What is a Highly Popular Post? Popularity Benchmark Models for Posts on Facebook Pages

Abstract

Posts have a very significant influence on the operational performance of Facebook Pages. Operators of Facebook Pages also focus on how to release posts that catch the attention of their fans. In past studies, no benchmarks were introduced as to how to calculate the popularity of posts. Therefore, with this study, a total of 125,302 posts are collected from Facebook Pages that fall in the four categories, namely figures, brands, media, and others applying the big data concept. After that, sequentially the categorical analysis for Stage 1, k-means clustering for Stage 2, and logistic regression analysis for Stage 3 are performed in order to establish benchmark models for the popularity of posts. The study has rendered a total of ten benchmark models with high and medium popularities, respectively, for figures, brands, media, others, and the general category (for which no categorization is done). It has also been found during the study that for Facebook Pages that are categorized as figures and others, their major operation strategy shall be to secure a high number of likes while for brands, the focus shall be on increasing the number of shares. In addition, among the different categories, comments do not contribute significantly to turning post highly popular.