Work

Utilization of text analytics in improving customer experience

Year: 2023
Type: dissertation
Abstract: Collecting data is not enough to generate value, because data needs to be turned into information and insights to be valuable. Particularly unstructured data, such as text, is difficult to turn into valuable insights due to its lack of predefined structure. The goal of this case study was to explore how text analytics, specifically topic modeling and sentiment analysis, could be used to measure and improve customer experience of the case company. \n \nThe study consists of a literature review that focuses on text analytics and customer experience, and an empirical portion that is based on qualitative research; involving interviews with customer experience managers, focus group sessions, and secondary data from the case company. The results from research were used to form an understanding of the case company’s current state of customer experience, develop a topic model and a way to utilize it, and lastly implement improvement suggestions. \n \nThe research indicates that text analytics offers an additional method for measuring customer experience, complementing traditional measures like NPS and CES. It can also be integrated to a customer journey to measure touchpoints more accurately and gather customer understanding more efficiently. A topic model should be built iteratively by following a set of guidelines and in cooperation with experts that can provide business understanding. Furthermore, integrating the topic model with customer journey data and mapping facilitates evaluation of customer experience’s continuous improvement. To evaluate continuous improvement, text analytics should be used in touchpoint monitoring and prioritization, and implementing the suggestions can guide the case company to monitor touchpoints more effectively, provide enhanced customer understanding, and make prioritization more data-driven.
Author: Kianto, Visa
Language: English
Open Access status: green
OpenAlex ID: W7072169287
Cites: 51
Cited by:
Related to: 10