Accurate perception of socioeconomic status and timely identification of emergencies are critical to smart social governance, however, traditional public sector data and statistical analysis methods cannot meet the accuracy and real-time requirements. Recently, large-scale data accumulated by the private sector, with many advantages including low acquisition cost, real-time updates and high spatio-temporal resolution, provide new directions for tackling the problem. This paper overviews the application of carrier data in combination with deep mining analysis algorithms on precise poverty alleviation and emergency management, and further discusses some prospects of applying carrier data to quantitatively evaluate the effect of poverty alleviation and disaster relief and to improve the decision-making efficiency and governance capability.