What happened with Slic
In October 2024, SAR 6,216 in ads brought Slic 44 orders and SAR 12,000 in sales on Zid. GA4 recorded zero checkouts against those 44 real orders. Judged by GA4, the ads that were selling would have been switched off.
So for Ramadan and Eid 2025 I moved reporting to Zid, the store's order system, and used the platform numbers only to steer the budget between Meta, TikTok and Snapchat.
- 216
- Orders recorded on Zid
- 129
- Purchases counted by the platforms
- SAR 50,518
- Sales on Zid, 1 Mar to 20 Apr 2025
Why the numbers split
Three things sit between an ad click and the order system, and each one can lose the purchase:
- The checkout leaves your domain. On Zid, checkout starts with a login step on
sso.zid.store, then returns to the store. A store on its own domain that does not set GA4 and the pixels up for that hop can lose the visit, and the purchase that follows it. - Browsers block part of the tracking. Privacy settings and ad blockers stop some pixel events from firing at all.
- Each platform counts by its own rules. Meta, TikTok and Snapchat each credit a purchase only when they can tie it to their own ad inside their own attribution window, and none of them sees the others.
The gap can go either way. Platforms can count more than the store, when two of them claim the same order, or fewer, as they did for Slic. In both cases the store is the only system that records every order once.
The rule I work by
The store's number is the result. The platform's number is a compass.
- Report sales, orders and cost per order from Zid, Salla or Shopify.
- Inside one platform, compare campaigns and ads with that platform's purchases, because the counting rules are the same on both sides of the comparison.
- Between platforms, read cost per reported purchase as a direction, then confirm every budget move against the store's total.
How the Slic budget moved
| Platform | Spend (SAR) | Reported purchases | Cost per reported purchase (SAR) |
|---|---|---|---|
| Meta | 17,465 | 90 | about 194 |
| TikTok | 8,281 | 21 | about 394 |
| Snapchat | 2,568 | 18 | about 143 |
Snapchat started late, took 9% of the budget and showed the lowest cost per reported purchase. It deserved a longer and larger test. The season closed at SAR 50,518 from 216 orders and 205 customers, on SAR 28,314 of spend including VAT: about 1.8x on spend, a seasonal result reported as it was. Read the full Slic case.
The check to run before the season
Before the season budget goes up, run this on your store:
- Place a real test order from an ad link and follow it: does it appear in Zid, in GA4 and in each platform's events manager?
- Watch the address bar during checkout. If it leaves your domain at any step, add that domain to cross-domain measurement in GA4.
- Compare one full week: orders in the store against purchases in each platform. Write the gap down; it is how you read platform numbers for the rest of the season.
- Check that purchase value and currency match on every platform: SAR, and the same treatment of VAT.
- Decide before launch which number goes in the report. For Slic it was Zid.