The recent international financial crisis and the steady decrease in development assistance have put many poor countries under increasing pressure to target more accurately their public spendings at the poor and the population in need. However, further progress is hampered by the lack of accurate and operationally reliable methods for identifying the targeted population affected by poverty. Therefore, this paper develops low cost and fairly accurate models for improving the targeting efficiency of development policies. Using household-level survey data from Malawi, this research applies various econometric methods along with out-of-sample tests to develop operational poverty targeting models for the country. Though there is a scope for further improvements, the results show that the developed models can considerably improve the poverty outreach of development policies compared to the currently used targeting mechanisms in the country. Likewise, this research can be replicated in other developing countries.