Sanoma and Google set out to develop smart customer analysis in a machine learning based personalisation engine. Machine learning is the study of computer algorithms that improve automatically through experience. A personalization engine is a tool used to collect and analyze customer behavior and data to create a customized user experience, in this instance on the digital front-page of Helsingin Sanomat. The aim of implementing these tools was to improve the subscription income of existing subscription channels. The expectation was that more detailed information on customer groups would help to define the likelihood to pay as each customer group has different sales potential and a reader is more likely to subscribe if the content is relevant to them.

Since implementing its personalization engine and propensity-to pay model, Helsingin Sanomat has increased its subscriber base to 400,000, growing subscriptions from 23% of sales to 60% in 2020. “Our understanding is that we are now one of the leading players within the news media industry in utilizing content serving algorithms on our front page,” adds Vesa Lindqvist from Sanoma. “The complexity and data utilized in Helsingin Sanomat’s front page algorithms surpass the level of many media companies’ work that we have benchmarked against. Moreover, our algorithms have many different problem solving solutions. Helsingin Sanomat’s uplift in engagement proves the point in practice.”