Video sells the idea. Static sells the decision. They are not competing formats and running one without the other leaves money on the table in both directions.
The question "does static or video perform better" has no answer because it is the wrong question. The right one is which job you are asking the ad to do, because each format is good at a different one.
What video is genuinely better at
Explaining something the viewer does not yet understand. If they need to see the product work, see the scale, see the before and after, no static can carry that. Sequence is the whole advantage.
Building belief. A person speaking to camera about a result carries a kind of credibility a caption cannot fake. Testimonial is the clearest example.
Cold traffic at scale. Video gives the algorithm richer engagement signals to work with. Watch time, retention and completion all feed delivery in ways that a static impression does not.
What static is genuinely better at
Speed of message. A static communicates in under a second. No hook required, no retention curve, no risk of losing someone at 0:03.
Price, offer and comparison. Anything numeric lands harder held still. Discounts, bundles, spec comparisons, side by side. Video makes the viewer wait for a number that a static hands them immediately.
Retargeting. The viewer already knows the story. They do not need it told again, they need a reason to act now. Static is usually the more efficient reminder.
Cost. A static costs a fraction of a video to produce and iterate. For pure offer testing, that ratio matters. You can test eight price framings in statics for less than one video.
The split that works
For most DTC accounts, a reasonable starting allocation:
- Cold prospecting: 70 to 80 percent video, 20 to 30 percent static
- Retargeting: 50 to 60 percent static, the rest video
- Offer and promotional pushes: mostly static, because the message is a number
Then adjust from your own data rather than defending the split. The allocation is a starting position, not a policy.
The mistake that costs the most
Treating static as the cheap option and briefing it accordingly.
Most brands put real thought into video and then produce statics by exporting a frame from the video with the logo dropped on. That static will underperform, the team will conclude static does not work for them, and a genuinely useful format gets abandoned for a reason that was self inflicted.
A good performance static is designed, not extracted. It has one message, high contrast, text legible at thumbnail size, and it survives being seen for 0.4 seconds. That is a specific craft and it is not a byproduct of video editing.
The one thing both formats share
The first moment does all the work.
In video it is the hook. In static it is whether the eye lands on the message before the thumb moves. Same problem, different medium. An ad that fails at that moment cannot be rescued by anything that comes after it, which is why the nine hook formats transfer almost directly into static headline writing.
How to test them fairly
Do not put statics and videos in the same ad set and compare CPA. They accumulate impressions at different rates and cost different amounts per impression, so the comparison is contaminated before it starts.
Run them in separate ad sets with equal budgets, judge each against its own format's history, and make the allocation decision at the campaign level rather than the ad level.
Judged that way, most accounts discover the same thing. Video wins on volume of new customers. Static wins on efficiency within an audience that already knows them. Both are true simultaneously, and any test designed to produce a single winner was going to mislead you regardless of which one it picked.