Encouraged by the emergence of a new and growing literature on the use of trademark data as complementary indicators of innovation, this paper strengthens the case for such use by offering both qualitative and quantitative evidence in its support. Based on two large trademark and patent databases built for this purpose, the paper makes the argument that careful use of trademark data can improve our understanding of innovation processes across all sectors of the economy. Our databases generate a high correlation coefficient between patents and trademarks in four different scales of aggregation: firm, sector, country, and global. We argue that the combined use of patent and trademark data is better at capturing the whole gamut of innovation that takes place in the economy than the standard patent data alone. We embed our argument and analysis throughout in the main structural changes that have occurred in the global economy since the latter part of the twentieth century. We find that in certain cases, e.g. in service sectors, in sectors where patents are not widely used, and in the case of exporting entities from developing countries, trademark data are better at capturing the extent of technological and non-technological innovation that occurs therein.