Interviews with individuals experiencing homelessness can be challenging for various reasons, including mental and physical health issues, substance use, and negative experiences with authority figures. Researchers have used photos to facilitate communication and empower participants during data collection. We analyzed data from a previous study to explore the use of researcher-generated photos during interviews about health with 13 women experiencing homelessness. Conversation analysis revealed clear patterns regarding the use of the photos during the interviews. The photos were referred to 118 times over the total interview length, 6 hours and 23 minutes, with the interviewer making 62% of the referrals and the women accounting for 38%. Fifty-nine percent of the referrals occurred within the first 5 minutes of the interviews. The women used the photos to trigger associations and emotions, to describe photo content, or in a minor role during the interview. Interpretations from an advisory board of six women with lived experiences of homelessness suggested that the photos did not engage participants as intended, highlighting the importance of considering participants’ perspectives when designing photo-elicitation methods. The feedback also provided valuable insights into interview locations and incentives in research that may have influenced the women’s willingness to use the photos. This study emphasizes the importance of understanding the complexity of choosing researcher-generated photos in interviews with underserved, hard-to-reach populations.