Why your Meta ads report wrong numbers — and how the Pixel and Conversions API fix it

The short answer
Meta’s numbers differ from your store’s for four main reasons: the browser Pixel loses events (ad blockers, privacy settings, slow pages), purchases get counted twice, Meta credits sales up to 7 days after a click, and in Egypt many cash-on-delivery orders are later refused. The fix is the Pixel plus the Conversions API with one shared event ID, a purchase sent once with the right value, and judging ads by delivered orders.
Your store says 120 orders this week. Meta says 160 purchases — or 70. Which one is right? Usually neither tells the whole story, and if your ads learn from wrong numbers, they spend your money on the wrong people. Here is why the numbers drift apart and how to make them trustworthy.
How Meta knows someone bought
There are two ways to tell Meta about a purchase:
- The Meta Pixel — a small script in the visitor’s browser that reports page views, add-to-carts and purchases.
- The Conversions API — your website’s server reports the same events directly to Meta, without depending on the browser.
Used together, with the same event ID, they cover each other’s gaps and Meta counts each event once.
Reason 1: the browser loses events
Ad blockers, privacy settings on phones and browsers, and pages that close before the Pixel loads all stop events from reaching Meta. The result is fewer purchases in Meta than in your store — and ads that cannot find more buyers like the ones they missed.
Reason 2: the same purchase is counted twice
If the thank-you page sends a purchase every time it opens — a refresh, a back button, an order-tracking link — one order becomes two or three. The same happens when the Pixel and the server both send a purchase without a shared event ID. The result is more purchases in Meta than in your store.
Reason 3: Meta credits sales differently
By default, Meta counts a purchase for an ad if it happens within 7 days of a click on the ad, or within a day of someone seeing or engaging with it — even if they finally bought through Google or came back directly. Meta also reports the sale against the day of the ad, while your store counts the day of the order. Both are “right”; they answer different questions.
Reason 4: cash-on-delivery orders that never get paid
In Egypt, many orders are cash on delivery, and some are refused at the door. Meta counted them as purchases the moment they were placed. If you judge your ads only by Meta’s number, you will think they earn more than they do.
The checklist to fix it
- Install the Pixel and the Conversions API together, sending the same event ID for each event.
- Send the purchase once — only the first time the order is confirmed, never on a refresh.
- Send the right value and currency — the amount the customer pays, in Egyptian pounds.
- Send customer details safely — Meta accepts the e-mail and phone number scrambled (hashed) so it can match buyers better; say so in your privacy policy.
- Check Meta’s Events Manager — use “Test events” to see each event arrive, and read the diagnostics for duplicates and missing details.
- Compare every week: Meta purchases, store orders and delivered orders side by side.
- Judge ads by delivered revenue, not by orders placed — that is your real return.
What right numbers change
In one account we took over, an e-commerce brand was getting back 63 piastres for every pound spent on Meta ads. The work started with an audit of the account and its tracking, then a new campaign structure, new creative angles and fixes on the landing pages and checkout. A month later the return was 3.55 pounds per pound, and purchases went from 83 to 465 — see the account turnaround case. Tracking alone did not do that, but nothing else works on numbers you cannot trust.
Want your numbers checked?
We set up the Pixel and Conversions API with each event counted once, and run ads judged by real, delivered revenue. See our performance marketing service.
Questions about this
What is the Conversions API?
A direct connection from your website’s server to Meta that reports visits, leads and purchases, so they are not lost when a browser blocks the Pixel.
Do I still need the Pixel if I use the Conversions API?
Yes. Meta recommends both together, sending the same event ID from each so every event is counted once.
Why does Meta show more purchases than my store?
Usually double counting — a thank-you page opened twice, or the Pixel and the server sending without a shared ID — sales credited to an ad someone only saw, or cash-on-delivery orders that were refused later.
Why does Meta show fewer purchases than my store?
Mostly lost browser events: ad blockers, privacy settings and pages that close before the Pixel loads. Some orders also really came from other channels, like Google or a direct visit.
Who wrote this
By the GRWPTH team — GRWPTH is a growth agency: we build websites and online stores, automate the work around them, and grow them with performance marketing, in Arabic and English.
See our work




