the Google account reported 9.16x. Split by search term, 87.0% of that value came from people typing the brand’s own name, and everything else returned 1.62x. We reported the split rather than the blend. In the twelve months to 26 August 2026 the store took $1,661,291.68 against $1,491,343.64, up 11.4%, on ad spend of $66,838.61.
Key results.
On Shopify’s own numbers, not platform attribution:
Challenge
Aaron Basha is a family house founded in 1906, known for its enamel baby-shoe charms. On the first call, 1 May 2025, the owner described the account she was paying for: “I have no idea what the hell is going on with it.” She had cut her spend out of caution rather than evidence, a freelancer hired the month before had gone quiet, and Mother’s Day, which she called “my Christmas”, was ten days away.
The account was effectively one campaign. Thirty existed, two were enabled, one was spending, with no non-brand Search, one negative keyword, no shared negative lists. Meta was not running at all. The catalogue was worse: of 1,510 items in Merchant Center, 551 were not eligible and 61 limited, so 40.5% could not enter an auction, and 531 pointed at dead pages. Merchant Center itself filed those dead links under low impact, below five clicks, which is what a product that cannot serve will always have.
Her name was undefended: the branded campaign had served $0 on a stale disapproval while a reseller ran 10% off against Aaron Basha searches.
Solution
Step #1 Report the split, not the blend.
The audit window, 30 June to 29 July 2026, showed $2,710.66 returning $24,826.74, or 9.16x, the number most agencies would have sent. Split by search term it was brand at 30.01x, carrying 87.0% of the value on 26.6% of the spend, against 1.62x on the other 73.4%. Two single days were 62.8% of the month’s value; without them it reads 3.41x. The rule since: quote the split, never the blend, and judge this account on 60 to 90 day windows. It also ruled out the textbook restructure, since negating brand out of Shopping would have deleted 87% of the account’s value.
Step #2 Fix the feed where it broke.
The 531 dead links were verified three independent ways that agreed, and the cause sat upstream: the feed’s primary source was a hand-maintained Google Sheet, blind to a Shopify deletion. On 13 August 2026 we retired it, made the Shopify sales channel primary and resynced all 589 products. That sheet also held the account’s best asset, a curated title reading “18K Pink Baby Shoe Charm & Diamond Strap | Aaron Basha” in place of Shopify’s “The Mary Jane”, worth $7,427.50 on $75.57 of spend in thirty days, so 821 curated titles went into a supplemental feed before the switch.
Step #3 Defend the name, then add the second channel.
We rebuilt the disapproved branded ad and moved Brand onto Target Impression Share. Google’s auto-apply had pulled 11 keywords out on 4 August, including the exact-match brand term; we restored all 11 and turned auto-apply off. Meta launched from zero in January 2026. On 13 August we also paused eight of our own live Meta ads, one producing 11 of the account’s 33 purchases, over a “Starting at $75” line when the entry rung was $125.
Results
Book a discovery call about your business goals
- ✓Non-binding introductory meeting
- ✓Explore your growth potentials
- ✓Talk to our experts