When Day Trio, a marketplace for whole-day trips run by local operators, opened in September 2026, the tracking went in before any money went into ads. Travellers on Day Trio book and pay online, so every booking can be traced back to the ad, post or search that brought it. Day Trio is a marketplace we work with. Here's what was set up, and seven lessons from the first weeks that apply to any tour operator or hotel taking bookings on its own website.
The short version: decide what a booking looks like in your analytics before you spend, judge every campaign by paid bookings rather than clicks or "results", and check your ad placements and website settings in the first week, because that's where small budgets leak.
1. Track the whole booking, step by step
A booking isn't one moment. On Day Trio it's four steps, and each one is sent to Google Analytics as its own event:
Viewed a trip (view_item): someone opened a trip page.
Started booking (begin_checkout): they picked a date and opened checkout.
Opened payment (add_payment_info): the payment window opened.
Paid (purchase): sent only after the payment is confirmed, with the amount and currency.
With four steps you can see exactly where people drop off. If lots of people start booking but few open the payment window, the problem is the checkout, not the ads.
2. Don't trust GA4's "Checkout journey" report for trips
GA4's built-in Checkout journey report was designed for online shops. Its second step is "Add delivery", the shipping address, and it only counts someone at a later step if they passed every earlier one. A day trip has no delivery, so the report shows zero purchases even when bookings are coming in. The fix takes ten minutes: build a Funnel exploration (Explore, then Funnel exploration) with your own four steps, and check the purchase event itself under Reports, Engagement, Events.
3. Send the booking to Meta from your server, not just the browser
The Meta Pixel runs in the visitor's browser, and browsers lose events: ad blockers, privacy settings and Instagram's own in-app browser all get in the way. Day Trio sends the purchase to Meta through the Conversions API from its own server, after the payment is confirmed. Two things are worth knowing when you test it:
Server purchases don't appear in Events Manager's "Test events" tab unless they're sent with a test code. Look in Overview instead.
Overview can lag 40 to 60 minutes. Make one small real booking, wait an hour, then check.
4. Tag every link, so each booking knows where it came from
Every Day Trio ad carries the same URL parameters:
utm_source=meta&utm_medium=paid_social&utm_campaign={{adset.name}}&utm_content={{ad.name}}
Meta fills in the ad set and ad names automatically, so each booking records exactly which ad brought it. Tag your Instagram bio and YouTube channel links the same way (utm_source=instagram or utm_source=youtube, with utm_medium=social). Without tags, most of this traffic shows up as "direct" and you can't tell what's working.
5. Check Audience Network in the first week
With Advantage+ placements switched on, Meta can send a large share of a small budget to Audience Network: ads shown inside other apps and games. Those clicks are cheap and can produce plenty of "checkouts started", but very few of those people go on to pay. Open the campaign, choose Breakdown, then Placement, and compare spend against paid bookings. If Audience Network takes the money without bringing bookings, exclude it in the ad set's placements, and untick "Allow limited spending to excluded placements", which Meta ticks for you.
6. Re-check every duplicated ad
Duplicating an ad in Ads Manager is quick, but it can quietly switch Meta's automatic enhancements back on: spotlights, video effects and AI-written summaries of your website. Those summaries can state things your page doesn't promise. After every duplicate, open the ad and check each enhancement before you publish.
7. Make sure Google can see your photos
A robots.txt rule meant to keep Google out of internal search results (any address with "q=" in it) also blocked every photo on the site, because image addresses on many modern websites contain "q=75" for picture quality. Google's free Rich Results Test shows the problem: under More info, then Page resources, blocked images are listed one by one. One extra line in robots.txt fixed it. Run the test on your home page and one trip page; it takes a minute.
A quick checklist before your first campaign
Four booking events in GA4, with the purchase sent only after payment.
A GA4 funnel exploration with your own steps.
Meta Conversions API sending the purchase from your server.
UTM tags on every ad, bio and channel link.
The placement breakdown checked after the first few days.
The Rich Results Test run on your main pages.
For the full setup, read our tracking setup guide for travel businesses or see our analytics and tracking service. If you'd like us to check yours, book a free 30-minute strategy call.