Results with Consent Mode

👋 In this document you'll discover the results of Google Consent Mode implementations on several sites. Here we provide some comparative remarks and potential side effects.

Results with Consent Mode

After configuring dozens of containers integrating the consent mode, we can now make an initial assessment of the benefits of behavioral modeling.

In every analysis you carry out, it's essential to know the CMP acceptance rate of the site concerned. This will give you an idea of how much data is being modelled in Google Analytics.

An example of a site with access to data modeling

This e-commerce is eligible for data modeling in its Google Analytics 4 interface. Data from its CMP indicates that 70% give their consent. Modeling is therefore performed on 30% of the site's traffic.

In order to perform this type of comparison, you need to activate and deactivate the integration of modeling in reports. To do this, go to Administration → Reporting identification. The data Mixed integrates modeling as opposed to data Observed.

💡Data modeling eligibility requirements:

• The consent mode must be activated on all pages.
• Google tags load before the consent banner appears.
• Property GA4 receives at least 1000 events with analytics_storage='denied' for 7 days.
• Property GA4 receives at least 1000 users with analytics_storage='granted' pendant 7 jours.</aside>

💡 Blended vs Observed

• Mixed: Data from consenting users + a model for data from non-consenting users
• Observé : Uniquement les données des utilisateurs consentis.</aside>

The acquisition report

One of the most frequently consulted reports is the traffic acquisition report. It sheds light on where users come from, and how they behave once they've arrived on the site.

A site's RGPD compliance will therefore have a major impact on the web marketer's analysis of its acquisition.

In the case presented below, we can see that the consent mode and its modeled data boost user volume and traffic by 38%, events by 34% and conversions by 26%.

Even more importantly, Google Consent Mode measures the impact of SEA even more clearly, with a 49% increase in traffic share compared to +35% for SEO. The increase in sales associated with these two channels is identical, at 28%. The readings are different, particularly on the conversion rate between the Observed and Mixed methods :

Event report

Once on the site, what do users do? As Google Analytics has an event-based data model, this report highlights the actions most frequently carried out on the site. Using our e-commerce site as an example, we can once again see the benefits of Google Consent Mode.

The increase in page views using behavioral modeling is 25%, and 30% for product pages. Of course, it's also possible to go into much greater detail, to see changes by page and by product.

The conversion report

This report is probably the one you follow the most. It is the link between your marketing efforts and your business objectives. So, by viewing incomplete data, your decisions could be based on the wrong signals. That's why you should always compare these figures with your internal data, to rule out any outliers. As in the other two cases presented above, the consent mode changes the reading grid. The site concerned recorded €20k in additional sales, i.e. an increase of 22%. At the same time, sales increased by 26%, implying a lower average shopping basket of 1.12€.

Comparing data from ten or so sites eligible for behavioral modeling, I can see that the data added varies between 15% and 40%. However, these figures will be greatly impacted by the consent rate, so it's in your interest to maximize your efforts to get visitors to give it.

In the case presented above, the figures presented by Google were close to internal data, but it's always essential to compare. To gauge the quality of the data reported, you need to compare it with your CRM and/or back-office data. Are basket additions, number of sales and sales figures consistent? If so, this is a good indicator for the rest of the statistics.

‍