Jild Leather

A premium catalogue competing on a crowded shelf.

Leather goods sell on craft and trust, but in the US market Jild was competing against enormous ad budgets across every surface a shopper touches. Brought in through an agency contract, the mandate was conversions — not impressions, not reach. That meant treating the product catalogue itself as the strategy and getting the data foundation right before spending a dollar at scale.

Segment the catalogue, then let bidding specialize.

Products were segmented by category in Google Merchant Center — jackets, bags, wallets, accessories — so each group could carry its own margin logic and bidding target rather than being averaged into one blunt campaign. Those segments fed portfolio bidding strategies, letting high-margin lines push harder while thinner categories stayed disciplined.

• Merchant Center category segmentation
• Feed optimization & attribute hygiene
• Portfolio bidding per product group
• Margin-aware ROAS targets

PMax to convert, DV360 to surround.

Performance Max campaigns were scaled gradually over time — expanding budget only as each product group proved it could hold its ROAS target. Alongside that, DV360 programmatic ran display and YouTube for prospecting and consideration, with cart retargeting closing the loop. Audiences were built from interest signals and real search terms, so targeting followed demonstrated intent rather than assumed demographics.

• Performance Max, scaled in stages
• DV360 display, video & retargeting
• Audiences from interest + search terms
• Search-term mining & negatives

CAPI, clean events, and a system that runs itself.

Conversion tracking was rebuilt properly: Meta’s Conversions API sending purchase, add-to-cart and checkout events server-side, mirrored into GA4 and Google Ads. That protected event quality against browser privacy limits and gave the bidding algorithms honest signal to optimize on — the difference between automation that compounds and automation that drifts.

• Meta CAPI server-side events
• Event-match quality monitoring
• GA4 + Ads conversion alignment
• Automated reporting cadence

306 conversions on $29.5K, and a machine that scales.

The account now runs as a system rather than a set of campaigns — segmented feed, portfolio bidding, staged PMax expansion and programmatic support, all reading from clean server-side data. Every new product line plugs into the same structure instead of starting from zero.