Why Meta Ads sales don't match your store
Meta Ads sales almost never match store orders. Attribution windows, report dates, delayed payments, and how to compare both without fooling yourself.

Meta Ads Manager sales almost never match the orders in your store. That gap is normal. Your store counts what entered (or is still pending in) your cash flow. Meta counts conversions it can attribute to an ad under its own rules.
If Ads Manager ROAS and the day's revenue in Shopify, WooCommerce, or another checkout disagree, start by checking what each side is counting.
Do Meta and the store answer different questions?
Yes. The store answers: "how many orders happened in this period, and with which status?". Ads Manager answers: "how many purchases can I credit to these ads, given this ad set's attribution setting?".
Meta itself explains that Ads Reporting conversion counts can differ from third-party tools because tools detect, define, and calculate conversions differently. See About Conversion Count Differences Between Meta Ads Reporting and Third-Party Reporting Tools.
Your store includes organic sales, bio links, WhatsApp, email, affiliates, and manual orders. Meta only counts what fits the ad set's attribution rules. The store updates status (paid, canceled, refunded). An ads report may keep showing a conversion that no longer counts in cash.
Adding Meta + Google + another channel often exceeds paid orders: each platform only sees its own ads and may claim the same purchase.
What is an attribution window in Meta?
Attribution is the rule that decides whether a purchase "belongs" to an ad. The attribution window is how long that link still counts.
In Meta's Help Center, About Attribution Models and Attribution Settings lists the standard settings supported for website conversions:
- Click-through: events within 1 day or 7 days after a link click on your ad.
- View-through: events within 1 day after an impression (the person saw the ad without clicking).
- Engage-through: events within 1 day after a non-link-click interaction.
In How Conversions Are Attributed in Commerce Manager, Meta states that the default attribution setting for new ad sets in Ads Manager became 7-day click or 1-day view, with a last-touch model: credit goes to the last ad clicked or interacted with; without a click, it can go to the last ad viewed.
Two effects follow:
- Meta can count more than your store would treat as a sale from that ad, for example with view-through credit.
- Meta can count less if the purchase fell outside the window, or if the pixel/CAPI never recorded the event.
Why doesn't the panel's day match the store's day?
The day a sale appears can differ.
On the same page, Meta explains that Ads Manager reports based on impression date (when the ad was shown), not necessarily the payment day. In its example, a Monday click is reported against Monday's impression. In several Commerce Manager reports, the date used is the conversion date.
Your store usually shows the order on the day it was created or paid. A common path: click on Monday, delayed payment on Tuesday, confirmation on Wednesday. "Yesterday" in Ads Manager and "yesterday" in the store are rarely the same slice. In Brazil, Pix and boleto make this gap especially visible.
In Troubleshoot Common Third-Party Reporting Differences, Meta recommends aligning the attribution window, time zone, and whether you report by conversion date or exposure date.
What else makes the numbers diverge?
Created orders versus paid orders
Delayed payment methods create an order before money clears. If the store shows "orders" and the panel shows Purchases, you may be comparing different things. Separate created, awaiting payment, and paid.
Cancellations and refunds
The store discounts or changes status. The ads report may keep the attributed conversion, depending on what the pixel/CAPI sent and when.
Several platforms claiming the same sale
Someone saw Meta, clicked Google, then bought. Each panel can claim it. Summing platform ROAS is not business ROAS.
Duplicate or missing events
On the third-party differences page, Meta notes that ad blockers can stop the Pixel from firing, and that different metric definitions also change totals. The troubleshooting page suggests checking raw fires in Events Manager, currency and decimals on the purchase event, and possible duplicate fires.
Time zones
Ad account in one zone, store in another. An order at 11 pm becomes "another day" on one side.
You can usually name where each piece of the gap comes from. You rarely need a mysterious bug story.
How do you compare Meta and the store without fooling yourself?
Use a short, repeatable method. Aim to explain most of the gap rather than force it to zero.
- Pick a closed week, not "yesterday". Recent days still move in Ads Manager as older conversions arrive.
- In the store, use paid orders only (or the net report you use for cash). Note canceled and refunded orders separately.
- In Meta, note the ad set attribution setting (for example 7-day click or 1-day view) and the account time zone.
- Compare Events Manager with the store before Ads Manager. Events Manager shows received events; Ads Manager shows events attributed to ads. If Events Manager is already far from the store, the problem starts in event tracking. If Events Manager is close and Ads Manager is far, the gap is attribution.
- Separate what the store sold without ads: organic, email, messaging, affiliates. That volume should never match Ads Manager.
- Do not add Meta + Google to match total orders. Each can count the same purchase.
- Split decisions: for cash, paid orders ÷ ad spend; inside Meta, which ad set performs better under the same attribution rule.
Use Ads Manager to compare ads with each other. Use the store, with a clear paid-order definition, to decide whether the business can afford the spend.
Do you need to replace the pixel when the numbers disagree?
Confirm first whether the gap is attribution, order status, date, or a broken event.
Swapping pixel, CAPI, or tools mid-comparison only adds another variable. If Events Manager already receives Purchases near paid orders, basic tracking works; the Ads Manager gap is largely the credit rule. Keep the pixel if it feeds Meta's optimization: the platform report and your cash count can coexist.
When does it make sense to read orders another way?
When the question is "which ad brought this paid order?", rather than only "how many conversions did Meta attribute to this ad set?".
You need three linked pieces: the click or visit with identifiable campaign/ad; the order from the store or checkout, with status; and your own attribution rule applied to real orders.
Independent attribution tools exist for that. Clakta is one of them: it connects visits and clicks to orders your store confirms and shows credit under the model you choose. Revenue comes from the same store orders, with status; a pending delayed payment does not count as revenue yet; refunds and cancellations update the same order. Your Meta pixel stays in place: Clakta runs beside it and does not change the platform report (What is Clakta, How attribution works). For stores, see also the ecommerce page.
Meta and the store remain different numbers. What changes is having real orders per ad next to what Meta shows.
Frequently asked questions
Is it normal for Meta to show more sales than the store?
Yes, it can be. View-through credit, last-touch inside the window, conversions not yet refunded in the panel, and sometimes duplicate events inflate Ads Manager. Meta can also show less if some purchases fell outside the window or the event never arrived.
Which number should I use for budget decisions?
Use two. For business health: paid store orders and total media spend. To optimize inside Meta: Ads Manager metrics, always with the same attribution setting across the ad sets you compare.
Do delayed payments mess up the comparison?
Yes, if you mix created orders with paid orders. Separate statuses before comparing with Purchase in Meta.
Can I make Meta and the store match exactly?
Meta and the store measure different things, so an exact match is the wrong target. A useful goal is explaining most of the gap.
Do UTMs fix this on their own?
No. UTMs label the visit. You still need the order and a way to connect the two. UTMs help reconciliation; they do not replace attribution or payment status. For labeling Meta campaigns, see the Meta Ads UTM parameters guide.
Should I pause a campaign just because panel ROAS dropped?
Check the period, conversion maturation, order status, and what the store sold in the same window first. Pause based on spend versus paid orders you trust, not on one isolated Ads Manager day.