While the title of this post may seem self serving, it is so only by way of introducing a dilemma common to both readers and authors. It is the opposite ends of the same stick. Readers want to know how to find the kind of book that will particularly engage them, and authors want to find those readers particularly engaged by their books. With so many writers and readers in the world today, once we sort this out everyone should be happy. Right?

I am both author and reader and I am familiar with both ends of the aforementioned stick. Between its ends frustration lies. Amazon happily posts sales data for KDP authors––the numbers for books sold, books refunded, net books sold and, for KDP Select authors, the number of free pages read. Each number in the refunded column indicates someone who grabbed the stick by the middle. Whose fault is that? Did the author categorize the book incorrectly, or did the reader misinterpret the category? Or both?

The categorization process is more complicated than it may at first seem. Not just on Amazon, not just for on-line books, but anywhere a retailer shelves and sells. Classifying books, particularly fiction books, is akin to classifying brains. There is a wide breadth of focus, perspective, experience, and presumption brought to bear for each categorizing label. Therefore the retailer utilizing a category system must determine if a crime mystery is more a police procedural, or an amateur detective story, more of a thriller or horror story, more of a romance with murder or murder with a romance? Such categorizing is seldom precise.

Amazon provides the categories authors must use. While the selection seems limiting, I’m sure it was well researched. Regardless, and necessarily, the categories are too broad to truly define content. They can only narrow the field by indicating a direction. This leaves a lot of room for a wrong choice.

From an authors point of view, it may seem advantageous to solve this problem by entering a category in which the work will have the most visibility, such as a category that contains fewer books, whether or not that category best defines the author’s book. How much choice is offered? I could legitimately list any book in my series under the following Amazon suggested categories: action & adventure, crime, legends and mythology, fantasy, paranormal, ghost, historical, horror, literary, mystery & detective, police procedural, romance, suspense, western, thrillers, occult & supernatural or non-Classifiable. Unfortunately for me, I may select only two of these per book. The problem is no two of these categories fully express the content of my book.

Visibility is the Holy Grail of the author, however it can be achieved. It is therefore very tempting to select one or two of those seventeen categories which is least utilized. The goal for the author is to be listed within the Top 100 in any category, and hitting #1 is like setting off the siren in a casino slot machine. Why? Those books are most visible on the Amazon book site. Books beyond the first 100 in any category are virtually invisible. 

However, such tactics do not aid the reader. In fact, how does a reader chose the next book? Smashword researchers recently posted poll results for this question (I am interpreting the poll results from the perspective of fiction). The following poll results are from Coker’s 2011 poll––since then poll questions and analytics have become far more complex and detailed (and less useful when drawn out of context). But the results listed below have generally remained the same. 

So…the poll results for how readers select books:

1. Recommendations from fellow readers (but not so much from personal friends/family??)

2. My favorite author

3. Random browsing: (i.e. book covers, reviews, free ebooks, browsing paperbacks in bookstore then looking for it on line, ebook sampling, etc.)

From my experience as a heavy reader tactics one and two are soon exhausted. When a favorite author writes a new book or a reader who knows my taste makes a new discovery, it is Christmas in July. But until then, I must try other methods. Tactic number three is a hodgepodge, uncertain at best, involving much trial and error.

This is when a universal category/keyword system designed and maintained by an independent source would be useful. To be most effective, the categories and keywords should be selected not by authors but by readers, sort of a Wiki-label kind of thing. But that system does not exist.As I think about it, if I crunched all the reviews for each of my books on Amazon, somehow processing them into a ticker tape of single word definitions, a Wiki category definition is probably what I would have.

Anyone out there want to launch that project?