This paper attempts to provide an understanding to what extent institutional change of the functioning of microfinance influences some specific outcomes of interventions like landholdings, health and income of the ultra poor in Bangladesh. It examines impact of the „Challenging the Frontiers of Poverty Reduction-Targeting the Ultra Poor? (CFPR-TUP) program on those outcomes using a large balanced panel dataset (2002, 2005 & 2008) from rural Bangladesh. The main objective of this paper is to examine short- and long-term impact on landholdings, health, income and distribution of income of the ultra poor using conditional and unconditional difference-in-differences (DID). In addition, fixed effects estimation is also applied to check the robustness of our estimates on certain outcomes like total landholdings, access to land, land-man ratio, chronic food deficit, income and its distribution. In particular, this paper examines the distinction between treatment and control group in base year and the impact on coping ability. This study uses two different units for measuring the impact on landholdings: proportion of households and decimal. This study (based on proportion) finds the robustness of impact estimates in the long-term for homestead, cultivable and total landholdings using both DID approaches, which also find short- and long-term robust impact for cultivation of others? land, access to land and leasing. After using decimal as a unit of measurement, we find the long-term robust impact on homestead, total land owned, access to land and land-man ratio. Here we do not find robust impact on cultivable land and cultivation of others? land both in short- and long-term.