CAPI, server-side tracking, and dashboarding for Asian automotive manufacturer

The Problem.

The organization faced growing challenges in its digital marketing and analytics infrastructure. With the deprecation of third-party cookies and increasing limitations of browser-based tracking, the company experienced significant data inaccuracies that compromised its ability to measure campaign effectiveness and make informed decisions. 

 

Specifically, they lacked Conversions API (CAPI) integrations for key advertising platforms including Snapchat, LinkedIn, and Reddit -limiting their ability to capture conversion events reliably and optimize ad performance. Compounding this issue was a fragmented analytics setup that relied heavily on client-side data collection, resulting in frequent data loss and misattribution.

 

In parallel, the company’s existing dashboards, which supported multiple lines of business, began showing inconsistent and erroneous data. These errors led to mistrust in weekly reports, slowing down decision-making and impacting stakeholder confidence. A robust server-side analytics solution and a complete dashboard revamp were urgently needed to ensure data accuracy, platform integration, and reliable business insights across the organization.

The Solution.

To resolve the organization’s data accuracy and reporting challenges, a full server-side analytics setup was implemented, along with CAPI integrations for key ad platforms and a complete dashboard revamp.

 

The Conversions API for Snapchat, LinkedIn, and Reddit was deployed using Google Tag Manager Server-Side. Tags were developed to send structured event data directly to each platform’s API, ensuring accurate conversion tracking. Events were authenticated and validated using platform-specific tools to ensure high match rates and minimal data loss.

 

To address browser-based tracking issues, all analytics and marketing tags were routed through a GTM Server container hosted on Google App Engine. A custom tracking subdomain was configured to maintain first-party data collection, and all event forwarding logic was handled in the server container. Real-time monitoring and debugging were done using Google Cloud Logging and GTM’s preview tools to confirm consistent event delivery across platforms.

 

Finally, dashboards were rebuilt using Looker Studio and GA4 BigQuery exports. GA4 was enhanced with custom dimensions to capture geographic “nameplate” data, making it possible to view ad performance by province and vehicle type. The new dashboards enabled more strategic ad planning while also including built-in checks to flag inconsistencies, restoring stakeholder confidence in weekly reporting.