Abstract
The conversational nature of chatbots poses challenges to designers since their development is different from other software. Since human dialogue involves several variables beyond verbalizing words, it is vital to design wellthought dialogues for chatbots to provide a humanized and optimal interaction. The main objective of this work is to unveil textual, visual, or interactive design practices from text-based chatbot interactions and how they can potentiate or weaken some perceptions and feelings of users for the creation of the Guidelines for Chatbot Conversational Design (GCCD) guide. We used multiple research methods to generate and validate the guide. First, we conducted a Systematic Literature Review (SRL) to identify conversational design practices and their impacts. These practices were inserted into the GCCD guide through qualitative analysis and coding of SLR results. Then, the guide was validated quantitatively through a survey and qualitatively through a case study. The survey aimed to assess the guide’s clarity and usefulness based on its reading and the case study aimed to assess the guide’s usefulness based on its practical application. The survey showed that software developers with different levels of experience strongly agreed that the guide could induce greater user satisfaction and engagementm that the guide is clear, understandable, flexible, and easy to use. The case study confirmed the survey findings, as participants reported positive feelings toward the guide and an intention to use it. Their extensive perceptions given through the conducted interviews unveiled that their previous experiences with chatbots and in specific software development positions influenced their design and adoption of practices. The guide proved to be useful for developers with different levels of knowledge, with the potential to become a strong ally for developers in the conversational design process.
Publisher
Sociedade Brasileira de Computação (SBC)