MarkBot - A Language Model-Driven Chatbot for Interactive Marketing in Post-Modern World

Hosting conversational responses on the official websites of products and services companies is an essential marketing aspect. With Artificial Intelligence's help to make conversational interactivity more intuitive to existing and potential customers visiting the websites, managers can notch up the return on marketing investments. This motivated us to study empirically and develop the MarkBot framework, a chatter robot on the management design principles. The framework uses an Artificial Intelligence application to respond to a website visitor's browse through the product catalog. Neural network (NN) architectures are known to achieve remarkable performances in synthetic text predictions. We use a long short-term memory recurrent neural network (LSTM) to predict the user's responses through a chatbot in the current work. The proposed framework reduces the lead time for the firms to adopt MarkBot. We empirically prove using user-generated content on social media platforms like Twitter in responses and queries to digital campaigns on the same product. With new businesses failing to venture into the space of hosting a chatbot owing to no historical data or existing firms yet to host a chatbot, the proposed MarkBot fuelled by user-generated content can have a substantial managerial implication. The management frameworks used to theorize the MarkBot also make it a theoretical contribution for future Information Systems scholars to conceptualize in the marketing field.

Automated summary

Hosting conversational responses on official websites is important for marketing, and the MarkBot framework leverages AI to enhance the interactive experience. Using LSTM neural networks to predict user responses, the proposed framework reduces the lead time for firms to adopt MarkBot.