Why Meta Shows More Conversions Than Shopify, and Which to Trust

Meta counts clicks, views and modelled sales. Shopify counts orders. Why the two numbers differ, which to trust, and how to measure Meta ads after iOS 14.

By Yannis Spanenburg · 2026-08-04 · 8 min read

Say Meta reports 500 purchases for last month and Shopify shows 350 orders. Nothing was stolen and neither system is broken. They count different things, and the space between them is where a lot of ecommerce budget gets misspent.

Why does Meta show more conversions than Shopify?

Meta counts a purchase when it happens within its attribution window after someone clicks, engages with or views an ad, and it adds conversions it models but cannot observe. Shopify counts completed orders, once each. So Meta's number is usually higher, and a gap by itself does not mean anything is broken.

Where your gap comes from depends on which Shopify number you are comparing. If Meta claims more purchases than your store recorded in total, something is counting twice: most often a purchase event sent from the browser and the server without a shared event ID, or fired again on every reload of the thank-you page. If Meta only claims more than Shopify credits to Meta, that is attribution: Meta counts views, engagements and clicks from days earlier, while Shopify's own reports give the order to the last click that brought the buyer in.

Key takeaways

• A gap between Meta and Shopify is normal: Meta counts purchases inside its attribution window, including modelled ones, while Shopify counts completed orders.
• If Meta reports more purchases than your store recorded in total, check for a purchase event that counts twice before you change any budget.
• Since March 2026 Meta's default attribution is 7-day click-through, 1-day engage-through and 1-day view-through, and only link clicks count as clicks.
• Safari limits cookies set by scripts to seven days and iPhone users can refuse tracking, so the pixel alone now misses real sales.
• Trust your store or your accounting for what happened, judge total spend on blended return, and use Meta's numbers to steer inside Meta.

What changed in Meta attribution in 2026?

Two things. On 12 January 2026 Meta removed the 7-day and 28-day view-through windows, and in March 2026 it narrowed click-through to link clicks. Likes, comments, shares and saves now count as engage-through, with a one-day window. The default is 7-day click-through, 1-day engage-through and 1-day view-through.

Jon Loomer's guide to Meta attribution in 2026 walks through each setting. The practical effect is that Meta's reported purchases can drop with no change in real sales, so compare periods either side of these dates with care, and use Compare attribution settings in Ads Manager to see how much of a result comes from clicks and how much from views.

Why did the tracking gap get so wide?

Because browsers and phones now keep tracking data for a shorter time, or not at all. Safari limits cookies set by scripts, iPhone users can refuse tracking, and consent banners hold pixels back until a visitor accepts. Each one hides real sales from the pixel, and Meta fills part of the hole with modelled conversions, which are estimates rather than recorded orders.

Browsers shortened the memory

Since WebKit's Intelligent Tracking Prevention 2.1 in February 2019, Safari caps cookies set through JavaScript at seven days. A customer who clicks your ad on Monday and buys three weeks later arrives, as far as the pixel can tell, from nowhere.

iPhone users can opt out

Since iOS 14.5, released on 26 April 2021, Apple's App Tracking Transparency has required apps to ask before tracking people across other companies' apps and websites. When someone taps no, Meta can no longer follow them from an ad in its apps to a purchase on your site the way it used to, and it fills more of the gap with modelling.

Your consent banner sits in front of your pixel

If you sell into the EEA, your cookie banner is part of your measurement. Google has required consent signals for EEA traffic since March 2024 for its measurement and ad personalisation features to keep working. Set up badly, a banner loses your data with no visible error. Set up well, it passes a consent signal and lets the platform model around the gap.

Every platform claims the same sale

Meta credits itself for a purchase inside its window. Google credits the same order, and so does your email tool. Add the dashboards together and the total comes out higher than what you sold.

Which number should you trust?

Trust your store or your accounting for what happened, and the ad platform only for steering inside that platform. Meta over-reports because it credits every sale it touched and models the ones it cannot see. Shopify's own channel reports under-credit ads because they hand the sale to the last click.

The hardest number to dispute is blended return: total revenue divided by total ad spend across every channel. It cannot be double counted or modelled. Judge the overall budget on it, and use Meta's own figures to choose between campaigns, audiences and creatives, the way we tested creatives and audiences in our Meta ads work for Decrypting.

To learn what Meta caused rather than what it touched, run a holdout test. Our guide to whether your Meta ads are driving sales or taking credit for them shows how to run one without an enterprise budget.

How do the leading Meta ad tracking setups compare?

There are four common setups: the Meta pixel alone, the pixel plus the Conversions API, server-side tagging that feeds every platform from one event, and third-party attribution tools. The first three differ in how many real purchases reach Meta. The fourth adds a second, cross-channel opinion. None makes Meta match your store exactly.

• Pixel only: quick to set up, but it sees only what the browser allows, so it loses the most to Safari, ad blockers and consent choices.
• Pixel plus Conversions API: the purchase goes from the browser and from your server with one shared event ID, so Meta counts it once and still receives it when the browser drops it. On Shopify, the Facebook and Instagram app sends it when data sharing is set to Maximum.
• Server-side tagging: one server container, for example on Stape, receives each event once and forwards it to Meta, Google and TikTok, so all three work from the same purchase.
• Third-party attribution tools such as Triple Whale or Northbeam: their own pixel and model give one view across channels. Useful for comparing channels, but it is another model, and only as good as the purchase event underneath.

Whichever you use, compare each platform with your store's orders for the same dates and time zone, never with each other. For stores, it is part of how we approach marketing for online shops.

How do you measure Meta catalog ads after iOS 14?

Measure catalog ads by the extra sales they bring in, checked against your own store data. Send purchases through the Conversions API with the product IDs from your catalog, compare the campaign's reported revenue with your store's revenue for the same products and dates, and run a holdout test to see how many of those sales would have happened anyway.

Catalog ads often show people products they have already viewed, which is where Meta's credit is most generous: some of those buyers would have come back anyway, and a holdout is the only way to know how many. Our look at Meta's Advantage+ sales campaigns covers how to tell whether that automation is working or only active.

How to close the tracking gap in 5 steps

1. Put last month's Meta purchases next to your store's total orders for the same dates and time zone, and write the percentage gap down.
2. Place a test order and check in Events Manager that the purchase arrives once, with a value, a currency and an event ID.
3. Send the purchase from your server through the Conversions API as well as from the browser, with the same event ID on both, so Meta can deduplicate.
4. Configure your consent banner so tags wait for consent and Google receives Consent Mode signals, instead of tags firing before consent or failing silently.
5. Pick one source of truth, your store or your accounting, judge total spend on blended return, and use each platform's figures only to steer inside it.

The expensive fault is usually a purchase event that fires twice, or on every reload of the thank-you page: the platform then optimises towards the wrong signal, and nothing on the dashboard looks wrong.

Proof of Pixel's position: fix the measurement before you touch the budget, starting with how far your Meta purchases sit from your store's orders.

Proof of Pixel is a marketing and consulting agency for businesses that live on enquiries and orders, building the website, the tracking and the follow-up as one system, with offices in Dubai, New York, London, Antwerp and Kuala Lumpur. Our tracking and lifecycle email work starts with exactly this comparison.

Start with one number

Before you change a budget, put last month's Meta purchases next to your store's orders for the same dates and write down the percentage gap. If Meta claims more purchases than you had orders in total, look for double counting first. That one number tells you whether you have a normal gap or a broken event. If you want a second pair of eyes on it, talk to us.