The holiday period is where a year's advertising discipline gets tested, because everything that is normally a minor inefficiency becomes expensive at once. Costs rise, competition rises, and creative fatigues faster because everyone is bidding for the same attention.
Most of what matters is decided before the period starts.
Prepare early, for a reason that is not motivational
Auction costs climb as the season approaches, so testing in the expensive window means paying premium prices to learn things you could have learned cheaply. The point of preparing early is not organisation, it is that discovery is cheaper in October than in late November.
What to establish before costs rise: which products respond to video, which angle works for each, and which format your audience actually watches. Those are three tests, and none of them needs holiday budget.
Three assets, not thirty
The offer ad. Clear, direct, the discount or the proposition stated early. This is the one that does the volume.
The gift-framing ad. The same product positioned as something bought for someone else, which is a different argument with a different buyer. Many stores never make this version and it is often the one that finds new customers.
The urgency ad, for the final days. Shipping deadlines are the honest version of urgency and they work better than invented scarcity, which experienced shoppers now discount automatically.
Three ads per key product, generated from the same link, each planning its own angle. That is a manageable amount of work and enough to rotate through the period.
Rotation is the actual mechanic
The specific way holiday campaigns fail is that a good ad is run until it collapses. Frequency climbs, cost per acquisition climbs with it, and by the time somebody notices, a week of peak-price budget has gone.
Watch frequency and cost per acquisition together and rotate on the signal rather than on a schedule. Having the replacements already generated is what makes rotation possible in a week when nobody has time to make anything.
Formats, per placement
Generate 9:16 for Reels, Stories and TikTok, 1:1 for feed placements, 16:9 for YouTube and landing pages. Cropping one master across all three is the fastest way to put your offer text where the interface covers it, and in a season where every impression is expensive that is a costly shortcut.
Retargeting deserves its own creative
People who watched most of an ad and did not buy are the most valuable audience you will have all year, and showing them the same video again wastes them. A second video that answers the objection, shipping time, sizing, whether it arrives before the date, converts a segment the first one could not.
What it costs to prepare
Credits do not expire and each generation shows its cost before you confirm, so a batch generated in October is not a use-it-or-lose-it purchase. The €5 entry offer produces one short video from your own product, which is enough to decide in advance whether this is part of your plan.
The honest limit
None of this makes an uncompetitive offer competitive. In a season where every seller in your category is discounting, video improves how well your offer is communicated, not how good it is.
Related reading: A/B testing product videos properly and Facebook and Instagram video ads.