Isn't this a question that contains its answer? A book review is a review of a book. True enough, but what does a review consist of?
A book review is not a book report. A book report describes what happens in a book. A book review includes some of this, but is also evaluative--what the reviewer thinks about the book and why.
A good review is more than a simple thumbs-up or thumbs-down. Rather, it presents
(1) the reviewer's opinion about the book's content;
(2) an argument about the book's significance, accuracy, and quality (or lack thereof);
(3) evidence drawn from the text itself or from outside research. Such evidence is crucial in supporting the reviewer's claims, particularly when a review is unfavorable or controversial. In practical terms, this means substantive book reviews resemble other kinds of academic writing, in both form and content.
What and how book reviewers evaluate and to what depth varies by publication. Like other kinds of writing, book reviews are written for different audiences. Generally speaking, a review on Amazon has a different purpose than one from Publisher's Weekly or Kirkus, just as the latter are different from the lengthy academic reviews published in scholarly journals.
If most book reviews contain some synopsis and opinion, what else goes into constructing an argument and gathering evidence? A reviewer may consider audience, accuracy/quality, purpose and context:
If you have a book review assignment or presentation, follow your professor's instructions, of course. Beyond that, here are a few general approaches to creating book reviews: