Choose a subject area that:
If you are lucky they will be all three at once!
1. Allow enough lead time to fully absorb the book over time.
When you first open the book, jot down some initial thoughts. This is important. You won't see the book the same way again as you did when you first cracked it open. Set the book aside and come back to it later to make additional assessments.
Read a little per day. Reading (or scanning) the book in one sitting is not recommended.
2. Keep your eyes open for similar books during this period.
Scan your library catalog for books of the same subject heading, published within the last few years, and even reaching back over the last decade. Consider older books only if they appear to be landmark books or were reviewed in the same review source. You may want to familiarize yourself with them in case you need to refer to them as benchmarks (or they may have newer editions on the horizon). Allow enough time to obtain books from other libraries to assist in your review, if necessary.
Most review sources will ask you to make critical evaluations about the following aspects of a book you are reviewing:
Authority What are the credentials of the author or editor? Is he or she prominent in their field? What else have they published? How do they compare with competing authors in their field? Is the book put out by a leading publisher in the discipline?
Scope Does the book focus on a particular aspect of a topic? Or is it a broad survey book? Is it in a unique enough niche to consider worthwhile but not so narrow in scope that very few would find a need for this book? Has it all been written about before in a hundred different ways?
Currency Is this book representative of the latest news or research in the field? Is it up-to-date? Or has the moment for the topic passed? Is it just an update of a previous edition? What is the nature of the revision? Or is there barely a difference from the last revision?
Relevancy Is this book relevant to a particular type of library collection? Is it unique? Or is it just one of hundreds of books on this topic? Is it appropriate for academic/public/school libraries? How many other libraries own the book? Is it indispensible, a book every library should own? Or would it be considered an optional addition to the collection?
Audience Who is the expected audience? Is it written for highly specialized practitioners or for general audiences? Is the book written appropriately for the intended audience?
Quality & Value Is the book physically solid? Will it hold up in a library collection? Is it well designed? Is it visually appealing? Are the illustrations necessary and of effective quality for the subject matter? Is this book designed for the amount of use it is expected to get? Is it too expensive for most to consider purchasing? Is it overpriced? Are there alternative new books in the same subject category or covering the same topic that might be preferred because they are cheaper?
Just jot notes at first, book in hand. Read the summary on the back to solidify the basic factual parameters of the work (e.g., number of annotations inside, exact number of decades spanned, number of artists interviewed, number of images, whatever the case may be...).
Read the introduction, preface, foreword. Get a feel for what the book is supposed to accomplish and where the author is coming from.
If a reference book, ask yourself some questions as if you are a researcher. Look up some hypothetical entries that matter to you. See if you can find your topic in the index (is an index provided?). How quickly can you locate what you are looking for? Does the layout make sense? Did you NOT find what you expected to be able to find?
If illustrated, what is the quality of the illustrations? Is color provided when necessary?
Nearly every book I have ever read has had at least one or two typos. How many typos do you see?
Write a first draft considering all these technical aspects and incorporating knowledge about existing books on the same topic. Be sure to indicate if any of these other books has been covered by the same review source and cite this. Be sure to also include a cursory introductory background about the topic in your review if allowed.
Let the review draft sit a few days and then re-read it. Revise it. Revise it again. Revise it until it is as succinct as possible. Cut out redundancies. Read it out loud so that that you can be sure the text flows well. Conclude your review with a recommendation on the book, offering alternatives when deemed necessary.
Work well within word limits. Restating bibliographic information in the text of a review is almost always a waste of valuable space, as is restating the basic chapter sequence, table of contents list, and so on. Limit your remarks to those that compare the authority and credentials of the book with other leading books in the field, stating the overall value and usefulness of the book in comparison with similar works and noting any deficiencies or extra features.