Abstract. Goal-orientated Conversational Agents are a specific family of conversational agents that are designed to converse with humans through the use of natural language dialogue to achieve a specific task. Traditionally, they utilise pattern matching algorithms to capture the values of specific attributes through their values through dialogue interaction with a user. This is achieved through the use of scripts which contain sets of rules about the domain and a knowledge base to guide the conversation towards achieving a specific goal. Such systems are ideal for providing clear and consistent advice 24 hours a day in many different scenarios, including advising employees about their organisations policies and procedures, guiding a user through buying a suitable product, and tutoring a student to understand a learning objective. This paper presents an overview of a methodology for constructing goal orientated conversational agents. Three case studies which employ this methodology are introduced and evaluated.