Ad Campaign Critique — by an AI council
Get an ad campaign torn apart before you spend the budget. Four rival AIs debate it and hand back: A critique with the strongest angle, the fix list and what to test.
A critique with the strongest angle, the fix list and what to test.
Signed off by an independent Final QA audit — with a “verify before you rely on this” list, never a black box.
Why this team
Ad money is spent on hunches dressed as strategy, so each seat attacks a different hunch: the Performance Marketer audits the media plan against the objective, David Ogilvy judges whether the ad makes a real promise or just decorates, and the Copywriter with the Conversion Strategist follows the click all the way to the landing page — where most campaigns actually die. The Customer Psychologist tests whether the message lands with the person seeing it, and the Data Skeptic asks how you will even know if it worked.
Each seat runs on a different frontier model — GPT, Claude, Gemini, Grok — so it’s a genuine cross-model review, not one AI in costumes. How a run works →
What the debate sounds like
An illustrative excerpt — A DTC skincare brand is about to put $15,000 behind a video campaign whose hook is the founder’s lab background, targeting broad 25–45 on two social platforms with the goal of “brand awareness and sales”.
The advertisement has no promise. A founder in a lab coat is a credential, not a benefit — nobody ever bought a face cream because the chemist looked serious. What does the product do for her skin that she can see in the mirror? Lead with that, prove it with the lab coat second. Credentials support promises; they do not replace them.
“Awareness and sales” is two campaigns wearing one budget, and $15,000 is not enough to do both. Awareness buys reach you cannot measure at this spend; sales needs conversion-optimised delivery and a tight audience. Pick sales, narrow 25–45 to the segment the product actually serves, and let the platform optimise against purchases, not views.
The lab-credibility angle is not worthless — it answers a real anxiety, “is this brand safe or snake oil?” — but that is a mid-funnel objection, not a scroll-stopper. The stopping moment for skincare is recognition: her specific problem named in the first second. Lead with the problem she has, answer the trust question one beat later.
Questions
What materials does the council need to critique a campaign?
The creative (paste the script, copy or attach frames), the audience and channel plan, the offer, the budget and the objective — plus the landing page the click hits, because the critique follows the whole chain. Campaigns mostly fail after the click, and a creative-only review misses that.
Can it critique the campaign before I’ve made the creative?
Yes — run it at the angle stage. The synthesis names the strongest angle to lead with and two or three variations worth testing, which is cheaper than discovering the angle problem after production. Teams often run this brief twice: once on the concept, once on the finished creative.
Will it tell me where the budget should actually go?
It will tell you whether the budget matches the objective — the most common finding is a budget split across goals or channels that undermines both — and what to test first at your spend level. What it will not do is invent performance predictions; the deliverable includes the metric that decides the winner in the real data.
Your material is used only to run your review — never to train public models. Encrypted in transit and at rest. Security & privacy →
Want full control — pick your own minds, set the depth? Open the full council →