Bonus 03 · Hook + Hero-Shot Generator
Hook + Hero-Shot Generator
Twenty hook formulas that hold attention past three seconds, six rules for the thumbnail crop that earns the click, and the A/B framework that decides between them.
The formulas below are pulled from the top-performing shoppable videos across the Idukki customer base. Each pattern is paired with an example in the wild and a one-line note on why it works.
Section 01
Twenty hook formulas
Pick the one that maps to your category\'s typical objection. Each pattern is a fill-in-the-blank you can adapt to one of your real SKUs in under a minute.
- 01
POV: the moment you realise X.
“POV: the moment you realise the foundation actually matches your chin.”
Drops the shopper inside the customer's moment of recognition.
- 02
The X test most people fail.
“The fit-test most people fail in the changing room.”
Curiosity gap, plus implied authority.
- 03
I was wrong about X.
“I was wrong about thick eyeliner. It works on hooded eyes.”
Personal reversal triggers stop-the-scroll.
- 04
X but I tried Y instead.
“The internet told me to size up. I tried size down instead.”
Sets a contrast against expected wisdom.
- 05
X seconds is all you need to know if Y.
“Five seconds is all you need to know if this concealer creases.”
Time-promise hook for short attention spans.
- 06
No filter, no edit. Just X.
“No filter, no edit. Just the lipstick at 7am.”
Authenticity claim breaks past the polish wall.
- 07
If you have X, you need Y.
“If you have wide feet, you need to see the side stitch on this trainer.”
Specificity invites the right shopper to lean in.
- 08
Three reasons I almost returned this, and why I didn't.
“Three reasons I almost returned this hoodie, and why I didn't.”
Acknowledges objection while leading to resolution.
- 09
X months in.
“Six months in. Sofa still holding up.”
Time-tested social proof beats day-one excitement.
- 10
Watch this before you buy X.
“Watch this before you buy a "stainless" pan.”
Pre-purchase warning frame, high stop rate.
- 11
How I went from X to Y in Z.
“How I went from oily mid-day to matte all day in three weeks.”
Transformation arc with a measurable timeline.
- 12
My friend said X. Here's what happened.
“My friend said the candle would smell synthetic. Here's what happened.”
Borrowed objection, native social proof.
- 13
X dollars, three months later.
“Forty dollars, three months later, still my most-worn ring.”
Price + time = value claim.
- 14
The thing nobody tells you about X.
“The thing nobody tells you about silk pillowcases.”
Insider-knowledge frame.
- 15
I bought this for X. It also does Y.
“I bought this for the office. It also fits in my carry-on.”
Bonus utility surfaces hidden value.
- 16
Twenty seconds, three colours, one decision.
“Twenty seconds, three blush shades, one decision. Save this.”
Implicit promise of decisional clarity.
- 17
X review after Y wears.
“Honest review after twelve wears.”
Specificity boosts credibility.
- 18
What X looks like in real light.
“What this gold chain looks like in office light.”
Counters studio photography's flattery.
- 19
X mistake I made so you don't have to.
“Sizing mistake I made so you don't have to.”
Gift-frame, generous, builds trust.
- 20
This is the X every Y owner needs to know.
“This is the cleaning tip every cast-iron owner needs to know.”
Identity-targeted, invites a save.
Section 02
Six thumbnail rules
Thumbnails win or lose the click before the video ever plays. These six rules are the ones that survive across categories.
Rule 01. The face takes a third of the frame.
Watch-time correlates with a recognisable human face in the thumbnail. Smaller than a third looks staged; larger than a third feels like a selfie. Target one-third, with the eyes on the upper third-line.
Rule 02. The product needs one clear silhouette.
If the shopper cannot identify the product class from the thumbnail at 100x100 pixels, the thumbnail is failing. Crop so the product's outline is unambiguous before any colour or texture detail is even legible.
Rule 03. Use a fact, not a question, in the overlay.
Question overlays ("does this work?") test worse than fact overlays ("six months later") across the cohort we measured. Facts read as evidence; questions read as bait.
Rule 04. The colour palette should be three colours, two of them warm.
Thumbnails with three or fewer dominant colours pull better than busy ones. Two warm + one cool (or one neutral) reads as inviting; three cools reads as clinical.
Rule 05. The expression carries the hook.
If the hook is curiosity ("the thing nobody tells you"), the face should be wide-eyed. If the hook is contrast ("I was wrong about X"), the face should be sideways-glance. Surprised + product is the highest-stop combination in our customer data.
Rule 06. Test three crops, not three thumbnails.
A 16:9 hero, a 1:1 grid, a 9:16 vertical. Each is a different shopper context. Run all three for the same video; never assume the hero crop works as the grid tile.
Section 03
The A/B framework
A six-step routine to decide which hook and which crop is actually working, instead of guessing.
Hook + thumbnail A/B framework, condensed. 1. Pick two hook formulas from the list. Pick the two that feel closest to your category's typical objection. 2. Shoot two videos. Same SKU, same person, same lighting. Only the first three seconds change. 3. Cut three thumbnail crops from each. 16:9 for desktop hero, 1:1 for grid, 9:16 for vertical surfaces. 4. Run them in rotation for at least 72 hours or 1,000 impressions per variant, whichever comes later. 5. Decide on watch-time at the 50% mark, not on click-through. CTR optimises for curiosity; watch-time optimises for fit. 6. Promote the winning hook to your evergreen rail. Retire the loser. Pick a new challenger from the list. Repeat monthly.
