Formats and creative executions
A good message in the wrong format is a lost message. The same promise can run as a video with a person talking to camera, as a static image with the product, as an explanatory graphic — and each form has a different economy: different production cost, different testing speed, a different type of trust it builds.
Your arena's market has already voted between these forms. The vote is public — it just has to be read.
What you'll be able to do after this chapter
- read your arena's format distribution: what dominates, what's missing, what's been running long;
- tell the dominant format from the obligatory one — they're not the same thing;
- choose your test format on signals and your own production economy, not on fashion.
The principle
In the Ad Library, the media facet (video / image) cuts the arena on the execution axis, and the format facet — on the product's form (powder, capsules, gummies...), which is a positioning choice of its own. Execution style (person on camera, demonstration, studio cut) is not a facet: you read it by opening the long-running ads. The correct reading has two layers, in the now-familiar order: first what exists (the distribution — how many ads per form), then what lasts (the age — which forms appear in the long-running ads).
The gap between the layers is often the most valuable information of the chapter: a format can dominate numerically because it's cheap to mass-produce, while longevity accumulates in another, more expensive and rarer. The first is what the market makes; the second, what seems to carry it.
And a format absent from the arena has two possible explanations — it was tried and died, or nobody tried seriously — and no table separates them from a distance: you check in neighbouring categories whether the format lives there on the same audience.
The workflow in RavenBI
- On your arena in the Ad Library, read the media facet: the video / image ratio, as the general dimension.
- Go into the format facets and note the full distribution — including the forms with a small presence; absences are data, not gaps.
- Apply the Exposure sort and go through the first two or three dozen ads: what forms are the veterans cast in? Compare with the general distribution — here appears the gap between "what gets produced" and "what lasts".
- For the format that tempts you, filter and study its longest-running examples: what do they share beyond the form?
- If the format you care about is missing from the arena, look for it in neighbouring categories on the same audience — does it live there? The answer decides whether the absence is a graveyard or an opportunity.
How to read the signals
- A format dominant in both distribution and longevity = the arena's standard — you must master it, even while testing something else.
- A format rare but present in long-running ads = an execution the veterans nevertheless sustain — a differentiation signal for whoever can afford it.
- A format dominant only in new ads = a wave of cheap production in testing — wait for longevity's vote before falling in line.
- A format absent here, alive in neighbouring categories = the chapter's most accessible differentiation hypothesis — to be tried in a limited test, with written expectations.
The typical mistakes
- choosing the format by what you like, not by what the arena votes;
- confusing numerical dominance with validation — cheap to produce doesn't mean it holds;
- ignoring your own economy — the proven format you can't afford in constant production isn't your format;
- reading a format's absence as opportunity without the neighbouring-categories check.
Exercise
Write your arena's format distribution and, separately, the formats of the first ten ads in the Exposure sort. Where the two lists differ, you've found the gap between production and endurance. Close with a written decision: your main test format and — if it exists — the absent format you'll try in a limited test.
The form is visible; what makes it work is underneath — the emotion it activates, the trust it builds, the urgency it creates. For that layer, the ad alone isn't enough: the funnel has to be opened. The next chapter: the psychology behind it.