character is the
  main cause of drift.
- No supporting cast — Professor High alone.
- Any supporting role may be recast freely between packs, never inside one — a
  recast mid-pack turns a format test into a casting test. Briefs and swap
  rules: /social/cast
TONE: Flat and unimpressed, edging into quiet irritation. He is not ranting; he is
stating something he finds indefensible. Sincere throughout, never winking
at the camera.

ENVIRONMENT: Working research lab. Dark pine-green backdrop, out-of-focus
shelving, glassware and papers behind him. Warm key light from camera left.

SHOTS + DIALOGUE:
1. (0:00-0:02) Static chest-up, deadpan, direct to lens. No graphic.
   SAYS: "THC percentage is the biggest scam in cannabis."
   ON SCREEN: THE BIGGEST SCAM
2. (0:02-0:09) Slow push-in. One brow raised, skeptical. He slides a THC
   distribution chart into the lower third with the 30% threshold marked.
   SAYS: "Fewer than one in two hundred strains actually reaches thirty
   percent. The average is around twenty. So why is half of every menu
   claiming thirty-plus?"
   ON SCREEN: UNDER 1 IN 200 REACH 30%
3. (0:09-0:13) Static chest-up, deadpan, direct to lens, chart gone.
   SAYS: "Either you found the rarest flower in the state, or somebody
   rounded."
   ON SCREEN: OR SOMEBODY ROUNDED
4. (0:13-0:15) Static chest-up, closing card.
   SAYS: "Highest number you have seen? Comments."
   ON SCREEN: HIGHEST YOU'VE SEEN?

ON-SCREEN TEXT STYLE: Bold sans-serif, white with heavy stroke, max 6 words
per line, lower third. Key terms in pineapple yellow #D98E04. Word-synced.
Keep 250px clear at the bottom and 120px at the top.

AUDIO: Lo-fi lab bed, low under voice. Voice normalized loudest.

DO NOT: No smoking, no lit product, no cannabis paraphernalia in frame. No
medical claims. No real dispensary, producer, or lab names. No logos.

HT-006 — Data Drop


Runtime	15 seconds
Hook family	Data Drop (`DD-02`)
Register	Data
CTA type	Comment
Test variable	Hook family, payload B held constant

Script

Time	Voiceover	On Screen	Shot
0:00-0:02	”Fewer than one in two hundred strains hits thirty percent THC.”	UNDER 1 IN 200	neutral, direct to lens + static chest-up + big number card
0:02-0:09	”The average is about twenty percent. The highest figure ever recorded is under thirty-five. Those are the actual numbers.”	AVG 20% · CEILING UNDER 35%	neutral, sliding chart in + slow push-in + THC distribution chart
0:09-0:13	”So a menu full of thirty-plus is not a strong menu. It is an unreliable one.”	NOT STRONG. UNRELIABLE.	neutral, direct to lens + static chest-up + statement card
0:13-0:15	”Highest number you have seen? Comments.”	HIGHEST YOU’VE SEEN?	neutral + static chest-up + closing card

Caption

Fewer than 1 in 200 strains hits 30% THC. Average is around 20%. Highest ever recorded is under 35%. A menu full of 30-plus isn’t a strong menu, it’s an unreliable one. This is why I’m high → thisiswhyimhigh.com

Hashtags #thcpercentage #cannabisdata #cannabistesting #cannabiseducation #dispensary Cantina Prompt

Professor High — 15S HOOK TESTS — HT-006
Runtime: 15 seconds | 9:16 vertical
CHARACTERS
- professor-high-v1 — LOCKED. Already loaded with his image, voice and
  personality. Do NOT re-describe him — re-describing a saved character is the
  main cause of drift.
- No supporting cast — Professor High alone.
- Any supporting role may be recast freely between packs, never inside one — a
  recast mid-pack turns a format test into a casting test. Briefs and swap
  rules: /social/cast
TONE: Precise and even. Reading out verified numbers without editorializing until
the last line. Sincere throughout, never winking at the camera.

ENVIRONMENT: Working research lab. Dark pine-green backdrop, out-of-focus
shelving, glassware and papers behind him. Warm key light from camera left.

SHOTS + DIALOGUE:
1. (0:00-0:02) Static chest-up, direct to lens. Large number card on screen.
   SAYS: "Fewer than one in two hundred strains hits thirty percent THC."
   ON SCREEN: UNDER 1 IN 200
2. (0:02-0:09) Slow push-in. He slides a THC distribution chart into the
   lower third with the average and ceiling both marked.
   SAYS: "The average is about twenty percent. The highest figure ever
   recorded is under thirty-five. Those are the actual numbers."
   ON SCREEN: AVG 20% · CEILING UNDER 35%
3. (0:09-0:13) Static chest-up, direct to lens, chart gone.
   SAYS: "So a menu full of thirty-plus is not a strong menu. It is an
   unreliable one."
   ON SCREEN: NOT STRONG. UNRELIABLE.
4. (0:13-0:15) Static chest-up, closing card.
   SAYS: "Highest number you have seen? Comments."
   ON SCREEN: HIGHEST YOU'VE SEEN?

ON-SCREEN TEXT STYLE: Bold sans-serif, white with heavy stroke, max 6 words
per line, lower third. Key terms in pineapple yellow #D98E04. Word-synced.
Keep 250px clear at the bottom and 120px at the top.

AUDIO: Lo-fi lab bed, low under voice. Voice normalized loudest.

DO NOT: No smoking, no lit product, no cannabis paraphernalia in frame. No
medical claims. No real dispensary, producer, or lab names. No logos.

HT-007 — Problem Agitation


Runtime	15 seconds
Hook family	Problem Agitation (`PA-06`, new)
Register	Data
CTA type	Save
Test variable	Hook family, payload B held constant

Script

Time	Voiceover	On Screen	Shot
0:00-0:02	”You paid extra for that number and it did nothing.”	YOU PAID FOR THAT NUMBER	neutral, direct to lens + static chest-up + no graphic
0:02-0:09	”Fewer than one in two hundred strains actually reaches thirty percent, so the premium you paid was for a rounding error, not a stronger high.”	A ROUNDING ERROR	skeptical, sliding chart in + slow push-in + THC distribution chart
0:09-0:13	”Buy the terpene profile. Let the THC number be boring.”	LET THC BE BORING	neutral, direct to lens + static chest-up + statement card
0:13-0:15	”Save this for the counter.”	SAVE THIS	neutral + static chest-up + closing card

Caption

Fewer than 1 in 200 strains actually reaches 30% THC. That premium you paid was for a rounding error, not a stronger high. Buy the terpene profile. Let the THC number be boring. This is why I’m high → thisiswhyimhigh.com

Hashtags #thcpercentage #dispensary #terpenes #cannabiseducation #cannabismyths Cantina Prompt

Professor High — 15S HOOK TESTS — HT-007
Runtime: 15 seconds | 9:16 vertical
CHARACTERS
- professor-high-v1 — LOCKED. Already loaded with his image, voice and
  personality. Do NOT re-describe him — re-describing a saved character is the
  main cause of drift.
- No supporting cast — Professor High alone.
- Any supporting role may be recast freely between packs, never inside one — a
  recast mid-pack turns a format test into a casting test. Briefs and swap
  rules: /social/cast
TONE: Direct and slightly protective — he is annoyed on the viewer's behalf, not
at them. Sincere throughout, never winking at the camera.

ENVIRONMENT: Working research lab. Dark pine-green backdrop, out-of-focus
shelving, glassware and papers behind him. Warm key light from camera left.

SHOTS + DIALOGUE:
1. (0:00-0:02) Static chest-up, direct to lens. No graphic.
   SAYS: "You paid extra for that number and it did nothing."
   ON SCREEN: YOU PAID FOR THAT NUMBER
2. (0:02-0:09) Slow push-in. One brow raised, skeptical. He slides a THC
   distribution chart into the lower third.
   SAYS: "Fewer than one in two hundred strains actually reaches thirty
   percent, so the premium you paid was for a rounding error, not a stronger
   high."
   ON SCREEN: A ROUNDING ERROR
3. (0:09-0:13) Static chest-up, direct to lens, chart gone.
   SAYS: "Buy the terpene profile. Let the THC number be boring."
   ON SCREEN: LET THC BE BORING
4. (0:13-0:15) Static chest-up, closing card.
   SAYS: "Save this for the counter."
   ON SCREEN: SAVE THIS

ON-SCREEN TEXT STYLE: Bold sans-serif, white with heavy stroke, max 6 words
per line, lower third. Key terms in pineapple yellow #D98E04. Word-synced.
Keep 250px clear at the bottom and 120px at the top.

AUDIO: Lo-fi lab bed, low under voice. Voice normalized loudest.

DO NOT: No smoking, no lit product, no cannabis paraphernalia in frame. No
medical claims. No real dispensary, producer, or lab names. No logos.

HT-008 — Contrarian


Runtime	15 seconds
Hook family	Contrarian (`CT-05`)
Register	Contrarian
CTA type	Comment
Test variable	Hook family, payload B held constant

Script

Time	Voiceover	On Screen	Shot
0:00-0:02	”High THC is the reason you are not getting high.”	HIGH THC = LESS HIGH	deadpan, direct to lens + static chest-up + no graphic
0:02-0:09	”Chasing the number builds tolerance faster, and fewer than one in two hundred strains even reaches thirty percent. You are paying more to feel less.”	PAYING MORE TO FEEL LESS	skeptical, sliding chart in + slow push-in + THC distribution chart
0:09-0:13	”The people who stopped chasing it are the ones still getting somewhere.”	THEY STOPPED CHASING IT	neutral, direct to lens + static chest-up + statement card
0:13-0:15	”Disagree? Comments.”	DISAGREE?	neutral, faint smile + static chest-up + closing card

Caption

Chasing THC% builds tolerance faster — and fewer than 1 in 200 strains even reaches 30%. You’re paying more to feel less. Disagree? Comments. This is why I’m high → thisiswhyimhigh.com

Hashtags #thcpercentage #tolerancebreak #cannabismyths #cannabiseducation #cannabisscience Cantina Prompt

Professor High — 15S HOOK TESTS — HT-008
Runtime: 15 seconds | 9:16 vertical
CHARACTERS
- professor-high-v1 — LOCKED. Already loaded with his image, voice and
  personality. Do NOT re-describe him — re-describing a saved character is the
  main cause of drift.
- No supporting cast — Professor High alone.
- Any supporting role may be recast freely between packs, never inside one — a
  recast mid-pack turns a format test into a casting test. Briefs and swap
  rules: /social/cast
TONE: Confident and provocative, inviting an argument he is sure he wins. Not
smug. Sincere throughout, never winking at the camera.

ENVIRONMENT: Working research lab. Dark pine-green backdrop, out-of-focus
shelving, glassware and papers behind him. Warm key light from camera left.

SHOTS + DIALOGUE:
1. (0:00-0:02) Static chest-up, deadpan, direct to lens. No graphic.
   SAYS: "High THC is the reason you are not getting high."
   ON SCREEN: HIGH THC = LESS HIGH
2. (0:02-0:09) Slow push-in. One brow raised, skeptical. He slides a THC
   distribution chart into the lower third.
   SAYS: "Chasing the number builds tolerance faster, and fewer than one in
   two hundred strains even reaches thirty percent. You are paying more to
   feel less."
   ON SCREEN: PAYING MORE TO FEEL LESS
3. (0:09-0:13) Static chest-up, direct to lens, chart gone.
   SAYS: "The people who stopped chasing it are the ones still getting
   somewhere."
   ON SCREEN: THEY STOPPED CHASING IT
4. (0:13-0:15) Static chest-up, the faintest smile, closing card.
   SAYS: "Disagree? Comments."
   ON SCREEN: DISAGREE?

ON-SCREEN TEXT STYLE: Bold sans-serif, white with heavy stroke, max 6 words
per line, lower third. Key terms in pineapple yellow #D98E04. Word-synced.
Keep 250px clear at the bottom and 120px at the top.

AUDIO: Lo-fi lab bed, low under voice. Voice normalized loudest.

DO NOT: No smoking, no lit product, no cannabis paraphernalia in frame. No
medical claims. No real dispensary, producer, or lab names. No logos.

Reading The Results

This pack produces a real answer if it is logged properly. Fill the performance log for all eight, then:

Rank by 3-second hold rate within each payload

Payload A is HT-001 to HT-004. Payload B is HT-005 to HT-008. Ranking across payloads mixes two variables and tells you nothing.

Check whether the winner repeats

If the same hook family tops both payloads, that is a finding about the audience and it should reshape the A-hook choice in every other pack. If the winners differ, the payload was doing the work.

Compare completion against the 30s and 60s packs

Fifteen seconds should post the highest completion rate in the library. If it does not, the problem is the writing, not the runtime.

Promote the winning family

Whatever wins becomes the default A-hook family for the next batch across all shows.

Eight videos is enough to see a direction, not enough to be certain. Treat a clear winner as a strong prior worth acting on, and re-run the pack with two fresh payloads a month later to confirm.

New Hook IDs Introduced

Add these to the Hook Library when this pack ships.

ID	Family	Hook
`CM-07`	Comedy	”Your four favorite strains are the same strain. Sorry.”
`PA-06`	Problem Agitation	”You paid extra for that number and it did nothing.”

Hook Library

Every hook ID and family.

Testing Framework

How these results become decisions.

Production Spec

The Cantina prompt block format.

Start Here

Research

Character

Show Bible

Strategy

Scripts

The 100

Tentpole Shows

TIWIH Series

High IQ Series

Recurring Shows

Story Arcs

Cross-Platform Franchises

Viral Formats

Seasonal & Specials

15-Second Hook Tests — Scripts

What This Pack Is For

Why 15 Seconds

Completion rate ceiling

Rewatch behavior

Cheap to produce

Hook is the variable

PAYLOAD A — The Myrcene Monoculture

HT-001 — Bold Claim

HT-002 — Data Drop

HT-003 — Question

HT-004 — Comedy

PAYLOAD B — The 30% THC Lie

HT-005 — Bold Claim

HT-006 — Data Drop

HT-007 — Problem Agitation

HT-008 — Contrarian

Reading The Results

New Hook IDs Introduced

Hook Library

Testing Framework

Production Spec

​What This Pack Is For

​Why 15 Seconds

Completion rate ceiling

Rewatch behavior

Cheap to produce

Hook is the variable

​PAYLOAD A — The Myrcene Monoculture

​HT-001 — Bold Claim

​HT-002 — Data Drop

​HT-003 — Question

​HT-004 — Comedy

​PAYLOAD B — The 30% THC Lie

​HT-005 — Bold Claim

​HT-006 — Data Drop

​HT-007 — Problem Agitation

​HT-008 — Contrarian

​Reading The Results

​New Hook IDs Introduced

​Related

Hook Library

Testing Framework

Production Spec

What This Pack Is For

Why 15 Seconds

PAYLOAD A — The Myrcene Monoculture

HT-001 — Bold Claim

HT-002 — Data Drop

HT-003 — Question

HT-004 — Comedy

PAYLOAD B — The 30% THC Lie

HT-005 — Bold Claim

HT-006 — Data Drop

HT-007 — Problem Agitation

HT-008 — Contrarian

Reading The Results

New Hook IDs Introduced

Related