E-commerce & Retail · India · online
GIMBAS Perfumes
E-commerce Meta Ads Strategy + AI UGC Production System · 18 pages · 2026
“AI should make honest advertising cheaper to produce, not make dishonest advertising easier to hide.”
01 · The challenge
In a store, GIMBAS has a tester strip, a salesperson and a conversation. Online, the ad, the product page and the offer have to do all of that work, for a product the buyer can’t experience before paying.
02 · The insight
In e-commerce, creative is the targeting. So build a creative engine, not a set of ads. And the store is a free UGC factory.
Once audiences are broad and the account is simple, the advertisement becomes the main thing you control, so creative volume becomes the growth constraint. Ten good variations a month beat two beautiful ones a quarter. And GIMBAS has an advantage most online-only brands pay for: every customer reacting to a first spray in-store is a potential ad.
03 · The system I built
- 01A three-layer funnel in which cold, warm and hot audiences each get a different job and a different creative
- 02A four-campaign account: broad prospecting, a permanent creative-testing lane, catalogue retargeting, retention
- 03A tracking foundation where every line is marked “to be validated” before a single rupee is scaled
- 04An offer built around a Discovery Set, which solves the “can’t smell it” objection instead of arguing with it
- 05A 15-second performance cut specified second by second, and an eight-hook library shot against one body of footage
- 06Four UGC sourcing routes and a six-step AI video pipeline with a human review step
- 07A written AI-creative ethics policy: no synthetic customers, no fake reviews, no fake product shots
04 · What was delivered
- 18-page strategy
- Account structure
- Tracking checklist
- Offer architecture
- Hook library
- UGC briefs
- AI production pipeline
- AI ethics policy