Sound is on and the screen is full. The first word decides it.

On most feeds the first second is silent and the caption does the work.

Snapchat is different. The ad arrives full screen with sound already playing.

That removes a hiding place.

On a muted surface a weak opening line can be rescued by a strong caption.

Here the viewer hears the weak line before they read anything.

So the read has to be written to be heard rather than skimmed.

Short first sentence, a real claim, no throat clearing, and no music bed swelling before a person speaks.

It also means the audio quality of the take matters more than on other surfaces.

A phone in a quiet room is fine. A phone in a car park is a problem you cannot fix afterwards.

The cheapest improvement available is a better room rather than a better camera.

A phone at arm's length in a carpeted room beats a good camera in a kitchen with hard walls.

Snapchat's audience has grown up with advertising as a native part of every feed.

They identify a sales structure faster than almost any other group of viewers.

Which means the standard persuasion shape, problem then agitation then solution, is recognised as a shape.

It is not experienced as an argument.

What survives is directness. Say what it is, say what it costs, say who it is wrong for.

That last part reads as confidence rather than as a warning.

Humour survives too. It is the one register where a brand can be obviously an advertiser without being penalised for it.

What does not survive is a corporate voice pretending to be a person.

Being an ad is fine here. Being a fake friend is not.

The practical version is to drop the framing devices. Start at the claim and let the viewer decide, because they were going to anyway.

Upload one talking-head take, up to three minutes.

Cutroom transcribes the read, drafts the cut, places b-roll against the words and returns a finished 9:16 MP4 with captions burned in.

You direct on the transcript. Delete the run-up so the ad starts on the claim.

Highlight a phrase and a clip lands over exactly those words. Change the pace and it re-cuts.

The fast pace setting suits this surface, and it caps b-roll coverage at 52 percent.

So the person still holds most of the runtime, which is what a direct claim needs.

Your product footage sits in the workspace library and lands on the words that name it.

Captions still matter, because plenty of viewing happens with the volume low.

Six packs plus colour, weight, size and position, with per-word emphasis on the claim.

A finished thirty second ad is about 110 credits: 100 for the batch, 10 to render.

Roughly 5 to 8 percent of creatives carry spend, per Motion's analysis of 550,000+ Meta ads.

On a surface this unforgiving about the first second, the opening is most of the variance.

So the useful unit is a set of first lines against one body.

Eight openings from one take, run together, rather than one ad watched for a fortnight.

Basic is $39.99 for 2,500 credits, roughly 22 finished ads.

Premium is $79.99 for 5,000 credits, roughly 45. A $15 top-up adds 1,000 credits.

Eight openings is about 880 credits, which fits inside a single month on Basic.

There is room left to replace the losers in the same month.

The trial is 7 days and 300 credits with no card, which is two finished ads.

Music is a toggle, a track and a level. There is no mixing and no noise reduction.

A take recorded somewhere loud stays a take recorded somewhere loud, so the room is the decision that matters.

The free b-roll search covers the world around the claim, from millions of clips.

Your product and anything you filmed yourself go into the workspace library.

Highlight the words they belong on and they land there.

Exports are 9:16 MP4 only, which is exactly what this placement takes.

Source takes cap at three minutes and one project makes one finished ad.

Somebody has to be on camera.

If nobody will, the avatar module builds a presenter from a photo and a voice-over at 320 credits per minute of render, with voice-over at 70 per minute.