At the end of the month you open your Meta ad account. It says 40 orders and a ROAS of 3.2. Then you open Shopify and see 25 orders. Same store, same period, two completely different stories. The question that follows is always the same: which number is right, and where did the money go?
The answer is less dramatic than it seems. Neither number is lying. They measure two different things. The problem only starts when you switch off an ad based on a misreading — an ad that was actually driving sales.
Platforms measure credit, the store measures money
Meta, Google and TikTok report the conversions they attribute to themselves. If someone saw your video on Monday and then bought through search on Wednesday, both Meta and Google will claim the order. Same order, two reports, one till. So adding up sales from every platform means counting some customers twice.
The reverse is true too. Platforms don’t see everything. Blocked cookies, phone privacy settings, and customers who browse on mobile but buy on desktop all poke holes in the reporting. Plenty of campaigns bring in more than the account shows.
The one main number you can’t afford to lose
The most useful metric for a store owner isn’t the ROAS in the ad account — it’s the ratio between the store’s total revenue and total ad spend over the same period. If you spent 6,000 euros in a month and generated 42,000 euros in revenue, the ratio is 7. You track that number month over month.
The advantage is simple: no one can pretend everything is fine just because one ad set shows a pretty number. The till is the till.
A shorter attribution window tells you more about creative
When you compare individual creatives against each other, set the attribution window to one-day click and one-day view. That way you see which video actually triggered immediate action — not which one happened to be seen by a customer who was already on their way to the cart.
At Red Eye Monkey we used this approach to separate the creatives that created demand from the ones that were only harvesting it. The result was 400 percent growth in online sales. Not because we found one magic video, but because we stopped switching off the right ads based on the wrong numbers.
Two checks you can run this week
First: the checkout question. Add a “How did you hear about us?” field with five options. After two hundred orders you’ll have a picture no ad account can give you.
Second: the switch-off test. Turn one campaign off for seven days and watch the store’s total revenue. If it didn’t move, the campaign wasn’t bringing in new customers — it was paying for people who would have bought anyway. That’s the most honest metric there is, and it costs nothing.
What to ask for at your next agency meeting
If the report contains only the platform ROAS, half the picture is missing. Ask for three numbers: total revenue, total spend, and the ratio between them. Alongside that, a look at the creative: which one brought the most orders in the tighter attribution window, and how long it held up before it started to fade.
Metrics aren’t there to be perfect. They’re there to help you with a single decision: where the next euro goes.
Let’s look at your numbers together
If the numbers in your ad account and your store tell you two different stories, let’s look together at where the truth is. Book a 15-minute call or write to jaka@vvvdigital.eu.