SUBJECT-0042·SaaS·2026-04-11
THUMBSTOP SCORE
I fired my top salesperson. Revenue went up 40% the next quarter.
Thumbstop Score
0/100
CURIOSITY
0
SPECIFICITY
0
PATTERN INT.
0
EMOTION
0
STAKES
0
Lab Active·0 HOOKS ANALYZED·V1.0
The science of
the scroll-stop.
Engineered hooks for short-form video. Every hook scored across five behavioral vectors — before it ever reaches a feed.
Median decision
0.0s
Vectors scored
0
Corpus size
0.0M+
Engine version
v1.0
Fig. 01 — Specimen analysisMeasuring
§1Anatomy
The anatomy of a Thumbstop.
Five measurable vectors determine whether a thumb stops. The Lab scores each — so you can engineer the hook instead of guessing at it.
/01Vector
Curiosity Gap
The knowable unknown.
A hook works when the viewer feels a question they need answered — and believes the answer is within reach. Vague curiosity loses. Targeted curiosity converts.
Typical range40–95
050100
/02Vector
Specificity
Numbers, names, stakes.
Precision at speed creates credibility. “$47,000” beats “a lot of money.” Every abstraction you leave in costs you a thumb.
Typical range30–98
050100
/03Vector
Pattern Interrupt
A break from the expected.
The brain prioritizes novelty. A visual, verbal, or structural break in the first frame forces the eye to process — and the thumb to pause.
Typical range35–92
050100
/04Vector
Emotional Charge
Arousal drives retention.
Shock, delight, outrage, relief — the polarity matters less than the amplitude. Neutral hooks lose. Charged hooks compound.
Typical range20–96
050100
/05Vector
Stakes
What's lost or gained.
Consequence anchors the watch. A hook without stakes is trivia. The viewer needs to feel that something changes if they keep watching — or if they don't.
Typical range25–94
050100
§2Method
From topic to thumb-stopper in eight seconds.
A three-step protocol built for content teams shipping volume. No blank-page paralysis. No committee reviews. No prompt engineering.
Step 01/ 03
Input · seconds
Input context.
Niche, angle, brand voice, taboo list. One paragraph is enough. The Lab infers tone, constraints, and audience before a single hook is written.
Step 02/ 03
Process · 6–8s
Generate & rank.
Twenty hook candidates, each scored across five behavioral vectors. Top three surface instantly. The others stay ranked in the dossier for A/B rotation.
Step 03/ 03
Output · formatted
Export with script.
Full thirty-second script, shot list, on-screen text, and CTA. One click to Google Docs, Notion, or your CMS. The hook arrives production-ready.
§3Library
Specimens from the lab.
A rotating slice of hooks analyzed by the Lab, across fifteen niches. Every card is real output. Scored by the same engine that will score yours.
Streaming · live sample
Median score: 82·Niches sampled: 15
SUBJECT-0042SaaS
“I fired my top salesperson. Revenue went up 40% the next quarter.”
Score
89/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0118Fitness
“I lost 30 pounds eating the foods my doctor told me to avoid.”
Score
85/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0207Finance
“The $12 bank fee that's quietly costing you $47,000 in retirement.”
Score
87/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0291Ecommerce
“We hid our 'Add to Cart' button for 48 hours. Sales tripled.”
Score
85/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0344Food
“Restaurant pasta hits different because of one ingredient you can buy for $4.”
Score
75/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0408Creator
“I had 200k followers and made zero dollars. Then I changed one line in my bio.”
Score
84/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0471Real Estate
“This $180k fixer-upper sold for $1.2M. The buyer never renovated it.”
Score
86/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0523Education
“Harvard studied 268 men for 75 years. The #1 predictor of happiness wasn't money.”
Score
82/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0589Personal Dev
“I journaled every day for a year. I don't recommend it.”
Score
78/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0637B2B SaaS
“Our churn dropped 60% after we deleted our onboarding emails.”
Score
84/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0702Career
“The interview question that gets you hired isn't about you.”
Score
78/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0768Parenting
“My 4-year-old ended our dinner-table fights with one question.”
Score
82/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0815Marketing
“Our worst-performing ad made us $2M. We're running it again.”
Score
83/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0884Health
“Your sleep isn't the problem. Your bedroom is lying to you.”
Score
79/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0946Productivity
“I cancelled every recurring meeting on my calendar. Nobody noticed.”
Score
78/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0042SaaS
“I fired my top salesperson. Revenue went up 40% the next quarter.”
Score
89/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0118Fitness
“I lost 30 pounds eating the foods my doctor told me to avoid.”
Score
85/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0207Finance
“The $12 bank fee that's quietly costing you $47,000 in retirement.”
Score
87/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0291Ecommerce
“We hid our 'Add to Cart' button for 48 hours. Sales tripled.”
Score
85/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0344Food
“Restaurant pasta hits different because of one ingredient you can buy for $4.”
Score
75/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0408Creator
“I had 200k followers and made zero dollars. Then I changed one line in my bio.”
Score
84/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0471Real Estate
“This $180k fixer-upper sold for $1.2M. The buyer never renovated it.”
Score
86/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0523Education
“Harvard studied 268 men for 75 years. The #1 predictor of happiness wasn't money.”
Score
82/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0589Personal Dev
“I journaled every day for a year. I don't recommend it.”
Score
78/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0637B2B SaaS
“Our churn dropped 60% after we deleted our onboarding emails.”
Score
84/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0702Career
“The interview question that gets you hired isn't about you.”
Score
78/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0768Parenting
“My 4-year-old ended our dinner-table fights with one question.”
Score
82/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0815Marketing
“Our worst-performing ad made us $2M. We're running it again.”
Score
83/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0884Health
“Your sleep isn't the problem. Your bedroom is lying to you.”
Score
79/100
CUR
SPC
PAT
EMO
STK
SUBJECT-0946Productivity
“I cancelled every recurring meeting on my calendar. Nobody noticed.”
Score
78/100
CUR
SPC
PAT
EMO
STK
All specimens illustrative · scores are calibrated to nicheView the full catalog→
For teams shipping volume
At a hundred videos a month, hook quality is not an art problem. It’s a throughput problem.
Fig. 02 — Lab Telemetry
Updated 2026-04-17
0.0s
Median time before a thumb decides
Source · Lab corpus, 2026
0.0M+
Short-form hooks analyzed
Corpus · rolling 24 months
0×
Ideation speed vs. manual
Measured · pilot cohort
0
Behavioral vectors scored per hook
Engine · v1.0
§4Research
Notes from the Lab.
An open dossier on what the Lab is measuring, how, and why. Opened for review before the app opens for use.
Volume 01 · Issue 04
Revised 2026-04-17
A.
A 0–100 composite measure of a hook's likelihood of halting a scroll within the first 0.8 seconds of exposure. The score aggregates five behavioral vectors — Curiosity, Specificity, Pattern Interrupt, Emotion, and Stakes — each calibrated against a corpus of two million short-form videos with labeled retention outcomes. 80+ outperforms the median by a measurable margin.
The Lab is open
Stop making thumbs
scroll past.
Filed under Short-form · Growth · Creative Ops
Review cohort Agencies · in-house teams · operators
Engine v1.0 · calibrated 2026-04