In their seminal book, Bartolini and Mair (1990: 44–5) proposed the cleavage salience index (CS) to capture ‘the salience of the cleavage, that is, its importance and weight within the general context of the electoral behaviour of a given country and/or period’. In this article, we demonstrate theoretically and empirically that the CS index cannot be used to measure and compare the strength of cleavages over time and for different cleavages. The reason is fairly straightforward; the index does not take into account the actual, absolute level of electoral support for the cleavage block parties, which influences the potential range of the index values. As a better alternative to the original CS index, we propose the block-weighted cleavage salience index (WCS) and provide empirical data about class and religious cleavages in 11 European multiparty systems from 1950 to 2010.