Meta shows fewer conversions than Shopify, and it makes you doubt one of the two. Shopify says 120 orders last week. Ads Manager says 74 purchases. Which one is wrong?
Usually neither. This Shopify vs Meta Ads Manager mismatch is normal to some degree, because the two systems count different things, over different windows, using different rules. Part of the gap is expected and part is genuine Meta ads underreporting caused by events that never arrived. This guide is a reconciliation process: split the gap into those two parts, then work on the part you can fix. The fix for the second part is usually server-side tracking. The numbers in this post are made up for illustration only.
The two numbers answer different questions
| Measure | Shopify | Meta Ads Manager |
|---|---|---|
| What it counts | Every order, from every channel | Purchases that Meta attributes to your ads |
| Credit rule | Shopify's own reporting rules | The attribution setting on each ad set |
| Data source | The order in the database | Pixel and server events Meta received |
| Date logic | Date the order was placed | Depends on the report view and attribution window |
So Meta is expected to show fewer conversions than Shopify whenever some of your orders came from email, organic search, direct visits or other ad platforms. That is not underreporting. It is a different question.
Step 1: split the gap into two parts
Do this before changing anything. In Meta Events Manager, find how many Purchase events Meta received from your pixel and server for the period, regardless of whether they were attributed to an ad. Compare that number with your Shopify orders for the same period.
- Received is close to orders. Tracking is mostly fine. The gap is attribution: windows, view-through and channel mix. Go to steps 2 to 5.
- Received is far below orders. Events are missing. That is real underreporting. Check steps 6 and 7, then read the server-side section.
Conversion reconciliation, step by step
2. Match the period, time zone and currency
Set both reports to the same dates and the same time zone. A shop on one time zone and an ad account on another will shift late-evening orders into different days. Compare orders and attributed purchases in the same currency.
3. Match the attribution setting
Meta applies an attribution setting at the ad-set level. For new ad sets the default has been 7-day click and 1-day view, but you can change it, and different ad sets can differ. Add the attribution setting column in Ads Manager and check it. Meta describes this in its help article on multiple attribution settings. Confirm the current defaults in your own account.
A shorter click window means fewer attributed purchases. A longer window means more, because a purchase a few days after a click is still credited. If you compare a 1-day click number in Meta with total Shopify orders for a store with long consideration times, the gap will be wide for reasons that have nothing to do with tracking.
4. Account for view-through
A view-through conversion is a purchase credited to an ad that the person saw but did not click. It can push Meta's number up relative to what a click-based report would show. If you remove view-through from the setting, the Meta total usually falls. Shopify, which does not track ad views, never sees it. So view-through can make a gap look smaller or larger depending on which direction you look at.
5. Check for duplicates
If Meta shows more than Shopify, look for double counting. This happens when a browser pixel and a server source both send the purchase without a matching ID. Meta deduplicates on event_id and event name, as described in its deduplication documentation. Events that share a key within 48 hours are discarded as duplicates, and the same document notes the fallback with fbp or external_id has limits. Check the event sources in Events Manager if the counts look inflated.
6. Check consent
Visitors who decline marketing consent should not be sent to Meta. Those orders appear in Shopify and not in Meta. That is a correct gap, and it varies with your banner and audience. Do not try to close it. See Consent Mode v2 for the consent side.
7. Check delivery and timing
Server events are queued and sent shortly after the order, so a purchase can appear in Meta a little later than it appears in Shopify. Do not compare to the minute. Look at completed days. If a destination returns errors, SyncBeacon retries up to 24 attempts, and failures show in the dashboard.
Causes at a glance
| Cause | Effect on Meta vs Shopify | Fixable? |
|---|---|---|
| Orders from non-Meta channels | Meta lower | No, expected |
| Short click window | Meta lower | Your choice of setting |
| View-through credit | Meta higher | Your choice of setting |
| Ad blockers, closed tabs, thank-you page not reached | Meta lower | Yes, with server-side events |
| Declined consent | Meta lower | No, by design |
| Browser and server both sending, no shared ID | Meta higher | Yes, match the event ID |
| Event with no usable identifier | Meta lower | Partly, depends on data available |
Where server-side tracking fits
Server-side tracking addresses one row of that table, and a big one: the purchases the browser never reported. With SyncBeacon on Shopify, purchases come from Shopify order webhooks, whichever payment method the shopper used. They carry the Shopify order ID as the purchase event_id, and SyncBeacon merges its browser and server events for the same order. If you also run a separate browser pixel for the same purchase, that pixel has to reuse the same ID, or Meta may count both.
- Meta needs at least one strong identifier (email, phone,
fbcorfbp). An event without one is dropped, not sent. - If marketing consent is explicitly denied, the event is stored but not forwarded. Without a consent signal, your SyncBeacon consent setting decides whether identifiers are used.
- Cancellations are recorded as
OrderCancelledevents. We do not promise refund-adjusted reporting in Meta, so a refunded order can still count as a purchase there.
Setup details are on the Meta Conversions API and Shopify pages. If your tracking is broken rather than just different, start with the Shopify conversion tracking diagnostic checklist.
When every platform claims the same order
Each ad platform credits itself using its own rules, so the sum of platform-reported conversions can exceed your real orders. To compare channels on one basis, you need a single record built from your own orders. SyncBeacon's Beacon Attribution Engine builds a journey ledger from browser touches (UTMs and click IDs) and server-side orders, with first touch, last touch, linear, time-decay and position-based models. It is included on Growth and Enterprise plans. Read the attribution page for what it shows. It is not a way to make the totals match exactly.
Set the right expectation
A perfect match between Shopify and Meta does not exist, and chasing one wastes time. A useful goal is a gap you can explain line by line: this much from other channels, this much from the attribution setting, this much from declined consent, and a small unexplained remainder that you monitor over time. When the remainder grows, something broke.
Related reading
- Meta Conversions API for Shopify: browser pixel vs server-side
- Ad blockers, conversion tracking and ROAS
- Safari ITP and ecommerce tracking
- Ecommerce conversion tracking overview
Get started
Close your conversion gap today
Connect Meta CAPI, Google Enhanced Conversions, and more - in one setup, no developer required.
Attribution defaults and menu names change over time. Check Meta's current documentation for your account. Example numbers are illustrative only. Not legal advice.