Persona

Video ad formulas for agencies

Agency creative carries a constraint no in-house team shares — every beat has to pass a client review before it reaches an auction. That filter shapes the structure as much as the brief does.

34 ads decodedValidation leads at 14%44s average

How should agencies structure their video ads?

Across 34 decoded ads, agencies run about 44 seconds across 6.7 of the seven beats, with Validation taking the largest share at 14%. Agency-run ads carry heavier Context and Validation than the cross-category average, because both beats are the easiest to defend in a review. The cost is a thinner Opening — the beat clients are most likely to soften. The constraint that shapes it: Every hook passes a non-expert approval round before it reaches an auction.

Derived from 34 decoded video ads run by agencies.

Ads decoded

Avg runtime

44s

Avg beats

6.7

Peak beat

Validation

Key takeaways

34 decoded ads — Validation takes the most runtime at 14%.
Average ad runs 44 seconds across 6.7 of the seven beats.
Beat Efficiency peaks at 1.15x on Validation and bottoms at 0.94x on Shift.
Every hook passes a non-expert approval round before it reaches an auction.

Time allocation

Where the seconds go

Share of runtime per beat, with the cross-category baseline marked on each bar.

Opening14%
Context21%
Tension12%
Delivery20%
Validation14%
Shift10%
Close9%

Opening14%avg 14.8%-0.8

Context21%avg 19.6%+1.4

Tension12%avg 12.4%-0.4

Delivery20%avg 20.9%-0.9

Validation14%avg 12.2%+1.8

Shift10%avg 10.6%-0.6

Close9%avg 9.5%-0.5

The white tick on each bar marks the cross-category average. Green deltas are beats this persona invests more in than a typical ad.

Diagnostic

Beat Efficiency Ratio

Runtime share divided by the cross-category average. The tick on each dial marks 1.0x.

Opening

Context

Tension

Delivery

Validation

Shift

Analysis

Why agencies creative looks like this

The decoded agency set shows a consistent signature: strong middles, cautious openings. Context and Validation both run above baseline because those are the beats a client can evaluate against something they already believe about their brand. The Opening is where approval rounds do the most damage, and it shows in the numbers.

The practical response is to build the interrupt into the proof rather than presenting it separately. A surprising, checkable number is simultaneously defensible in a review and disruptive in the feed — which is why data-led openings are the most common agency structure in the decoded set.

The second lever is presenting structure rather than execution. A client who is shown a beat map and the reasoning behind each beat argues far less about the specific opening line than one who is shown three scripts.

Reality check

The constraints you're actually working under

These explain more of the gap between good creative and shipped creative than any category difference.

Every hook passes a non-expert approval round before it reaches an auction

Brand-safety limits remove most pattern-interrupt and negative openings

Creative volume is capped by client turnaround, not by production capacity

Application

What to prioritise

Protect the Opening in the brief

Agree the hook's job before the creative exists. A hook that gets softened in round two is the single largest performance loss in agency creative.

Lead reviews with the beat map

Presenting the structure first moves the conversation from taste to function, which is where an agency actually has the expertise advantage.

Bank Validation early

Proof beats are the least contested in review and the most portable across a client's whole account. Build a reusable library of them.

Failure modes

How agencies creative goes wrong

Designing for the pitch deck

A structure that presents well in a boardroom and predicts well on paper still has to beat a stranger's thumb. Those are different tests.

Letting the client write the hook

Clients reliably optimise the Opening toward brand comfort and away from performance. Own that beat contractually if you can.

How this was produced

Based on 34 decoded video ads, observed in public ad libraries and creator platforms and read by the Klip Kanvas research team. Each ad is classified by opening hook, creative format, persuasion mechanism and close using the taxonomy shared across this library, which is what makes patterns comparable between brands and categories.

These are editorial research figures intended to guide creative decisions. They are not audited platform metrics, they do not come from any advertiser's ad account, and they describe the sample named above rather than a brand's complete advertising. Test against your own audience before treating any of it as a rule.

Frequently asked questions

What ad structure works best for agencies?

Agency-run ads carry heavier Context and Validation than the cross-category average, because both beats are the easiest to defend in a review. The cost is a thinner Opening — the beat clients are most likely to soften. Across 34 decoded ads the average runs 44 seconds over 6.7 beats, weighting Validation to 14% of runtime — 1.15x the cross-category average.

Why do agencies lean on Validation?

Proof that it worked for people like the viewer — reviews, results and receipts arriving right when skepticism peaks. The decoded agency set shows a consistent signature: strong middles, cautious openings. Context and Validation both run above baseline because those are the beats a client can evaluate against something they already believe about their brand. The Opening is where approval rounds do the most damage, and it shows in the numbers.

How long should a agencies ad be?

Decoded winners average 44 seconds — shorter than the 46-second cross-category average — across 6.7 of the seven beats. Extending past that adds little, because the surface and the funnel do not reward it.

What constraints shape agencies creative?

Every hook passes a non-expert approval round before it reaches an auction. Brand-safety limits remove most pattern-interrupt and negative openings. Creative volume is capped by client turnaround, not by production capacity. Those constraints matter more than category convention — they explain why this persona's shape differs from the vertical it happens to be selling into.

Why is Shift compressed for agencies?

At 10% of runtime — 0.94x the baseline — Shift is the beat this persona leans on least. A structure that presents well in a boardroom and predicts well on paper still has to beat a stranger's thumb. Those are different tests.

Should I follow my category shape or my persona shape?

Start from the persona when your constraints are unusual, and from the category when they are not. An agency running ads for a supplement brand faces the supplement category's buying friction and the agency persona's approval constraints at the same time — the practical answer is to take the beat order from the category and the runtime discipline from the persona.