Shopify Facebook ads: how to set up a store for Meta, and what decides whether it sells

To run Facebook ads for a Shopify store you need three things working before you spend: tracking that reports purchases back to Meta, a clean product catalogue, and one simple campaign that is not starved of budget. Most of the setup is a one afternoon job. The part that decides whether the ads make money is the creative, and that is the part almost every setup guide skips.

This piece covers the setup quickly, then the bit that matters more. We run Meta ads for ecommerce brands every week, so the observations below come from live accounts, not from a tutorial.

What do you need in place before you spend on Meta?

Three things, in this order.

1. Purchase tracking you can trust. Meta can only optimise toward sales it can see. The usual route is Shopify's own Facebook and Instagram sales channel, which sets up the pixel, server side event sharing and the catalogue in one flow (OptiMonk's Shopify guide walks through it). The standard events to confirm are page view, view content, add to cart, initiate checkout and purchase.
2. A catalogue with no rejected products. Products sync from Shopify to Meta automatically. Variants can sync as separate items, which inflates the count, and rejected items quietly drop out of your ads, so look for them before launch.
3. A budget that matches the plan. One guide gives the arithmetic: splitting a small daily budget across five ad sets leaves each one with a fraction of what it needs to learn (Easysell's 2026 setup guide).

A caution on those sources: they are written by app vendors and agencies, and menu paths and default settings change. One of them reports that Shopify changed a default pixel setting in January 2026 that can suppress some events. We could not confirm that against Shopify's own documentation, so check your own event log rather than trusting any guide, ours included.

The test is simple. Place a real order on your own store and see whether it shows up in Meta as a purchase. If it does not, nothing you build afterwards will optimise properly.

How should the first campaign be structured?

Simply. Brand new accounts do best with very few moving parts: one campaign aimed at purchases, a small number of ad sets, and enough budget in each that it can actually spend. The guides that suggest starting with an automated shopping campaign are describing Meta's default direction, where the system finds buyers for you and the ad itself does most of the targeting work.

That last point is the one worth sitting with. With broad targeting, you are not choosing the audience any more. The ad is. Meta reads the picture, the words and the opening seconds of a video to decide who sees it. We went through what that means for volume in how many ads you should run on Meta.

So the real structural decision is not ad sets. It is how many different ideas you give the account to work with. A launch with one image and one video is a launch with one idea, and when it fails you learn almost nothing.

Why do two Shopify stores with identical setups get different results?

Because the setup is the same and the ads are not. Tracking, catalogue and structure are table stakes. Every serious store has them, so they stop being an advantage the moment you have them.

What separates accounts in our experience is which message the ad leads with. A few things we have seen recently, anonymised and described against each account's own target rather than in money:

• A claim shown beat a claim stated. On a UK luxury home brand, the argument that a small dose per wash makes the product last ran two ways. Filmed as a side by side comparison, it beat target. Written as a static headline, it spent and brought almost nothing back. The same words, a different outcome, because one showed it working. We wrote that up in the Christmas gift ads issue of August Weekly.
• Proof beat a guarantee. On the same brand, ads built on social proof came in above target, while ads built on a money back promise took far more of the budget and returned about half of what was needed. Details in our issue on proof against guarantees.
• A stranger beat the founder. The ad that came closest to target on a premium brand one week was a street interview with passers by reacting to the product. Founder led ads on the same account came in much lower (that issue is here).

None of these is a setup problem. A store could have perfect tracking and still lose on every one of them.

What creative should a new Shopify store launch with?

Not one ad, and not twenty ads that are the same ad. Launch with a small set that tests different reasons to buy.

• Different buyers. The person who hesitates on price needs a different ad from the person who already loves the category. In our accounts the ad written for the hesitant buyer has been among the strongest more than once, on more than one premium brand.
• Different formats. A static and a video of the same idea do not behave the same way. We covered when each wins in static against video ads on Meta.
• Different proof. Reviews, a demonstration, a stranger's reaction and a before and after are separate bets. Treat them as separate ads, not as options inside one.

Then read results against your own target cost per purchase, not against a benchmark from another industry. An ad that has spent less than about three times your target cost per purchase has not told you anything yet, in either direction. Calling it early is how stores kill ads that would have worked.

What goes wrong in the first month?

Four things come up again and again.

Judging too soon. Daily results on a small budget swing wildly. A single bad day says nothing.

Warm audiences treated as free money. It is tempting to assume past visitors and buyers will convert cheaply. On one account we run, ads aimed at past buyers and engaged people returned well under a third of target, while several cold audiences cleared it. Retargeting is not automatically your cheapest sale.

Too few ideas. One winning ad fades. The bedtime ad that led a kids' supplement account two weeks earlier had slipped by the next read, and the account needed a new moment to replace it. If you only have one winner, you have a countdown.

Tracking assumed, not tested. The cheapest fix in the whole list is the one people skip: check that purchases are counted before you read a single result.

Where does this leave a new Shopify store?

With a clear order of work. Get tracking and the catalogue right once, keep the first campaign simple, and then put most of your effort into the ads, because that is where the results move. If you would like someone to look at your store and your numbers before you commit to anything, we do a free look, and you can see what this has meant for other brands on our case studies page.

FAQ

How do I set up Facebook ads for a Shopify store? Install Shopify's Facebook and Instagram sales channel, confirm that a test purchase reaches Meta, check your catalogue for rejected products, then launch one simple campaign optimised for purchases with enough budget to spend.

Do I need the pixel and server side tracking for Shopify? Yes, you want both reporting purchases. Browser tracking alone misses sales, so server side sharing gives Meta a fuller picture to optimise from. Test with a real order to confirm.

How much should I spend to start on Shopify Facebook ads? Enough that each ad set can actually spend and learn. Guides we read suggest a modest daily budget on a single campaign. Splitting a small budget across many ad sets starves all of them.

How many ads should a new Shopify store launch with? A small set that tests different ideas, not one ad and not many copies of the same one. Vary the buyer, the format and the proof so you learn which reason to buy works.

How long before I know if the ads work? Judge each ad on its own spend. Until an ad has spent about three times your target cost per purchase, it is too early to call either way.