Using text mining and crowdsourcing platforms to build employer brand in the US banking industry
Crowdsourcing platforms are a valuable source of information for employees and firms. They have become an essential part of human resource practices, such as recruiting, employee engagement, and the marketing of a company's brand. Crowdsourcing platforms can help enrich an employer's brand, with employee reviews acting as a terra-incognita for the early-stages in the employer branding process. In response to these dynamics, this research explores the employer value proposition preferences and the sentiments of employees in the US banking industry. Approximately 11,000 reviews on the crowdsourcing platform Indeed.com, covering the top 18 banks listed on the Fortune 500, were parsed. The text mining techniques topic modeling and sentiment analysis were carried out using Latent Dirichlet Allocation. The topic modeling analysis revealed work-life balance, management, economic, social, and development values as the most preferred employer value proposition, while brand-image, application, and interest values were the least preferred. Sentiment analysis showed that employees exhibit negative sentiments for management, work stress, work-life balance, and economic values, and show positive sentiments for development, social, and interest values. The paper highlights avenues for future research and discusses the managerial and theoretical implications for managers and HR practitioners.