Black Friday video ads: the offer is the hook

The scale is not in dispute. Adobe Analytics put United States online spend at $11.8 billion on Black Friday 2025, up 9.1% year on year, $14.25 billion on Cyber Monday, and $44.2 billion across the five days from Thanksgiving to Cyber Monday.

Those figures describe demand, not your results. What they explain is the crowding. Every brand with a budget is buying attention in the same few days, which is why costs rise and why an ad that needs a week to prove itself never gets one.

So the calendar decides more than the creative does. Anything filmed in November is being uploaded untested into the most expensive week of the year.

The alternative is unglamorous. Film in October, test small in early November while attention is cheaper, and walk into the window with creative that has already been judged on something.

One thing worth being blunt about: no plan makes a discount week worthwhile if the margin is not there. Creative decides which offer gets seen. It does not decide whether the offer was a good idea.

For most of the year the hook is a claim, a problem or a surprise. Inside a discount window it is the number.

Everyone in the feed is announcing a sale, so the differentiator is not the phrasing. It is what the offer actually is, and how fast it is legible. Say the number out loud in the first two seconds and put it on screen at the same time, because a large share of the feed is watched with the sound off.

What varies is the framing of the same offer, and that is where variants come from. One 30% can be a bundle, a tier, a gift with purchase, an early-access window for the email list, or a single hero product at a price that stops a thumb.

The deadline is the second half of the hook. A discount with a date attached is an ad. A discount with no date is a price change.

And the blunt part again: weak offers are not rescued by good creative. If the best available is 10% off something nobody was hesitating over, the video is not the problem worth solving this month.

One filming block covers the whole window if the list is written first. Each of these is a separate take, thirty seconds or under, recorded back to back at the same table.

Film the first nine in October. The tenth arrives during the window itself, which is exactly why the schedule needs a gap in it.

Ad fatigue in a normal month is a slow curve you can watch. In a five-day window there is no time to watch anything, so the rotation goes into the calendar in advance.

The practical version: a new variant every two or three days across the window, with the offer held constant and only the opening seconds changed. A new hook on the same body is a new ad as far as the auction is concerned, and a cheap one to make.

Sequence matters more than volume. Lead with the plainest statement of the offer, because that is what a cold audience needs first. Move to the bundle and gift-guide framings mid-window, once the easy buyers are through. Keep the shipping-deadline ad and one reaction cut in reserve for the final days, when urgency is the only thing left to say.

This is also the argument for a stack rather than three ads. Motion's Creative Benchmarks 2026, built on $1.29 billion of realised Meta spend across 578,750 creatives, found about 5% of creatives became winners and that those winners carried 55% of the spend. Walking into the busiest week of the year with three creatives is a sample size, not a strategy.

The trade-off is why most brands skip it. More variants cost more to produce, and that cost is precisely what pushes production into November when there is no time left to make them. Filming in October is how you buy the volume back.

Block one day. Set the table by the window, put the ten angles on a sheet of paper, and record each one twice: once plainly, once with a different opening sentence.

Get the dates right on camera. A shipping cut-off spoken wrongly is an ad you have to pull and a support inbox you have to answer for a fortnight.

Name the files as you record them. Angle, framing, variant number. By the middle of the window nobody remembers what version four was, and the account fills up with things called final_final.

Keep one recording slot open during the window for the two things that cannot be filmed in advance: your reaction to a real customer, and the restock.

The remaining work is the edit, and at fifteen or twenty variants that is where an October plan quietly dies. Cutroom is built for that step: each take, up to 3 minutes, comes back as a finished 9:16 MP4 with every word transcribed, the hook written, your product footage placed on the words that mention it, and captions timed in one of six styles, all directed from the transcript instead of a timeline.

A batch is 100 credits and export adds 20 credits per finished minute, so a 30 second cut is about 110 credits and fifteen of them come to roughly 1,650. Lite is $19.99 a month for 1,250 credits, about 11 videos of that length, so a season built on fifteen or twenty variants belongs on Basic at $39.99 for 2,500 credits. Cancel is one click when the window closes.