Real estate platforms, like Zillow, Redfin, and Apartments.com, have transformed real estate information access in the U.S. since they launched in the early 2000s. They assemble information about real estate listings from various sources, including public information (sales and tax history), information from other platforms (Google Maps, Yelp), and listing information that was previously only accessible to real estate agents (through the Multiple Listing Service). On the real estate platforms, home-seekers encounter a range of neighborhood scores – numerical rating and basic heatmaps that rank the schools, walkability, transit, crime and noise of a neighborhood. While some scholars and policymakers praised the platforms and these metrics for making information more universally accessible and potentially correcting historical inequalities in access to housing, others raised concerns that platforms would continue consolidating power for a few actors in housing. Both perspectives fail to consider the multiplicity of the user interactions with the platforms. By contrast, this dissertation takes an interpretivist approach to investigating the platforms and their users. I draw on 32 in-depth interviews with home-seekers from Las Vegas, NV or Oakland, CA, additional interviews with housing experts and activists, and online and in-person participant observation as I recruited interview participants and conducted my own online housing search. Instead of treating information as self-evident, my research shows that people interpreted the platforms’ metrics in considerably varied ways. Different users, for example, could draw opposing conclusions from the same neighborhood score. I also present evidence of online housing search activities that extend beyond real estate platforms and analyze user folk theories about the price-estimating algorithm, the Zestimate, that challenge media narratives that hype its influence. This research reveals key gaps in current policy framing of both housing search and platform use.