Z
illow has pulled its climate‑risk overlays from listings, a move that has sparked a wave of pressure on other major real‑estate portals to do the same. The catalyst is the California Regional Multiple Listing Service (CRMLS), one of the country’s largest private listing databases that feeds Zillow’s inventory. CRMLS has raised concerns that the flood, wildfire, wind and heat risk scores supplied by First Street—a private risk‑modeling firm—are inaccurate and may be harming home sales.

    First Street builds predictive models for climate‑related hazards and sells the resulting risk scores to banks, insurers, asset managers and real‑estate sites such as Redfin, Realtor.com, Homes.com and, until recently, Zillow. CRMLS CEO Art Carter publicly criticized First Street’s forecasts, claiming that its “future predictions ended up being very wrong” for California flooding. He pointed to listings that carried a high projected flood probability despite a long history of no flood events, and warned that such labels can deter buyers.

    In response, CRMLS has asked Realtor.com, Redfin and Homes.com to remove First Street’s flood probability data and flood‑risk maps from their platforms. Zillow, in turn, said it had “updated our climate risk product experience to adhere to varying MLS requirements and maintain a consistent experience for all consumers.” The company notes that First Street data remains linked to its listings, but is now less prominently displayed than on Redfin and Realtor.com.

    Redfin’s spokesperson Angela Cherry said the site will continue to provide risk estimates for fires, floods and storms, but sellers can request removal if they believe the data is inaccurate. Realtor.com’s Sara Wiskerchen confirmed the company is working with CRMLS and its data providers to investigate the concerns.

    The debate over data accuracy is not new. First Street’s founder and CEO Matthew Eby claims its models outperform California’s state flood maps, citing the 2024 Los Angeles wildfires where its wildfire‑risk maps identified over 90 % of homes that burned in the Eaton Fire as “severe” or “extreme,” compared with only 21 % flagged by the state. The state has since begun revising its own maps.

    Zillow’s internal analysis supports the notion that climate risk data influences sales. Homes listed in June 2024 with a high flood risk sold at a 52 % rate by March 2025, versus 71 % for low‑risk homes. First Street argues that its data is more reliable than limited federal sources such as FEMA flood maps.

    However, independent researchers point out that private firms like First Street keep their modeling algorithms confidential, making external validation difficult. In a 2024 study, climate‑risk scientist Oriana Chegwidden found wide variations in property‑level risk predictions across nine private companies, with only two willing to share their models. Chegwidden stresses the importance of transparency, noting that risk assessments can affect insurance eligibility and property values.

    First Street maintains that its models are built on transparent, peer‑reviewed science, and that providing private data to consumers is essential to protect buyers from long‑term financial harm. Eby argues that the current real‑estate market—characterized by low inventory and high prices—exacerbates the pressure to close sales, making climate‑risk data harder to ignore.

    In short, Zillow’s removal of extreme‑weather overlays has set off a chain reaction among major listing sites, as CRMLS and other stakeholders question the accuracy of private climate‑risk models. The outcome will shape how buyers access—and trust—information about environmental hazards in their future homes.