User-Centered Conversational AI for Small Business Customer Service: A Cost-Effective and Accessible Framework

Abstract

The rapid digitalization of small and medium-sized enterprises (SMEs) has highlighted the need for cost-effective and accessible customer service solutions. Traditional commercial conversational AI platforms, such as Dialog flow and AWS Lex, often present high costs and steep learning curves, which pose significant barriers for SMEs. This paper proposes a user-centered conversational AI framework based on an open-source technology stack, including the Microsoft Bot Framework (MBF), OpenStreetMap, Azure Cognitive Services, and MongoDB. The framework is designed to support essential customer service functions—automated order processing, FAQ management, and voice-enabled interactions—while maintaining accessibility compliance (ADA). We demonstrate that this approach reduces operational costs, simplifies deployment, and enhances usability for SMEs. Experimental evaluation comparing our system with commercial alternatives shows competitive accuracy and latency, alongside improved accessibility satisfaction. Finally, we discuss the practical implications, limitations, and future directions for expanding knowledge automation and sentiment analysis integration.