The big data concept started to form in the early 2000s; since then, organizations have been challenged with increasing, even alarming amounts of various types of data, generated by multiple sources such as social media exchanges, digital processes, or sensors. An early definition from the analyst Doug Laney characterizes big data by the three Vs: volume, velocity, and variety.

The book presents a new viewpoint to approach the big data concept that tries to distance itself from previous approaches. It introduces a consistent framework to handle and apply data strategies. Starting with the very definition of big data, the author stresses their role in innovation, in the discovery of new and useful structures and correlations. Furthermore, the material emphasizes the need to move the processing and data analytics steps to a lower level, in order to extract meaningful insights for the targeted domain, and thus to reduce data transfer costs. So big data and data science concepts are interlinked and used alternatively throughout the book. Based on some usual misjudgments associated with big data, the author elaborates a novel data deployment approach. The material identifies some basic judgment mistakes that usually affect big data interpretation and impact. One of them is the false impression that a higher bulk of data ensures a higher precision, the counterpart of the “curse of dimensionality” in data classification. A second judgment flaw is not to relate data to the actual context of their user, or not to correctly specify the real problem to be solved and take for granted whatever solutions are found. As some recent big data researchers have already pointed out, the book accentuates the importance of the correct problem formulation; of the identification of appropriate variables, statistical methods, and criteria; and of the accurate assessment of the analytical methods. “It is not the amount of data that matters. It’s what organizations do with the data.” The book condenses the most important issues about big data, specifically addressing companies and organizations.

The second chapter, the core of the book, presents a scheme for data management, meant to integrate data coming from several sources, which reports to the “golden record” (“a unique, well-defined version of all the data entities in an organizational ecosystem”). A lean scheme for data deployment is proposed that works in feedback loops, consisting of a hierarchy of five stages, where the central stage is data acquisition that should receive feedback from the analytical framework component and the business and go along with modeling and optimization steps.

The data stage development structure is the proposed model to implement a revenue-generating data strategy, intended to help assess the current position of a company and its big data capabilities. The model describes four different levels of growth for the data development structure (starting with the primitive one, the bespoke, factory, and scientific levels) and reflected in four aspects: cultural, data, technology, and talent. The cultural facet involves leadership support. The last part of the chapter leaves the theoretical and descriptive layers and discusses more practical facets related to its implementation versus company profile. This model also highlights the fact that big data is a team affair, involving IT leaders, managers, data scientists, and strategists.

The rest of the book acknowledges some important facts related to big data.

The third chapter discusses the main contemporary challenges to big data systems, among which the most important are data security and the ethics behind sharing data with some degree of privacy. Some possible threats are analyzed, and possible solutions are sketched. Some other provocations involving big data management issues are mentioned, for instance initial public offerings (IPO), growth strategy, emerging markets, and data ecosystems.

Another question discussed is an emerging position in the big data context: the data scientist. This role is analyzed, as well as the way it relates to the other job roles involved in big data handling: the businessman, domain expert, statistician, and computer scientist. The possible data scientist personality classification is based on a survey from 2013, fully presented in the appended materials.

The future trends include the Internet of Things (IoT) and cloud technologies. Finally, the role of application programming interfaces is highlighted as a means to create applications to access, secure, and use big data.

The appendix contains some dictionaries of some technological terms; the definitions are very short and probably considered sufficient from a manager’s point of view.

The book would be interesting reading for big data analysts or company managers because it discloses some important facets of this new technological challenge, although some subjects are quite hastily explained.