Juniper: Turning insight into conversion
CRO
2025
✓ +34% quiz starts on LP ✓ 88% hook rate on Ad ✓ 100K+ plays / 20k Ad spend Juniper is a DTC healthcare brand focused on helping people manage their weight through clinically-backed treatment and support.
The problem:
Juniper was running every Meta ad to the same generic landing page. Menopause ad? Price page, Scientific explainer ad? Price page Food noise ad? Price page. The message that got someone to click had nothing to do with where they landed.
The Bet:
I pitched and owned a full-funnel campaign, an animation explainer paired with a dedicated landing page, both built around one insight: food noise. End-to-end: concept, script, animation, LP design, copy, imagery, build.
Audience:
We found our insight through chatting to Juniper members via user interviews, and also reading top comments within the Facebook community. People with food noise don't have language for it yet, it's usually something they've experienced but can't put there finger on it.
The Ad:
I created a new animation style to help support with our current creative diversity fatigue and help feed metas andromeda algo.
88% hook rate, people stopped scrolling.
3.5 CTR.
The Landing Page:
The page needed to do a simple job, click a food noise ad, land on a food noise page and learn about it, with Juniper coming through as the solution. The copy mirrored the ad with a clear path to start the quiz.
Testing & Results:
We ran an in-platform A/B test within Meta. The setup was straightforward:
Control: the generic landing page used across all ads
Variant: the food noise-focused landing page
We selected 10 ads with primarily food noise messaging and split traffic 50/50: half sent to the standard LP, half to the food noise LP.
The results, +34% LP → Quiz Start
Final thoughts:
This test proved matching pages to messages unlocks new audiences without hurting efficiency.
One test. One proof of concept. A playbook for every top-spending ad group. Super personalised flow for all our audience buckets moving forward.
