Social Media Competitive Analysis and Text Mining: A Facebook Case Study in a Local Television Market

Introduction

As more Americans use Facebook as one of primary sources of news consumption, television broadcasters have increasingly added this social media platform in their distribution strategies for audience engagement and program optimization. Recent industry studies revealed that Facebook is a dominate platform for local broadcasters; use of Facebook stands out as a central social medium and a streaming distribution service (Knight Foundation, 2016; Pew Research, 2018). In particular, television newsrooms utilize Facebook News Feed and Facebook Live to prioritize local news, strengthening their important roles in serving as trusted resources for essential community information (Regan, 2018). The social media revolution has gradually reshaped news organizations and their news operations, generating a vast amount of social media data (Gramlich, 202, Guo & Sun, 2020). Those audience data and analytical insights provide television broadcasters with valuable audience graphs for deep segmentation and optimizing content to better serve audience needs. Those media outlets can examine social media users’ news consumption patterns, identify potential threats, assess vectors for growth, and provide metric-driven insights that will inform and improve production pipeline.

To increase competitive advantage and effectively assess the competitive environment of business, television broadcasters need to constantly monitor and analyze social content and audience data from both their own social media platforms and competitors’. In the U.S. television industry, there are usually three to five local stations within a designated market area (DMA), each of which is affiliated with one of the top five national English-language broadcast networks (i.e., ABC, CBS, FOX, NBC, and The CW). These network affiliates offer similar products (i.e., local news programming) to target similar audiences (i.e., local communities). The on-going set of competitive actions and competitive responses occur almost every day online and offline, as these local television stations compete against each other directly for advertising dollars. Social media competitive analysis therefore provides them with an additional way to engage with digital audiences and gain competitive advantages over their competitors.

By integrating quantitative analysis and text mining, the purpose of this study is to investigate the dynamic Facebook competition among five local television stations by applying a social media competitive analytical framework. This investigation employs two sets of social media competitive analysis metrics to investigate a mid-sized television market in the U.S. The first set of metrics focuses on the local television stations’ Facebook posting activities and content strategies. It applies a text-mining technique to analyze unstructured text content posted on Facebook by these five local television broadcasters. The second set of metrics focuses on measuring Facebook user engagement behavior interacting to those news posts. Recommendations are discussed to help local television outlets understand how to perform a social media competitive analysis, transform social media data into knowledge for decision-making, and develop their own social media strategies to gain competitive advantages over their opponents.