This book aims to minimize barriers to data collection and analysis for smaller businesses, both those that may not have the resources for a dedicated data science department as well as those that would benefit from the services of a large analytics firm. Here, small and medium businesses are defined as having between 50 and 200 employees.

The content and examples are written for subject matter experts without prior experience in the data collection and analysis life cycle. The book focuses on concrete business analytics services, foregoing an in-depth explanation of more complex topics like data governance, the business analytics project life cycle, or risk analysis. These topics are mentioned, but only with enough detail to understand their context in the complex landscape of modern data science.

On the other hand, there is a secondary benefit to this book: it contains a good high-level overview of advanced analysis and predictive techniques from deep learning, which will help the reader to create a mental map of what current artificial intelligence (AI) techniques are useful in which business cases, ranging from customer analysis and promotion evaluation to sales prediction. Having a systematic, holistic description of business data analytics may well fill in the gaps for those whose knowledge depends on media articles or data science blogs.

The sample use cases are illustrated with RapidMiner and Gephi, two open-source tools for data mining and visualization, although it is also possible to use others (for example, ChartBrew or Lightdash). A complete e-commerce case study is provided as a workshop at the end, and it makes sense to go through it in parallel with the chapters.

Readers can expect to learn the basics of the life cycle of a descriptive data science project, starting from data capture, preparation, and curation. Several examples cover how to manually clean, complete, and normalize input data, as well as how to plot data into charts, graphs, maps, box plots, and so on, in RapidMiner. The summarization and comparison of results are self-explanatory. More advanced descriptive techniques like clustering are discussed in chapter 8.

Predictive data science--the most valuable area of business data science--is also discussed in detail. The most common models, including linear regression and predictive neural networks, are mentioned in relation to future business sales; examples in RapidMiner are also provided.

In short, by reading the book and working out the use case, subject matter experts will be able to get a coherent roadmap to the main techniques available for both descriptive and predictive data analytics, as well as be able to provide simple services related to their company data and future prospects.