A consortium comprised by the Global Earthquake Model Foundation, ImageCat Inc. and the Humanitarian OpenStreetMap Team has been chosen by the Global Facility for Disaster Risk Reduction and Recovery to develop an open exposure database for multi-hazard risk assessment, as part of the Challenge Funds supported by the Department for International Development of the United Kingdom. This database is capable of storing different assets, while considering relevant attributes for six natural hazards: earthquakes, floods, volcanoes, strong winds, tsunamis and drought. The development of an open database to characterize the built-up environment at a global scale requires a uniform methodology to classify the elements exposed to the natural hazards. This deliverable describes a comprehensive multi-hazard GED4ALL taxonomy capable of classifying the building stock, lifelines, crops, livestock, forestry and socio-economic data. This taxonomy will be applied to the country of Tanzania to develop an exposure model at the national scale, and to the city level of Dar Es Salaam to demonstrate how an exposure dataset at a building-by-building resolution can be created. Moreover, this process will be demonstrated for five countries around Tanzania (Ethiopia, Uganda, Kenya, Malawi and Mozambique).