Arguing that photo-interviewing yields richer data than that usually obtained from verbal interviewing procedures alone, it is proposed that this method of data collection be added to "standard" methodologies in instructional development research and evaluation. The process, as described in this paper, consists of using photographs of "episodes" in a project as visual cues both in interviewing subjects and in reports for stakeholders in the outcome of the project. It is noted that the technique can be particularly helpful for pilot-stage projects where innovation and development are organizational priorities. Some strengths of photo-interviewing are cited, and a nine-step strategy is proposed: (1) identify major questions and research or evaluation perspective; (2) prepare the data gathering approach; (3) photograph episode; (4) organize and label photographs; (5) prepare protocol; (6) pilot interview; (7) conduct and videotape interview; (8) analyze photographs, field notes, interview notes, and videotape; and (9) summarize findings. Discussions of the individual steps include examples from two attempts to incorporate photo-interviewing with inquiries involving separate program innovations, e.g., an intensive 40-hour hypermedia mini-course for higher education faculty and graduate students, and research, with high school juniors and seniors, which considered ways in which matrices could function as systematic organizing structures for concept and simple rule learning. Cognitive, ethical, and technical issues related to photo-interviewing are discussed, and an outline of implications drawn from these issues concludes the paper. A section of a photo-interview script that emphasizes critical incidents is appended. (31 references) (BBM)