Idea platforms have become a well-established means for organizations for gaining valuable ideas from a diverse set of stakeholders such as customers, partners, or citizens. One major challenge in this field is how to make use of the full potential of those ideas. Hurdles towards this aim include a lack of necessary information in the idea descriptions to comprehend and effectively process them, redundancies among individual ideas and an exceedingly large pool of ideas, that make it difficult for decision makers to identify high-potential ideas. We address these challenges in the idea elaboration stage to lay the foundation for choosing the best ideas easier for further processing or evaluation by others. We choose a conversation based approach to challenge the initial idea contribution by a single contributor through structured facilitation by a non-human facilitator in the form of a conversational agent (CA). By instantiating the concept for a chat based CA in a Wizard of Oz study, we found exploratory indication that automated facilitation through a CA has the potential to be a feasible approximation to human facilitation in situations where the latter is not possible or where social control by a human facilitator might impede creativity or openness