Book a call
← All case studies
Case study · Tourism · Luxury hotels · GA4 · Consent Mode · Google Ads

From 40–50 % to around 10 %:
how a luxury resort found its bookings in GA4 again.

A resort in the luxury segment sells its stays through an external booking engine on its own domain. The booking system showed far more website bookings than GA4, even though Enhanced Conversions and cross-domain tracking were already in place. We narrowed down the causes step by step, repaired the tracking, connected the new affiliate programme at Awin and rebuilt the reporting. Today the gap between the booking system and GA4 is around 10 %. We are now optimising Google Ads on this data.

40–50 % → 10 %gap between the booking system and GA4, before and today
2 domainswebsite and booking engine measured as one customer journey again
3 scenariosconsent given, rejected and no choice made, each tested separately
Brand · non-brandpaid and organic cleanly separated in the dashboard
Starting point

Good campaigns, unclear figures.

We started with a simple request: the dashboard should clearly separate paid brand traffic from organic brand traffic. The real question soon followed. A large share of the bookings in the booking system did not show up in GA4 at all, and many that did were attributed to the booking engine as a referral instead of the campaign that had brought the guest.

Together with the client's Google team, Enhanced Conversions and cross-domain tracking had already been implemented. The gap remained nevertheless. We systematically checked the usual suspects: consent banner, browser tracking protection, the purchase event on all booking and payment paths, and how the domains were linked.

Google Tag ManagerLinker, conversion linker and purchase event checked and adjusted.
GA4Domains, DebugView and consent signals for each scenario.
Booking system & Data StudioBookings from the booking engine as a baseline, new reporting.
What we found and fixed

Four causes, four workstreams.

01 · Cross-domain

The guest became a new user along the way.

When moving from the website to the booking engine, the cookie with the GA4 client ID changed. GA4 could no longer connect the visit and the booking. We allowed all domains involved in GA4 and extended the linker in the tag manager, including a technical subdomain of the booking engine.

02 · Consent Mode v2

No consent, no continuous journey.

We tested all three cases. With no choice made, nothing is sent; with consent, the link across both domains works. If consent is rejected, GA4 sends cookieless signals without a client ID, as intended. Some of these bookings therefore end up as referrals from the booking engine. How large this share is depends on the consent rate in the consent tool.

03 · Purchase event

A test booking instead of guesswork.

With a real test booking, we checked the purchase event and Enhanced Conversions in the debugger and in GA4 DebugView. All parameters are correct for this booking path. We compare other payment paths with the booking system data so that no route is lost unnoticed.

04 · Affiliate

A new partner programme at Awin, cleanly connected.

For the launch in the Awin network, neither the master tag nor tracking was on the website. We implemented the mandatory components via the tag manager: master tag on all pages, conversion tag with fallback pixel, server-to-server tracking at booking completion and consent handover via the Awin framework. Awin receives the net booking value, checked with test bookings via the tracking link.

Reporting

A dashboard that answers questions.

Before we built anything, there was a proposal with sample data that the resort's team could comment on directly in the document. Only once the structure was agreed did we rebuild the Data Studio dashboard.

OverviewThe key figures, long-term trends and the main breakdowns on one page.
One page per channelGoogle Ads and organic search in detail, organic complemented by Search Console data.
Brand vs. non-brandPaid and organic brand traffic separated, so Google Ads does not pay for guests who would come anyway.

We hid the previous dashboard pages instead of deleting them, so a comparison remains possible at any time.

Collaboration

International, and without standstill.

The resort's team works in English and is spread across several countries, plus an external IT service provider and the Google team. We therefore ran the project with clear written updates: what is done, which question is open and who can answer it.

One dedicated contact from our analytics team led the project. During holidays, a colleague who had been briefed took over seamlessly. Every change in the tag manager was approved beforehand.

What comes next

Measure first, then optimise.

StepStatusWhy
Check and repair cross-domain, consent and purchase eventDoneWithout a connected journey, GA4 attributes bookings to the wrong source.
Connect the Awin partner programmeDonePublishers are only paid correctly if every partner booking arrives with its net value and consent.
Rebuild the dashboard, separate brand and non-brandDoneDecisions need a page that everyone understands.
Optimise Google Ads based on the new dataOngoingBudget where new guests book, not where they would come anyway.

A small remaining gap is part of every privacy-compliant setup: anyone who does not consent to measurement is only recorded cookieless. For reasons of confidentiality, we do not name the resort and do not show revenue.

Similar problem?

Your booking system and GA4 show different figures?

Tell us in 30 minutes where the gap is, or send a short message. We will tell you honestly whether and how we can help.

Armin Sanjari
Armin SanjariFounder & CEO · Your contact

Prefer to call? +49 711 500 908-17

Book a 30-minute call →
or send a short message

Armin replies personallyWithin 24 hours on working days · no obligation · no sales pressure

Request a free tracking check →

Book a 30-minute call

Open at Calendly ↗ · Send a message instead · Privacy settings