OLAP enables users to access information from multidimensional data warehouses almost instantly, to view information in any way they like, and to cleanly specify and carry out sophisticated calculations. Although many commercial OLAP tools and products are now available, OLAP is still a difficult and complex technology to master.
My review of this book is the story of 3 books really. As a reminder, this book is several years old at this point, and where relevant, I will mention that in this review.
The first 11 chapters, which are primarily theoretical, are still as relevant and useful today as the were when written, although some decision logic has likely changed where memory is discussed. In fact, I was really looking for a book that would help me build and design my own system (as a way to replace the no-longer-supported Python package cubes) and even though this book is pointed towards users of professional systems (Essbase/TM1/SQL Server Analytical Services), the book actually covers several topics in a way that allowed me to make meaningful progress on my program.
Chapters 12-18, which are the case study, are just poorly written. The author regularly tries to include flaws in logic by exploring them across several pages just to ultimately say they're wrong. That probably is a good format for a live class but is just miserable in a book. I guess if you have nothing better to learn from, they'd be OK. If you simply think of the context of your data (as described by dimensions) you will gain 80% of what is discussed in this part.
OLAP is still a relevant topic (Essbase is the #1 Financial Planning and Analysis tool; new players like Vena are trying to improve upon it), but it is worth noting the author now heads an AIML company - if you need to learn OLAP software, you'll probably have to get training from whatever vendor your firm uses.