Guide
The offer does most of it. Production value does least
The offer decides most of it. Then how specific the claim is. Then whether the right viewer reaches the ask. Production quality comes fourth at best, and in a feed a polished look often costs you performance. This page ranks the four levers, shows what a checkable claim looks like, and gives the coverage ceiling that keeps a face in the ad. By the end you will know where the next hour belongs.
- Levers, in descending order of effect
- 4
- What the opening has to work with
- 8 words
- Cutaway ceiling at normal pace
- 45%
The offer outranks everything, and no edit compensates for a weak one
What you sell, at what price, with what guarantee, to whom. That is the offer. It moves results more than every creative decision below it combined.
This is unwelcome because it sits outside the creative team's control. That is exactly why it gets skipped in favour of arguments about caption style.
Here is a useful test. Replace the ad with one plain sentence stating the offer. Would anybody buy? When the answer is no, creative work is polishing a decision that was already made badly.
Risk reversal is the most reliable component. A guarantee that removes the buyer's downside outperforms almost any adjective you could put in the script.
Once the offer is right, the ad's job narrows usefully. Make the offer understood, believed, and reachable by somebody who did not ask to hear it.
Specific claims get believed. General claims get classified as advertising.
Every viewer runs a filter. It sorts what they see into a person telling them something and a company selling them something. Specificity is the strongest signal on the person side.
A number, a date, a name, a consequence. The stain came out in one wash. The second version changed because eleven people asked for it. The return rate is under two percent.
Generic claims fail for a mechanical reason rather than an aesthetic one. Everybody says fast. Everybody says premium. Those words carry no information and get filtered before they are processed.
The detail nobody would invent is the strongest evidence available. The fridge humming in the background. The blown out window. The placeholder that went live by accident.
That is why a phone video in a real room beats a studio shot of the same script. The room is evidence, and evidence is what a viewer is short of.
The same claim, twice
Left is what most scripts say. Right is what a viewer can check, which is the only version that survives the filter.
| Filtered as advertising | Read as evidence | |
|---|---|---|
| Speed | Fast results | Out in one wash, on a shirt I had written off |
| Quality | Premium materials | Second version changed because eleven people asked |
| Trust | Customers love it | Return rate is under two percent |
| Value | Great value for money | Works out at about the price of a coffee a week |
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
Structure exists to get the right person to the ask, and to lose the rest early
A converting ad is not the one that keeps everybody. It is the one that keeps the people who might buy and lets the rest leave quickly.
The opening buys attention. Done properly it also disqualifies. Naming the audience in the first sentence loses people who were never going to convert. That lowers your retention chart and raises your results.
The middle has one job. Make the claim believable. Proof, demonstration and the specific detail belong there. A second example usually does not.
The ask has to be explicit and singular. An ad that ends with three possible actions gets none of them. An ad that never says what to do relies on the viewer inventing it.
Between those, the mechanics matter in a small way. Captions carry the message for muted viewers. Cutaways reset attention. Pace changes when the argument changes.
Where each job sits in a thirty second ad
The first block decides whether the rest happens. The last block is the only one that asks for anything.
- Hook and disqualify0-3s
- Problem, in their words3-8s
- Mechanism8-15s
- Proof15-24s
- One ask24-30s
Production value signals advertisement, and advertisement is what they are avoiding
In a feed placement, polish is a category signal. A graded, tripod-mounted, colour-matched shot tells a viewer this is an ad before a single word arrives.
That is not an argument for bad video. Unintelligible audio loses people faster than anything else on this page. Audio is the one place worth spending attention.
It is an argument against spending the budget on the wrong line. A better camera changes little. Standing closer to the microphone changes a lot. What you say changes most.
The one production decision that reliably matters is coverage. Bury the speaker under cutaways and the person the viewer was following disappears. That is why the ceiling sits at 40, 45 and 52 percent by pace.
Above roughly half the runtime a talking-head ad becomes a stock montage with narration. That register is exactly what people learned to skip.
Where the result actually comes from
A rough ranking rather than a measurement. Every account should test it, and almost none has.
- The offerMost of it
- Specificity of the claimA lot
- Structure and the askSome
- Production polishLeast
Cutroom carries the fourth lever, so the first three get your whole week
The top three levers are yours. The offer, the claim, and the ask. Nothing automates those, and nothing should.
The fourth comes off your plate entirely. One talking-head take of up to three minutes returns a finished 9:16 MP4, directed on the transcript rather than a timeline.
Highlight a phrase and a clip lands over exactly those words. Delete a line and the cut rebuilds around the gap. Change the pace and it re-cuts, with coverage held under the ceiling.
A batch is 100 credits. An export is 20 credits per output minute. A thirty second ad is 110 credits. That is what buys back the hours the first three levers deserve.
Some products only convert when the viewer watches the thing happen, somewhere a phone in a kitchen cannot reach. For those the shot is the ad, so book a videographer for the demonstration.
Then run this route for the volume around it. That division of labour produces more tested creatives a month than either route alone, and the fourth lever stops costing you a week.
Questions people ask
- Does a better hook improve conversion or only views?
- Both, indirectly. A hook cannot convert anybody, and nobody converts who left at second two. Improving the opening raises the number of qualified people who reach the ask, which shows up in results and in cost per landing page view.
- Should the ad show the price?
- Test it. Showing price disqualifies early, which lowers click volume and often raises conversion rate on the clicks you get. For a considered purchase it usually helps. For an impulse purchase it usually does not.
- How many claims should one ad make?
- One, supported. Two claims properly evidenced beat five asserted, and five asserted is what most scripts contain. Each extra claim needs its own proof, and proof takes time the ad does not have.
- Is creative work the right place to start?
- Only once the offer has been tested against a different one. Creative multiplies the offer, so a weak offer with excellent ads buys expensive confirmation. Change the guarantee or the price first and rerun the same creative.
Fix the offer, say something checkable, ask for one thing. Cutroom carries the fourth lever at 110 credits an ad, so the argument about polish never costs you a week again.