The Server-Side Google Ads conversion is claiming no or too few conversions
Different types of conversions, such as through the server-side tag, GA4 imports, or on-site scripts, use various methods to collect data, which can lead to discrepancies in measurements. It is crucial to determine which conversion is the most accurate and to check if certain conversions might be overestimating. For example, a conversion based on a thank-you page pageview might result in overestimation compared to specific interactions tracked via the DataLayer, which may be more precise.
Although a server-side conversion has its advantages, it does not always guarantee the best results. GA4 import conversions, for example, are often more reliable as they utilize UTM parameters that help identify campaigns. This provides insights into which users arrived on the site via a specific campaign, which is especially valuable when optimizing marketing efforts. The data collected in GA4 also originates from your Server-Side Tagging setup but may be processed and modeled differently by Google Ads.
In addition to the inherent differences between conversion methods and their performance, other issues or settings should be checked to identify if something is genuinely wrong.
Issues with Gbraid and Wbraid Parameters for Apple Visitors
Apple visitors may encounter issues with the Gbraid and Wbraid parameters not being properly forwarded via the standard Google Ads tag in the server container. AdPage has developed a custom server-side tag that works with webhooks to transmit these parameters. For instructions on implementing this tag, refer to AdPage’s guide. If you can determine in GA4 that many of your site visitors use Apple devices, there is a high likelihood that server-side conversions may underperform.
Check the Attribution Model
Attribution models determine how Google Ads assigns credit to different channels in the conversion path. Various options are available:
• Last Click: Attributes the conversion to the last interaction point before the user converted.
• Data-driven: Utilizes machine learning to calculate the contribution of each touchpoint in the conversion path based on historical data.
The data-driven attribution model differs from other attribution models by using your own conversion data to calculate the actual contribution of each ad interaction. This model is specific to each ad account. Data-driven attribution analyzes all interactions on your Google Ads campaigns across Search, YouTube, Display, and Demand Gen. By comparing the paths of customers who convert with those who don’t, the model identifies interaction patterns likely to lead to a conversion. These valuable interactions are then given more credit in the model, helping you see which ads contribute most to your business goals.
To check which attribution model your ad account uses, navigate to ‘Goals’ in your Google Ads account. Open one of your conversions and go to the settings. Under ‘Attribution Model,’ you can choose whether to focus only on the last click or to use the data-driven model.
Consent Modeling
Consent Modeling accounts for users who do not consent to tracking. Google supplements the missing data using models for conversions where tracking information is unavailable. This helps estimate performance more accurately but may cause slight discrepancies, even between different conversion goals within Google Ads itself. Refer to Google’s Consent Modeling documentation for more details.
Although a server-side conversion has its advantages, it does not always guarantee the best results. GA4 import conversions, for example, are often more reliable as they utilize UTM parameters that help identify campaigns. This provides insights into which users arrived on the site via a specific campaign, which is especially valuable when optimizing marketing efforts. The data collected in GA4 also originates from your Server-Side Tagging setup but may be processed and modeled differently by Google Ads.
In addition to the inherent differences between conversion methods and their performance, other issues or settings should be checked to identify if something is genuinely wrong.
Issues with Gbraid and Wbraid Parameters for Apple Visitors
Apple visitors may encounter issues with the Gbraid and Wbraid parameters not being properly forwarded via the standard Google Ads tag in the server container. AdPage has developed a custom server-side tag that works with webhooks to transmit these parameters. For instructions on implementing this tag, refer to AdPage’s guide. If you can determine in GA4 that many of your site visitors use Apple devices, there is a high likelihood that server-side conversions may underperform.
Check the Attribution Model
Attribution models determine how Google Ads assigns credit to different channels in the conversion path. Various options are available:
• Last Click: Attributes the conversion to the last interaction point before the user converted.
• Data-driven: Utilizes machine learning to calculate the contribution of each touchpoint in the conversion path based on historical data.
The data-driven attribution model differs from other attribution models by using your own conversion data to calculate the actual contribution of each ad interaction. This model is specific to each ad account. Data-driven attribution analyzes all interactions on your Google Ads campaigns across Search, YouTube, Display, and Demand Gen. By comparing the paths of customers who convert with those who don’t, the model identifies interaction patterns likely to lead to a conversion. These valuable interactions are then given more credit in the model, helping you see which ads contribute most to your business goals.
To check which attribution model your ad account uses, navigate to ‘Goals’ in your Google Ads account. Open one of your conversions and go to the settings. Under ‘Attribution Model,’ you can choose whether to focus only on the last click or to use the data-driven model.
Consent Modeling
Consent Modeling accounts for users who do not consent to tracking. Google supplements the missing data using models for conversions where tracking information is unavailable. This helps estimate performance more accurately but may cause slight discrepancies, even between different conversion goals within Google Ads itself. Refer to Google’s Consent Modeling documentation for more details.
Updated on: 22/11/2024
Thank you!