47 Active Ad Sets and a White Background. Why This D2C Brand's Ads Were Bleeding Money.

The Brief

A D2C skincare brand came to us with a Meta Ads account that was technically functional but commercially unclear. Campaigns were active, reports looked busy, and the team had no simple way to understand which decisions were improving contribution margin.

Our mandate: restructure the account so testing, creative learning, and budget allocation could be managed with more discipline.

What Was Wrong

Three things killed their efficiency. First, they were running 47 active ad sets simultaneously with no clear testing structure — the algorithm was spread too thin to optimise effectively. Second, their creative library was almost entirely static images with product shots against white backgrounds. Third, their audience strategy was all broad targeting with no retargeting architecture.

The Framework We Built

Phase 1: Consolidation (Days 1–14)

We reduced the account to six campaigns: one prospecting campaign per product category (three total), one retargeting campaign, one retention/upsell campaign, and one testing campaign. This isn't interesting advice, but it's the single change that had the largest immediate impact on performance.

Phase 2: Creative Transformation (Days 15–45)

We rebuilt the creative library around three formats that consistently outperform on Meta for D2C: UGC-style testimonials (founder or customer, unpolished, authentic), problem-solution videos (15–30 seconds, hook in the first three seconds), and social proof carousels (before/after, results data, press mentions).

Phase 3: Audience Architecture (Days 30–90)

The retargeting architecture is where most D2C brands leave the most money. We built five retargeting audiences: website visitors (7 days), product viewers who didn't add to cart (7 days), add-to-cart who didn't purchase (3 days), purchasers (for upsell, 30 days), and lapsed customers (90 days).

The Results

The account structure became easier to manage: a clear creative testing process, a retargeting architecture that addressed different buyer states, and a prospecting approach that could be evaluated without guessing.

The lesson: most Meta accounts fail not because the platform doesn't work, but because there's no systematic process for testing, learning, and consolidating winners.