Overview of this campaign

For any news site, driving sales is the lifeline of revenue growth, and our campaign takes an innovative leap to unlock its full potential. The front page is the most powerful tool for this growth—it engages users already on our platform, requires minimal external cost, and delivers outstanding results.

Building on past insights, we know that delivering precisely tailored content to the right audience is key to engaging users and converting them into paying subscribers. The industry's biggest challenge in achieving effective personalization lies in the lack of robust reading history and user data for non-logged-in users—an opportunity we've embraced to innovate and excel.

Over the past year, we developed a groundbreaking Machine Learning model that seamlessly integrates diverse data sources to quickly identify which content types drive user subscriptions. Leveraging existing data sources, such as age and gender predictions from our advertising business and processed sales data, we've built a compact set of features that enable precise, real-time content recommendations with minimal input data.

This solution stands out by generating recommendations on-demand, rather than through traditional batch processes. This innovation ensures that recommendations improve instantly as new user data becomes available. No need for batch re-runs—our model adapts dynamically, factoring in contextual elements like time of day and day of the week to maximize sales impact.

Our model is fully integrated with overall content recommendation systems, blending its insights with editorial signals and other performance metrics. This ensures optimal performance while maintaining strong editorial control, empowering us to deliver both personalization and journalistic integrity in a single, unified approach.

Given that we don’t really have any data on our prospects for sales, being anonymous users, and we lack the lakes of data for all our users that our competitors in the app and social world, this is really about adopting the mindset of selling the strawberries you have at hand.  

Results for this campaign

The results of this machine learning approach to scoring articles for sale has been tremendous. Although one could argue that segmenting content based on how it generates sales within different segments is unclassy and generalizing both content and users, this method and the nuances beg to differ. 

The new model has been A/B tested against our existing models, as well as in positions high up on the page where it can reach more readers. 

Against our old yet optimized recommendation approach focusing on sales pr impression, the sales increased by on average 75% from articles on our front page with the new model. 

So reality is that although some data sources are scarce and the quality for anonymous users are below what would be optimal, the results have been staggering. 

Moreover, we are already on the steps on adding more data and further exploring the potential for this solution. 

During our development we tested 158 different data points for their impact on likelihood for promoting sales, and ended up with a dozen features.

This model has also unlocked the further potential for optimizing content for anonymous users in other areas to increase engagement rather than sales.

 

Contact

To contact a company representative about this campaign, click here for the INMA Member Directory