Using previous cases and our own pilots as data, we uncover the main difficulties in understanding and working with energy users. We argue that formal user research and interaction methods are helpful, yet insufficient for project success or even genuine user responsiveness.