Classified service analyzes your book to pick out the best categories

One of the most important aspects of ebook metadata is assigning BISAC codes, also known as picking categories for your book, and a new company is offering a free classification service which takes just seconds to assign a category to your book.

You can upload your book file to the Classification Demo at Kadaxis as an epub or doc file and have the BISAC code predicted, along with other derived metadata, such as locations mentioned in the book, character names and summaries of various word counts.

Major publishing houses use experts to assign BISAC codes for maximum marketability. The codes have an impact on placement on shelves in bookstores, online placement and discoverability in search engines.

It’s also, of course, very annoying when you’ve spent some time picking out categories for your books and then having them reassigned, seemingly on a whim, by Amazon, but that’s another matter, I suppose.

The assignment of codes is a crucial element of book publication, but as there are over 3,000 codes, it’s challenging for self-publishers or any non-expert to understand the best category for a book.

New York-based Kadaxis says BISAC Classifier has learned this expertise from publicly available data and can predict the most appropriate BISAC code, providing self-published and unsigned authors with access to the same expertise as industry pros.

It claims to extract keywords, phrases and other metadata from the text of a book, and selects and ranks this data based on trends and relevance. As different trends emerge, new keywords highlighting different elements of a book will attract more traffic and Kadaxis treats a book’s description and metadata as dynamic to help publishers attract a bigger audience over time.

The company says it will provide the access to the same technology to indie authors and you can sign up for the Author Checkpoint free beta at Kadaxis.

The company is also providing another service which looks like a worrying development. Slush Filter is claimed to be able to read a 200,000-word document in seconds and provide a marketability recommendation and metadata such as preliminary BISAC codes (for book categories), comp titles, locations and character names.

Kadaxis says Slush Filter is designed to help save editors and agents time on reviewing unsolicited manuscripts. I’d love to run some proven bestsellers and award-winners through the program to see what it comes up with. I wonder what it would make of the Joycean A Girl Is A Half-Formed Thing by Eimear McBride, which won the £30,000 Bailys Women’s Prize for Fiction this week.