The ad bill is real. The sales data is not.
Here is where tracking usually breaks.
The bill from Meta or Google arrives every month. The conversions column next to it shows almost nothing, while the bank account tells a different story. That gap is almost never the ads. It is tracking.
Why the numbers almost never add up by default
An ad platform only knows about a sale if something tells it a sale happened. By default, that something is a pixel sitting in a browser, firing a conversion event on page load. Browser privacy changes, from iOS settings to ad blockers to cookie consent, have made that single signal miss a growing share of real conversions. The sale still happened. The ad platform just never heard about it.
This is the single most common reason a business sees real revenue with almost no matching conversions. The ads did not fail. Nobody set up a second way for the platform to hear about a sale the pixel missed.
What server-side tracking actually fixes
Conversions API for Meta, and server-side tagging for Google, send the same conversion event from the business’s own server, not only the visitor’s browser. It does not replace the pixel. It backs it up, catching sales that ad blockers or privacy settings would otherwise hide from the ad platform. Setting this up correctly, matched to the real checkout flow, is usually the single highest-value fix here.
Checking it is actually working, not just installed
A pixel installed once and never checked again is one of the most common things an audit finds. Both major platforms have a live test tool built for exactly this: place a real test action and watch whether the event shows up within minutes. This takes an hour, not a project, and it tells you immediately whether the problem is tracking or something else.
When the gap is somewhere else
Not every mismatch is a pixel problem. Sometimes the ad platform tracks correctly, but a CRM does not record where a lead actually came from. The business cannot see which ads produced which sales, even though the ad platform itself is fine. Checking both ends is the fastest way to tell which one broke. Put the platform’s conversion count next to your own order data, for the same dates.
Fixing it before scaling spend
Finding a tracking gap before raising ad spend is almost always cheaper than scaling into the same blind spot and hoping the numbers catch up. Once tracking is confirmed with real test events, and matched against actual orders, decisions about which campaigns to scale or cut rest on what is actually happening. Not on a platform that has been missing a real share of the sales the whole time.
What to do in the next hour
- 01
Place a real test order and watch whether a conversion event fires in the ad platform's own test tool within minutes.
- 02
Check whether server-side tracking (Conversions API for Meta, server-side tagging for Google) is set up at all.
- 03
Compare the number of orders in your store or CRM against the conversions shown for the same date range.
- 04
Do not raise ad spend to "test if it's working" before confirming the gap. That just spends more into the same blind spot.
When to call us
Call us if ad spend is a meaningful cost and the tracking gap has been there for weeks. Also call if you are about to scale spend and want tracking confirmed first. Get a free plan →
FAQ
Is this an iOS or privacy problem we just cannot fix?
Privacy changes made browser-only tracking less reliable, that part is real. It is also why server-side tracking exists. A properly set up server-side connection recovers a large share of the events a browser pixel alone would miss. Most of the gap is fixable.
How do we know if the pixel is even working?
Both Meta and Google have a built-in test tool that shows events firing in real time. We check this directly, rather than assuming from the reported numbers alone.
Could this be a CRM problem instead of a tracking problem?
Sometimes. Orders can be tracked correctly while a CRM pipeline does not log where the lead came from. Then the ad platform looks broken when the real gap sits further down the funnel. We check both ends first.
What does a proper fix actually include?
A working pixel and server-side connection, test events confirmed live, and a check that the numbers match between the ad platform, GA4 and your own order data. Plus a short written note on what was wrong.