Pick the campaign before the shoot
Show your audience the ideas while they are still scripts, boards and animatics. Learn which one they remember as yours, what they take from it, whether they believe it, and whether it moves them.
- The family road trip
A minivan, a long drive, one box of cereal that lasts the whole way.
Remembered it, and named30% us17% a rivalof 58% who recalled itTook away the intended message46%Believe it41%Would pick us next shop, vs. control+3 ptsRevise: show the box in the first scene“Family time. I thought it was for the big-name brand.”
- The 6 a.m. kitchen
A tired parent, three kids, breakfast on the table in two minutes.
Remembered it, and named41% us4% a rivalof 49% who recalled itTook away the intended message72%Believe it68%Would pick us next shop, vs. control+9 ptsTake forward“It's fast and the kids will actually eat it.”
- The celebrity cameo
A famous chef tries to make breakfast and gives up.
Remembered it, and named21% us12% a rivalof 70% who recalled itTook away the intended message33%Believe it29%Would pick us next shop, vs. control+1 ptDrop: the chef is the story“That the chef is funny. I don't remember the cereal.”
The cameo was remembered most and sold least: people recalled the chef, not the cereal. Next, 40 real parents talk through the kitchen and the road trip with the AI interviewer overnight.
Run it with
Methods
Monadic and sequential monadic design
Each person sees one idea first for a clean read, then sees the rest and picks a favorite.
Branded recall and misattribution
Shows how many people remember the idea, how many name your brand, and how many credit a rival.
Key message takeaway
People play the idea back in their own words, coded against the message you meant to send.
Believability and persuasion vs. a control cell
Measures belief in the claim and the shift in brand choice against people who saw no idea.
Diagnostic interviews
Real people talk through likes, dislikes and confusion, so you know what to fix before production.
Remembered is not the same as remembered as yours
A memorable idea can still sell for a competitor, or let a celebrity steal the scene. Jean splits recall into people who named your brand and people who named someone else.
Test early, while changes are cheap
Twins react to scripts, boards and taglines in minutes, so you can drop or rewrite an idea the same afternoon. An agency pretest usually waits for finished work and takes weeks.
Their words for what it says
The AI interviewer asks real people from your audience to play the idea back in their own words. You hear the misreads before the media money is spent.
How you run it
- 1Minutes
Screen
A thousand digital twins answer first, so the weak options drop out.
- 2Overnight
Interview
The AI interviewer talks to real people and follows up on what they say.
- 3By morning
Decide
The number, the reasons behind it and the quotes that carry it.
Questions
What is campaign concept testing?
Showing early campaign ideas to your audience before production, to learn which one is remembered as yours, understood, believed and persuasive.
What metrics does an ad pretest report?
The usual set is branded recall, key message takeaway, believability, likeability and persuasion, plus diagnostics on what people liked, disliked or found confusing. Jean reports each one per idea, not one blended score.
Can I test storyboards and animatics, or only finished ads?
Rough work is the point. Scripts, storyboards, key frames, animatics, taglines and still ads all work. Jean asks people about the material; it does not track eye movement or measure attention second by second.
Monadic or sequential monadic: which should I use?
Monadic, where each person sees one idea, gives the cleanest read on recall and belief. Sequential monadic shows the rest afterward and tells you which one they would pick. Jean can run both in the same study.
Do you compare results to norms?
Jean does not have an ad norms database. Ideas are compared with each other and with a control cell that saw no idea. If you have results from past pretests, Jean can use them as your benchmark.
How is this different from an agency pretest?
An agency pretest usually runs on near-finished work and takes weeks. Jean screens rough ideas on twins in minutes and confirms the leaders with real interviews overnight, early enough to change the work.