The field of smart meter data analytic is a relatively young field that grew because of the wealth of data generated from the use of smart meters. This data has been used for several sustainability applications including power management and planning. This review paper aims at presenting a unifying taxonomy to classify the various domains of smart meter analytic and their underlying functions and techniques. The aim is to better understand the current research trends, approaches, and opportunities. The paper reviews the functions, applications, and techniques after examining a huge body of knowledge within the three main domains of the field, namely forecasting, knowledge discovery, and power management. In the forecasting domain, the functions are classified based on the scale and horizon. Other domains are divided into relevant functions, algorithms, and general techniques. The review is performed from the perspective of data science; emphasizing the data tasks such as data wrangling, analytics algorithms, and evaluation methods for each domain. A review of the various algorithms and techniques within each function is presented, and the paper concludes with a discussion of issues and research opportunities for smart meter data within the machine learning and data science fields