Skip to content

A cleaner record of every booking.

Activate’s bookings start on its website and finish in ROLLER checkout. Tracking had grown across native integrations, custom scripts and Google Tag Manager, so one booking could be recorded twice and credited to Activate’s own site. We gave tracking one home and repaired the hand-off to checkout.

Activate GamesEntertainment and attractions · Tracking repair

GTM preview · booking BK-58214playactivate.com → ROLLER checkout

  1. page_view · Mega GridFrom a Meta ad · Spring Weekend
  2. begin_checkout · ROLLERMega Grid · 4 players · Sat 15:00
  3. purchase · $96.00Sent once · source: Meta, Spring WeekendTags fired

One booking became two purchases, credited to Activate’s own site.

Bookings started on Activate’s website and finished in ROLLER. Tags in three places could record one purchase twice and credit Activate’s own site instead of the channel that sent the visitor, while a Reddit tag sent missing or zero revenue.

What happened

Website1 booking

What the reports recorded

2 purchasesSelf-referral$0 in Reddit

One booking event, one home for tags, one purchase sent to each platform.

Booking tags now live in Google Tag Manager, and a repaired bridge passes each ROLLER purchase to analytics and ad platforms once.

Activate Games’ environment

Website + checkout

Booking event bridge

  • Checks message origin
  • One listener, no repeats
  • Clears stale cart data

Google Tag Manager

  1. PurchaseOnce per booking
  2. ValueMapped to live data
  3. ProductsOne shared structure
Overlapping integrations removedLegacy tags pausedDocumented setup

One purchase per booking

  • GA4Purchase, value and channel
  • MetaMoved to a template
  • TikTokShared product data
  • RedditPurchase value mapped

How Activate’s booking data changed, step by step.

Four fixes, from giving tracking one home to sending the right purchase details to each platform.

  1. 01

    Give tracking one home

    We consolidated tracking in Google Tag Manager, removed overlapping native integrations, paused legacy tags and handed Activate a documented configuration.

    Before

    Tags spread across native integrations, custom scripts and GTM

    After

    One documented place to manage how booking events reach each platform

  2. 02

    Repair the hand-off to ROLLER

    The event bridge now checks where messages come from, prevents duplicate listeners and clears old ecommerce data before the next event.

    Before

    The same purchase could be sent more than once, or carry stale values

    After

    Duplicate tracking resolved

  3. 03

    Credit the channel, not the site

    At the February 26 handoff we confirmed the self-referral issue was resolved, and the team reviewed the effects in Google Analytics and Facebook.

    Before

    Bookings credited to Activate’s own website

    After

    Self-referrals resolved, checked in GA and Facebook

  4. 04

    Align purchase details

    We moved Meta from legacy custom code to a template, corrected Reddit’s purchase-value mapping and aligned TikTok’s product data with the shared ecommerce structure.

    Before

    Reddit purchases with missing or zero revenue

    After

    Purchase values and product data mapped correctly

One booking in ROLLER. One purchase event sent to each platform.

An example check of the purchase events each platform received against bookings in ROLLER. Consent choices and ad blockers leave small gaps, and ad platforms report only the purchases they’re credited with.

Purchase event checkROLLER bookings vs platforms
All locations ▾Last 30 daysEvent: purchase
Show
Bookings3,412in ROLLER
GA4 purchases3,30997% of ROLLER
Duplicate purchases0per booking ID
Self-referrals0credited to own site

Purchase events received

PlatformPurchasesOf ROLLER
ROLLER bookings3,412100%
GA43,30997%
Meta3,10491%
TikTok3,03789%
Reddit2,96987%

GA4 purchases by channel

MetaPaid social1,01531%
Organic searchUnpaid84826%
DirectUnpaid54516%
TikTokPaid social40512%
EmailOwned2678%
RedditPaid social2297%
Own websiteSelf-referral00%
Illustrative data

Bookings don’t add up across your platforms? We’ll trace one booking from ad to checkout and show you where the count splits.

Book a walkthrough