1 min read

Divante helps grow ROI by leveraging a tool made by Facebook

Care to share?

Facebook recently released an open-source code called Robyn that’s an automated marketing mix modeling (MMM) algorithm. As eCommerce experts we decided to create a solution based on Robyn that would enable the analysis and prediction of budgets and marketing results.

Internet giants such as Apple or Google started blocking third-party cookies that are responsible for the possibility of users’ behavior tracking. It has a significant impact on measuring and planning advertising activities. Those cookies are, for example, files not used by the website visited by the user but by its partners.

You can see the idea and comparison of cookies in the table below:


In light of the changes mentioned above, measuring digital marketing efforts is becoming more and more complicated or fragmented. Robyn, the algorithm made by Facebook, was created to enable the analysis of results and planning of activities regardless of access to third-party cookies. As Kyle Goldberg from Marketing Science at Facebook says:

We want to help all companies to have access to advanced, privacy safe, efficient and rigorous marketing effectiveness analytics in the context of changing digital marketing landscape and consumer behavior.

Robyn allows marketers to take control over cross-media budget allocation decision making with continuous modelling. This allows them to understand the performance of marketing in almost real time.

We believe that using this modelling tool, powered by artificial intelligence (AI), in an eCommerce environment will enable business owners to measure their activities in a safe, accurate, and actionable way. In addition to the ability to measure and predict the direct impact of marketing on revenue, we decided to enrich the offer of this solution with the impact of individual eCommerce elements. This includes page loading speed or a selection of payment methods. 

What does it mean for Divante?

At Divante, we always think about what’s next. For sure, AI-driven decisions on how to improve eCommerce is a step forward. We already make well-informed predictions on how packaged business capabilities (PBCs) help our clients to grow, but now we’ll be able to predict it. - Artur Wala, Head of Innovation Lab

You can read more about the product itself and our approach here.

Are you feeling concerned about measuring and predicting the data after privacy changes?  Contact us and we’ll guide you through.

Published August 11, 2021