For poor women paid work is not simply a pathway out of poverty, but has more deeper transformative potential, including both internal transformation (changes in women's personal and political consciousness and agency as citizens) and external transformation (changes in women's social position). Hence, measurement of women's empowerment requires identifying appropriate qualitative indicators to capture these dynamic processes of change that are not all observable. We were faced with two crucial measurement challenges: first, to estimate the magnitude and nature of women's paid work that is often unrecognised, and second, to assess a transformative process like women's empowerment. The paper describes the methods used for enumerating women's economic activity and measuring women's empowerment in the context of Bangladesh, using quantitative indicators estimated from a large household survey. Our experience suggests that to capture women's work adequately interviewer bias due to social perception of work can be minimised through training and capacity building and careful wording of questions that have been pretested. In addition, to gain a more nuanced understanding of the impact of paid work on the process of women's empowerment we included several new indicators for assessing various dimensions of empowerment that have been examined widely in the literature.