Mining voice of customers and employees in insurance companies from online reviews: a text analytics approach

Abstract

PurposeAlthough insurance companies contribute to about $600 billion of the US gross domestic product and employ 2.7 million people, the overall perception of clients and workers employed in this domain is poor. To the best of our knowledge, this study is the first to propose recommendations to improve the service quality of insurance companies using the voice of customers and employees that are available online.Design/methodology/approachThe proposed methodology consists of four stages: extracting text reviews posted by employees and customers, reviewing feedback using bigrams and trigrams, determining topics and identifying quality-associated managerial recommendations using the Strengths, Weaknesses, Opportunities and Threats (SWOT) and root cause analysis techniques.FindingsOur results indicate that primary causes for client dissatisfaction are improper client assistance services, inefficient claims processing, issues related to payments/fees and unresponsive ancillary services. Offering different types of payment methods, immediate car replacement policy and features such as smartphone applications are commended very well. The major reasons for workforce frustrations include long and overly complex training, lack of social events and incentive programs. Coworkers, workplace facilities, clean work environment and other amenities are shown to have a positive impact on employees' satisfaction level.Originality/valueTo the best of our knowledge, this study is the first to propose recommendations to improve the service quality of insurance companies using the voice of customers and employees that are available online. The proposed quality-related insights can assist insurance companies in improving the outlook of their workers as well as customers.

How to deeply understand the voice of the customer? A proposal for a synthesis of techniques for analyzing online reviews in the hospitality industry

“…In a different study, Rajendran et al. (2023) offered suggestions for enhancing the service quality provided by insurance businesses utilizing online testimonials from clients and staff.…”

“…Surveying research related to the VoC, text mining techniques to extract the VoC, especially in the field of hospitality, the research found some gaps: First, studies such as (Nicholas and Lee, 2017; Rajendran, 2021; Marcolin et al. , 2021; Balducci and Marinova, 2022; Rajendran et al. , 2023) just focused on exploring the VoC on texts in one language, that is English.…”

“…In a different study, Rajendran et al (2023) offered suggestions for enhancing the service quality provided by insurance businesses utilizing online testimonials from clients and staff. Proposed quality-related information will help insurers improve their customer's and personnel prospects.…”

“…Surveying research related to the VoC, text mining techniques to extract the VoC, especially in the field of hospitality, the research found some gaps: First, studies such as (Nicholas and Lee, 2017;Rajendran, 2021;Marcolin et al, 2021;Balducci and Marinova, 2022;Rajendran et al, 2023) just focused on exploring the VoC on texts in one language, that is English. Either the research of Alamsyah et al (2021) only mentioned Indonesian; Second, the techniques to extract features from the customer's voice are tokenization (Rajendran, 2021), LSA (Sezgen et al, 2019), and IDF (Nicholas and Lee, 2017).…”

Advanced Sentiment Analysis for Managing and Improving Patient Experience: Application for General Practitioner (GP) Classification in Northamptonshire

“…The package is preferred ahead of XML or HTML parser as it supplies functions for searching, modifying, and iterating the parse tree. The Beautiful Soup library along with the Selenium WebDriver assists the program to navigate through multiple web pages [ 41 ]. RAKE is an algorithm that requires simple input parameters to automatically extract keywords from a body of text by analyzing its coincidence with other collections of words and its frequency count [ 42 ].…”