A store with more than 1,000 products.
The campaign selling well today is, three weeks later, pushing products that are out of stock or that people have tired of.
So the media buyer keeps going in manually to change the catalogue, removing and adding. And the moment they're busy for two days, performance drops.
Most people solve this with attention. Which means solving it with their time.
The fix here wasn't in the ad at all.
It was a Shopify Flow workflow running daily on its own. It pulls the last 30 days of sales per product. Any product selling more than 50 units gets a bestseller tag automatically. Any product falling below 50 in the last 30 days loses the tag automatically.
And the Meta campaign runs on that tag, not on a fixed catalogue.
The result is that every morning the campaign shows the products genuinely winning right now, with nobody touching it.
The 50 isn't sacred. It's a threshold sized to the store. A smaller store might use 10 or 15.
Why does this matter so much?
Because most catalogue campaigns die from exactly one thing: the algorithm keeps pushing a product that used to work. It doesn't know the product ran out or the season passed. It only knows this thing was producing results.
So when the data itself decides what gets shown, the campaign stays fresh without anyone ageing alongside it.
He notes the workflow can be built easily through Shopify Magic, and left the prompt for each step.
Which means this isn't engineering work. It's a decision to convert that manual checking into a written rule.
The good media buyer isn't the one editing daily. It's the one who builds something that doesn't need them daily.
The lesson
Automation is part of media buying
