Applied trade analysis and modeling would not be possible without databases. Over the last few years, a considerable effort has been devoted to put together and harmonize a huge amount of data. This has lead to the construction of large-scale datasets on trade (imports and exports in volume and value terms). In parallel, increasing attention has been devoted to measuring tariffs and other trade barriers. These efforts are still under way. It is only recently that large-scale models, for example, have taken into account the lower tariffs applied under preferential agreements (the former generation of models only considered the "bound" tariffs that act as a ceiling in many countries). The changes resulting from the introduction of better data on tariffs have been very significant, and many institutions have revised considerably their estimates of, say, gains resulting from trade liberalization (Ackerman 2005). However, empirical difficulties faced by modelers suggest that there is still a need for improvement, especially in the area of applied tariffs and tariff rate quotas. The aim of this study is to review the main data sources dealing with world tariffs and trade. After discussing some issues and methodology, we provide a brief description of the main datasets available in terms of the origin of the data, its accessibility, reliability and shortcomings.