An Integrated Service Quality and Sentiment Analytics Framework for Assessing Customer Satisfaction in Indian Quick-Commerce Delivery Platforms

Abstract

The fast-growing Indian quick-commerce (q-commerce) has disrupted retail with fast delivery in 10–30-minute windows. This study presents an Integrated Service Quality and Sentiment Analytics Framework to measure customer satisfaction through a combination of operational measures that are structured and unstructured textual responses. The study evaluates core dimensions in terms of delivery time, accuracy of order and service quality analysing a dataset consisting of 500,000 reviews on such platforms as Blinkit and Zepto. The study is methodologically a hybrid as it deploys Multiple Regression, ANOVA and Machine Learning algorithms (Random Forest, Neural Networks) and NLP-based sentiment analytics. Results show a moderate level of satisfaction (3.003), which means that speed of delivery is no longer a differentiator, but rather a standard characteristic. It is interesting to note that there is a notable difference in satisfaction depending on the type of the platform (p < 0.01), but not on the geography. This study provides a data-based model of the digital service-dominant logic that managers can use to enhance last-mile delivery and customer retention in a high-velocity market.