This paper presents a machine learning methodology for predicting the success of online movie news with web analytics data from Google Analytics and social media analytics data from Twitter (now known as X). This methodology is built in two phases. The first consists of segmenting the data with the different categories that apply to the articles (in this case, all news published, news about movies, and news with a trailer) and performing a multiple linear regression to extract a prediction equation for each success variable. The second phase includes validating the prediction equations with test data, which helps to select the most reliable prediction depending on its accuracy. This methodology has shown that it can account for some of the variability of the success variables and can make their prediction. It provides a basis for future research to improve accuracy and test new variables that can enrich this analysis. This article also helps editorial teams make better data-driven decisions, such as more efficient resource planning or optimising articles to achieve a specific goal.