Anyone with a smartphone can take pictures anytime anywhere. This opens the opportunity for researchers to collect photos to augment traditional Web survey data. We conducted an experimental survey asking 2700 members of the Dutch LISS panel about their dwelling conditions. Depending on the condition, they were asked to either take several photos of their house or answer a set of survey questions about the same topics. This paper documents the feasibility of collecting pictures instead of answers in a web survey and studies the consequences of using pictures/questions or a choice for components of Total Survey Error. We find that respondents in our study were much more willing to answer survey questions than to take pictures, but this difference depended on the difficulty of the topic. We then compare the quality of the data from the pictures of the heating systems with the survey responses. Here we find that our pictures contain more useful information than the survey responses. Overall, we conclude that asking respondents to take a picture within a survey can be a feasible way to collect high-quality data.