The direct consensual method of poverty measurement developed by Mack and Lansley (1985) in their study Poor Britain has been one of the most important contributions to modern poverty research. There are nevertheless several problems with Mack and Lansley's methodology, a number of which are discussed in this paper. An alternative method for measuring direct consensual poverty is proposed and it is argued that the method improves on the approach of Mack and Lansley in the following ways: it is less sensitive to the coverage of the initial list of consumption items on which the direct consensual measurement of poverty is based; it avoids some of the arbitrary decisions made by Mack and Lansley; it increases the sensitivity between the measurement of poverty and the preferences revealed by public opinion; and, at the same time, it decreases the sensitivity to particular individuals' preferences. An empirical comparison of Mack and Lansley's original approach and the proposed alternative method is conducted using a Swedish data set from 1992. This shows that both methods, apart from some minor differences, generate a high degree of consistency in the results. One important conclusion is therefore that Mack and Lansley's approach appears to produce robust and reliable results.