Your ad platform and your analytics will never agree. Here is why.

You open Google Ads and it reports 40 conversions. You open your analytics or attribution tool and it says 28. Neither is broken. This gap is one of the most common sources of confusion in paid advertising, and once you understand the five reasons behind it, you stop trying to make the numbers match and start knowing which one to trust for which decision.

1. They count sales on different days

Google Ads and Meta report a conversion on the date of the click, not the date of the sale. If someone clicks an ad on Monday and buys on Thursday, the ad platform backdates that sale to Monday. Your analytics tool almost always reports it on Thursday, when it actually happened. Compare a single day and the totals will never line up, because the two tools are filing the same sale under different dates.

2. They use different attribution models

Ad platforms use their own attribution, often data-driven, which spreads fractional credit across several touches. A last-click tool gives 100% of the credit to one touch. So a sale that Google records as 0.4 of a conversion (because the customer also touched two other campaigns) shows up as a whole conversion, or none, in a last-click tool depending on which touch it credits. Different models, different totals, same sale.

3. “All conversions” is not what you think

Many advertisers compare their tool against Google’s All conversions column, which can include secondary actions like page views, add-to-carts, and multiple conversion actions firing for one purchase. If you have three conversion actions set up and sum them, one sale can count as three. Compare against your single primary purchase action instead.

4. Platforms model what they cannot see

When privacy settings or cookie loss stop a platform from observing a conversion directly, it does not always drop it. It models an estimate based on similar converters and adds it to the total. Those modeled conversions are real revenue in aggregate, but they are estimates, and your analytics tool, which only counts observed sales, will not include them.

5. Spend, currency, and timezone drift

Reported spend updates on a delay, currencies convert at different rates, and tools use different timezone boundaries for “today”. None of these is large on its own, but together they add noise to any direct comparison.

So which number do you trust?

Stop trying to reconcile them to the dollar; they are answering different questions. Use the ad platform’s number to feed bidding, because that is the number its algorithm optimises against, and it is only as good as the conversions you send it. Use a complete, first-party attribution tool to decide where to put budget, because it sees the full journey the platforms cannot.

The real problem is not that the numbers disagree. It is that the ad platform’s view is usually incomplete, missing the conversions the browser pixel dropped, so it optimises on a fraction of your revenue. Trakt recovers those missed conversions and reports them back, which narrows the gap for the right reason: the platform finally sees more of the truth.