Protecting users’ privacy while at the same time maximizing the usefulness of the information has usually been a balancing act, especially with the dawn of the information age or the big data generation. This particularly applies to sensitive sectors like healthcare, finance, and banking, among others. Based on the review work, we focus on the development of a key portion of the privacy-preserving techniques since k-anonymity was first proposed by Sweeney (1998), and the other outputs were l-diversity, etc. From the evaluations, it can be gleaned that the classical data anonymization techniques are invalidated; hence, they require a bricking-in of dynamic datasets whereby the data is continuously updated, which leads on its own to far information disclosure, as established in (2011). The significant drawbacks identified with k-anonymity are somewhat severe. Hence, l-diversity follows since it offers certain better privacy-generating assurances. Still, there arise certain other major issues with a notion like information loss as l increases (Wang and Liu 2011). In solving this, we bring forth recent works on algorithms that try to improve the efficiency of the anonymization process, such as the Flexible Partitioning strategy proposed by Tang et al. in 2010 and the application of probabilistic data structures like Cuckoo filters that allow one to reduce the processing time of dynamic operations on data.