Abstract This paper presents an application of a recently developed method for measuring innovative activity in a national economy based on new product announcements in trade and technical journals. The 'literature-based innovation output indicator' is not meant to capture all aspects of innovation, or to substitute for other indicators, but is seen as a useful addition to the range of indicators available. The method is described in detail, and then applied to the UK. A sample of 941 innovations is constructed, and then subjected to analysis. The results suggest that the method is a useful and relatively reliable way of measuring the degree of ‘radicalness’ of innovations being generated, and of surveying their distribution across sectors, across firm size, and across firms which are UK-owned or foreign-owned. The method captures product innovations well, but captures process innovations less well. The results for the UK suggest that the more radical innovations come disproportionate from companies based outside the UK. Innovations originating inside the UK are skewed towards product differentiation. There is also some evidence that smaller companies produce a slightly higher than average share of the more radical innovations. The paper suggests that this small-scale testing of this indicator has produced promising results, and that there would be benefits to the policy community from establishing it on a permanent basis.