An Australian menswear brand, scaled on profit per dollar, not vanity ROAS.

Spend that could not be trusted, so it could not be scaled.

Peter Jackson had a healthy online business and an established brand, but the paid-media account was being steered on numbers it could not stand behind. Conversion tracking double-counted and mis-attributed, so the reported return was not a number anyone could confidently fund against. When the truth of a dollar is uncertain, every decision to scale becomes a gamble.

Underneath that, the structure was working against itself. A large product catalogue with size and colour variants meant the shopping feed carried heavy duplication, splitting performance signals across near-identical items and starving the campaigns of clean data. And a meaningful slice of the budget was being spent recapturing demand the brand would have earned anyway, which inflated the headline return while doing little to grow the business.

The brief was direct: make the numbers honest, then scale them. Prove that paid media was adding incremental customers, not just claiming credit for them, and do it at a return the business could keep reinvesting against.

Fix the measurement, fix the feed, then fund the growth.

The work started with the foundation that everything else depends on: conversion measurement. We reset how conversions were tracked and valued so a reported dollar matched a real dollar in the back end. Once the measurement reconciled, the return figure became a number the business could actually budget against rather than a directional guess.

With clean signal restored, we rebuilt the shopping feed. Size and colour variants were de-duplicated so each product presented as one coherent unit, which concentrated the performance data instead of scattering it across dozens of near-identical listings. A cleaner feed gives the bidding system fewer, stronger signals to learn from, which is what lets a large catalogue actually compound.

We then rebuilt Performance Max so it was structured to be funded rather than fragmented, and removed branded targeting from the paid mix. Bidding on your own brand often buys back demand you would capture for free, which props up the reported return without growing the customer base. Stripping it out forces the account to prove genuine, incremental acquisition.

On Meta, we built a full top, middle and bottom of funnel structure anchored on Advantage+ Shopping, so the brand was introducing itself to new audiences, nurturing consideration, and closing intent rather than relying on a single conversion campaign to carry everything.

Holding it all together was the analytics layer. Rather than scale on ROAS, which can rise while the business stalls, we modelled value to acquisition cost on a 90-day horizon: what a newly acquired customer is genuinely worth over their first three months against what it cost to win them. That ratio became the throttle. Where the 90-day value justified the spend, we scaled hard; where it did not, we held. Budget grew because the model said the customers were worth more than they cost, not because a vanity number looked good.

• Reset conversion measurement so reported value reconciles to real back-end revenue.
• De-duplicated the product feed across size and colour variants to concentrate the learning signal.
• Rebuilt Performance Max to be fundable and removed branded targeting to force incremental acquisition.
• Built a full TOF, MOF and BOF Meta funnel on Advantage+ Shopping.
• Scaled against 90-day value to acquisition cost, not headline ROAS.

A return the business could keep reinvesting against.

Once the account was funded against a number it could trust, it scaled. Paid media returned 11.55:1 across the account and tracked A$942,000 in conversion value through 3,880 conversions, while cost per conversion fell to A$21.01, a reduction of A$4.76. Budget was increased materially while the return held, which is the proof that the growth was structural rather than a one-off spike.

Because the budget was governed by a 90-day value model rather than a daily ROAS target, the scaling held as spend increased. The account stopped being a number nobody trusted and became a growth lever the business could lean on, which is what earned the brand a Global Search Awards nomination for the work.