Utilising NLP to support Voice of Customer analysis in an industrial B2B company
Outinen, Anni (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025112529612
https://urn.fi/URN:NBN:fi:amk-2025112529612
Tiivistelmä
Voice of Customer (VoC) is a process for collecting insights into customer experience, analysing them, and reacting to them. Effective VoC analysis enables organisations to address customer pain points, thereby improving the customer experience and business performance. Customer experience can be measured through feedback surveys, for example, but analysing the results, particularly the free text responses, can be challenging due to the unstructured nature of the data and often because of the volume.
The aim of this research project was to study how a B2B organisation operating in industrial material handling can utilise text analytics for customer feedback analysis. The commissioning organisation has a feedback program in place, but currently lacks a way to collect insights from the free-text responses from customers. Natural language processing has been found to support feedback analysis, and so the objective of the project was to build a solution that would be able to process the comments and provide the organisation with insights into the customer experience.
The theoretical framework discusses key concepts of customer experience, as well as Voice of Customer. Natural language processing is discussed in the context of customer experience, with a focus on topic modelling and sentiment analysis as key natural language processing methods. Also, both the benefits and challenges of implementing these approaches are covered. The research project utilises agile methodology, as the commissioning organisation also operates using an agile approach to development. The data collection is conducted through industry benchmarking, collective brainstorming with business stakeholders, and A/B and usability testing. Collaborative analysis is utilised throughout the project.
The research project supported the production of a text analytics solution that processes customer comments and identifies key themes discussed and the sentiment behind them. The results show that natural language processing can support Voice of Customer analysis, as this process brings to light the organisation’s strengths and weaknesses from a customer perspective. The research findings support the literature, suggesting that a text analytics solution should be tailored to the organisational use case through additional context and terminology that may be company- or industry-specific.
The aim of this research project was to study how a B2B organisation operating in industrial material handling can utilise text analytics for customer feedback analysis. The commissioning organisation has a feedback program in place, but currently lacks a way to collect insights from the free-text responses from customers. Natural language processing has been found to support feedback analysis, and so the objective of the project was to build a solution that would be able to process the comments and provide the organisation with insights into the customer experience.
The theoretical framework discusses key concepts of customer experience, as well as Voice of Customer. Natural language processing is discussed in the context of customer experience, with a focus on topic modelling and sentiment analysis as key natural language processing methods. Also, both the benefits and challenges of implementing these approaches are covered. The research project utilises agile methodology, as the commissioning organisation also operates using an agile approach to development. The data collection is conducted through industry benchmarking, collective brainstorming with business stakeholders, and A/B and usability testing. Collaborative analysis is utilised throughout the project.
The research project supported the production of a text analytics solution that processes customer comments and identifies key themes discussed and the sentiment behind them. The results show that natural language processing can support Voice of Customer analysis, as this process brings to light the organisation’s strengths and weaknesses from a customer perspective. The research findings support the literature, suggesting that a text analytics solution should be tailored to the organisational use case through additional context and terminology that may be company- or industry-specific.