Tripling INSÉRIF’s revenue year-on-year across three channels

We rebuilt the account for scale, consolidating learnings into top campaigns and leaning into broad targeting and ASC+ to let Meta find efficiency at volume.

We created a worldwide catalogue using country-override feeds and condensed all catalogue activity into a single campaign for cleaner signal and stronger delivery. We then split prospecting by region (UK, USA, EU and ROW) to scale each market on its own merits, and launched a dedicated worldwide testing campaign with fresh concepts isolated in separate ABO ad sets to read true winners.

The result: Meta revenue grew 426% year-on-year, and remarkably ROAS improved around 30% even as we quadrupled spend.

We scaled search and Performance Max to capture surging demand, restructuring campaigns for control over new-vs-returning customers and feeding Performance Max with shopping data to prioritise winning SKUs.

We optimised product feeds to lift Performance Max output and built asset groups to best practice, pairing ad copy, video and images, with custom labels split by product category and seasonality for sharper control.

We cleaned up brand + generic search to improve ad rank and Quality Scores, tested AI Max, and layered new-customer-acquisition optimisation across all campaigns to bias spend toward incremental customers.

The result: Google revenue grew 256% year-on-year as we scaled spend roughly fivefold to capture market share.

We launched TikTok as a brand-new acquisition channel and scaled it from a near-standing start into a core part of the mix, driving thousands of purchases a month where there had been virtually none.

We utilised spark ads, catalogue campaigns and Smart+ as well as providing the algorithm with localised catalogues to scale internationalisation efficiently.

We overhauled the creative engine to feed three channels at once, building concepts off customer pain points and product benefits, scaling winners fast, and repurposing UGC and creator content across Meta and TikTok to keep the testing pipeline full and the cost of testing low.