What the ad platforms claimed against what the hotel actually took
Two ad platforms claimed 92.1% of one property’s bookings between them. Neither can see the other, so both counted the same guest.
4 min read
One property. June to August 2026 inclusive. Bookings counted, not valued.
| Figure | Value |
|---|---|
| Bookings claimed by Meta | 33 |
| Bookings claimed by Google Ads | 153 |
| Total claimed by paid advertising | 186 |
| Bookings the website recorded, every source | 202 |
| Share of all bookings claimed by paid | 92.1% |
Run on revenue over the identical window, the same subtraction gives 93.6%, so the two measures agree closely and neither depends on the other.
How it was measured
Claimed bookings are each platform’s own reported conversions for the window, taken from the platforms rather than from us: Meta on a 7-day-click and 1-day-view basis, Google Ads on its own click-based model. Recorded bookings are the reservations the property’s own booking engine wrote in the same window, from every source including direct, organic, repeat, phone and walk-in.
No attribution model is applied on our side and none is needed, because the argument is a comparison of two totals rather than an allocation between them.
Why the ratio is impossible rather than merely high
If paid advertising had caused 92.1% of reservations, then every repeat guest, every direct booking, every branded search, every email and every walk-in together accounted for 7.9% of the property’s business. No hotel operates that way.
The excess is double counting. A guest who saw a Meta advert without clicking it, then searched the hotel by name and clicked a Google advert, is counted once at full value in both systems, and neither system can see the other.
Limits
One property, one three-month window, one pair of platforms. The direction of the effect is structural and will appear in any account running both platforms, because the cause is that neither can observe the other’s claims. The magnitude is specific to this account and its channel mix, and a property with heavier direct business would show a larger gap rather than a smaller one.
The property is not named and no revenue total is published. A ratio and two counts do not identify a business.
Reproduce it on your own account
Count every booking you took in a period, remove the ones no advertisement could plausibly have caused, and compare what remains against the sum of what each platform claims. Anything above your ceiling is being counted twice.
Hostli reads both platforms against the bookings your engine actually wrote, so the overlap shows up as a number rather than an argument.
See how it works