Speech technology is gaining popularity and is used in a wide range of mobile applications and consumer devices. Speech-based interfaces circumvent the typing difficulties posed by soft keyboards. This paper describes the design and research of a speech-based interface for an enterprise sales application that combines speech recognition, natural language processing, and web services to drive the application. The paper further delves into the research methods applied—paper prototype testing, structured interviews, and a lab study—to gather feedback at various design stages. Finally, the paper describes how each of the three distinct research methods provided researchers with a more complete understanding of the user experience and led to a redesign.