Its ads, orders, stock and returns lived in different places. Read together, they showed money going to the wrong products and stock with nothing selling it. In one week in September, real orders came to 4.4 times the ad spend.
The retailer sells online through Shopify and advertises on Meta and Google. Each ad platform reported its own sales figure. Stock and returns sat in the shop, where neither ad platform could see them.
So no single number said whether an ad had paid for itself. We connected to all three, read-only, and began checking each one against the others every morning.
The first four came in the first month. The last two came from the wider read of stock and customers.
They looked like they sold. Once refunds were counted, each sale cost more than it made, and the ads were still paying to find more.
Ad spend by product against margin after returns · first monthMeta counts a sale if someone saw an ad and bought later, even without clicking. Some of them would have bought anyway, so its figure runs high.
Meta’s reported sales, split by click and view · first monthMoney spent showing people something they could not buy.
Live ads against stock in the shop · first monthThe stock was there and nobody was being shown it. By 24 September every in-stock line was reaching buyers on Meta and Google Shopping.
Stock against ad coverage · 14 Aug and 24 Sep 2026Counted from first-ever orders, net of discounts and refunds, before VAT and shipping. The ads were finding new customers. Repeat buying was the smaller share, with room to grow.
Shopify net sales · 180 days to 23 Sep 2026Anything that needs a judgement still goes to the owner. The rest is done before they wake up.
Break-even is the return below which the ads lose money on this retailer’s own margins. The figure counts all Shopify revenue in the week, so it includes some sales the ads did not cause. That is why each product is also checked against its own break-even.
Shopify revenue over Meta and Google spend, 17 to 23 September 2026. Break-even from the retailer’s own margins.
Orders, refunds, stock, ad spend and every live ad, each checked against the others.
Sold-out products come out of the ads and old prices are caught. The owner is told, not asked.
Return on real orders against break-even, what was fixed, and the few calls that need a person.
Keep, cut or reprice a product below its break-even. Approve new ads before they run. The rest of the day is theirs.
A person reads your Meta and Google accounts, read-only, against your real orders. You get what they really return, what is wasting money, and the first six things we would change.
We will email to arrange read-only access. The audit takes 3 to 5 working days once we can see the accounts.