For example, an insurer might use risk management tools to reduce premium leakage. Using geospatial imagery, underwriters first get an instant, comprehensive view of every property in their book. Then they can use computer vision tools to determine which properties have been appropriately priced according to the presence of real risk drivers. In many cases, they will find that certain attributes have been ignored or missed entirely (real roof condition, for example, often eludes the detection of physical inspection teams). Given a new, fuller understanding of real property condition, underwriters can more accurately price risk, reducing the possibility of premium leakage. This process can then be automated for future efficiency, especially when using a property intelligence platform equipped with a customizable flagging engine.