Understanding the drop in PPC performance: A result of Advanced Consent Mode in GA4
Issue overview
In some cases, there has been a significant drop in PPC performance, specifically a 5% to 20% reduction in reported conversions within GA4.
These conversions are instead attributed to the Direct channel, resulting in less precise data and creating challenges for decision-making, especially with already constrained budgets.
The background
Data gaps in user journeys are not new to digital marketing. Historically, users' use of multiple devices, ad blockers, and evolving privacy regulations have exacerbated these gaps. Google’s Data-Driven Attribution (DDA)* model was.
Introduced in 2013.
Integrated into GA4 in 2020.
Made the GA4 default reporting model in 2023.
DDA uses machine learning to provide a complete picture of user behaviour by attributing conversion credit to the appropriate interactions and channels.
However, new challenges have emerged with the introduction of Advanced Consent Mode**.
What’s happening now?
The implementation of cookie banners, designed to comply with privacy regulations, has significantly reduced the amount of data being tracked - this can range from 10% to 50% data loss.
Advanced Consent Mode was introduced to improve data collection while maintaining compliance.
The goal is to compensate for missing data using Google's modelling algorithms*** depending on how consent is collected and managed.
However, Google’s algorithms default to the Direct channel when insufficient or inaccurate data is available, which has contributed to this trend.
While Google's cautious approach is designed to avoid overcompensating for missing data, this shift is problematic for reporting, insights, and performance evaluations.
What you need to do to address the issue
While dealing with inconsistencies in digital data is frustrating, it’s important to remember that no single source provides perfect accuracy. The focus should be on improving data quality and working within the limitations of current technologies by:
Checking modelled data: Review your reports to see if modelled data (including estimated user data) is being included. This may help explain discrepancies.
Benchmarking comparisons: Compare current performance with historical benchmarks to identify any significant changes in attribution patterns.
Using multiple reports: Combine different reports to conduct a more comprehensive analysis and gain greater confidence in your insights.
Reviewing your cookie consent approach: Ensure your cookie banners are optimised for maximum consent rates while remaining compliant with regulations. This can help reduce the amount of missing data.
Consulting experts: Accord’s experienced Measurement and Data team specialises in addressing these challenges and improving client data quality, so you get more accurate insights from your digital media performance.
Finally, Google has released four updates**** this year alone to refine its attribution models, and we can expect continued advancements. For now, patience and proactive optimisation will help navigate these challenges while we wait for more reliable data from evolving technologies.
References:
*[GA4] Get started with attribution – Data-driven attribution
***[GA4] About modelled key events – Google’s key event modelling approach
Dan Ward
Dan is Accord’s Digital Director and GA Guru. Get in touch if you want to pick his brains on all things marketing performance.
