Google Analytics 4 (GA4) comes with a powerful attribution modelling tool: GA4 DDA.
GA4 DDA is leveraging machine learning to analyse converting and non-converting path on your website. The resulting data-driven model gains knowledge about how various touchpoints affect conversion results. The model takes into account variables like conversion time, device type, ad interactions, exposure order
, and creative asset type. The model can then assign credits for conversions to each touchpoint.
This is an easy-to-use attribution model; however, like any machine–learning based model, DDA is a black box. It is therefore fair to ask: can you trust GA4 DDA attribution data?
The answer will depend a lot on the channel. GA4 DDA rely on data from your web analy
itics. It is not able to account for interactions happening outside your website. For example, if a prospect interacts with a carousel ad on Instagram, but does not click to visit your website, GA4 won’t be aware of the interaction. This is a one key limitation of a click-based attribution model like GA4 DDA that we explore more in detail in our case study with Expondo.
Datart is a retail chain that sells electronics, appliances and other home goods. Their e-shop is ranked in the top 60 most visited in the Czech Republic, so they have a lot of data for us to look at.
As expected, GA4 is under evaluating interaction from social sites. Instead, GA4 tends to overestimate the impact of organic search as most social interaction results in a brand search.
ROIVENUE can provide better attribution data as we combine web analytics data with 3rd party tracking tools like Facebook pixel. This promises to give you far more reliable data for walled garden platforms like Facebook or Instagram.
Interestingly, Google is not using GA4 DDA to overestimate the impact of google ads. However, if you are heavily using social media ads, you might want to implement ROVENUE to get a better picture of your ad performance.
Thanks to Roivenue’s data-driven attribution analysis we were finally able to understand each channel’s true effectiveness. When we reallocated budgets towards the best-performing ones, we immediately saw an uplift in the overall performance.