ECOMMERCE · Meta
Set up Meta ads for an ecommerce store.
Four checks in order: whether your account is ready for Advantage+ Shopping or should stay manual, whether your catalogue can carry dynamic product ads, whether your pixel and event match quality are good enough to optimise on, and how to add budget without resetting what the algorithm has already learned.
Built by Jonas Sluijs, former Meta growth lead, $500M+ in ad spend.
What this is not. It does not connect to your ad account, your pixel or your catalogue. It applies ecommerce setup rules to the numbers you type, so it can tell you what your purchase volume supports but not whether your tracking is actually firing. Confirm that in Events Manager. For a review of your real account, book a free strategy call.
The order these four things have to happen in
Ecommerce on Meta fails in a predictable order, and fixing things out of sequence wastes money. Purchase signal comes first, because every other decision depends on it. Meta optimises toward the event you send it, so if purchases are under-reported or badly matched, the algorithm is optimising toward a distorted picture of who buys.
Catalogue comes second, because dynamic product ads are usually the lowest-CPA format available to a store, and they simply cannot run without a connected feed. Campaign type comes third, and scaling comes last, because scaling an account whose signal is broken only breaks it faster.
Advantage+ Shopping or manual
The deciding factor is conversion history, not preference. Meta's own guidance puts Advantage+ Shopping at its best for stores with existing conversion data, around 50 or more purchases a month, and explicitly does not recommend it for brand-new accounts with no history. Meta's published figures are a 17 per cent lower cost per acquisition and 32 per cent better ROAS against standard manual campaigns, which is worth having once you qualify and worth nothing before you do.
| Recent purchase volume | Usually the right call |
|---|---|
| Around 50+ purchases a month, consistently | Advantage+ Shopping, with manual running alongside for control |
| Some history, inconsistent month to month | Manual with broad targeting; revisit once volume steadies |
| Little or no purchase history | Manual, one ad set, and the budget concentrated so learning can complete |
The full comparison covers the cases where both should run together, and what to check when Advantage+ performance drops.
Why event match quality decides your CPA
A purchase event that Meta cannot attach to a person teaches it nothing. Event match quality measures how well the identifiers you send, email, phone, click ID and so on, match a real account. Raising it is usually the cheapest performance improvement available to a store, because it improves every campaign at once rather than one ad set. Server-side events through the Conversions API are what move it, since they survive iOS restrictions and ad blockers.
Common questions
Does this connect to my store or my ad account?
No. There is no login, no OAuth, no Shopify or ad-account connection. It applies ecommerce setup rules to the numbers you type on this page. Nothing is read from your systems and nothing is written to them.
Should I run Advantage+ Shopping or manual campaigns?
It depends on recent purchase volume rather than preference. Advantage+ Shopping needs enough recent conversions to learn from, and below that threshold broad manual targeting usually performs better. Many stores run both, with Advantage+ taking the bulk of spend once volume supports it. The comparison guide works through the cases.
Why is my catalogue not delivering dynamic product ads?
Usually the feed, the events, or the match between them. Dynamic product ads need product IDs in your ViewContent, AddToCart and Purchase events that match the IDs in your catalogue exactly. If they disagree, Meta cannot work out which product to show and delivery stays near zero. Check the feed refresh frequency too, since out-of-stock products kill performance quietly.
How much can I raise budget without resetting learning?
Raise gradually and infrequently rather than in one jump, and change one thing at a time. Large edits push an ad set back into the learning phase, where cost per purchase rises until it stabilises again. If you need a big increase, duplicating into a new ad set is often safer than editing the proven one. Scaling kills performance covers the mechanism and duplicate or increase budget covers the choice.
My ROAS looks fine in Ads Manager but the business is not growing.
Reported ROAS counts conversions Meta believes it caused, inside an attribution window you chose, and it can include purchases that would have happened anyway. Compare it with blended numbers: total revenue over total spend, and new-customer CAC. The MER and nCAC calculator works those out and the measurement guide explains why the two disagree.