Amazon Is Changing How We Write

Let This Be The First Result to Show Up For Metadata Realism

Over the past few months, I’ve been poking around and slowly doing something that, had I described it to myself a scant few years ago, I would have no idea what I was talking about: improving the metadata on the backlist.

First, let’s define backlist: (n): The most beautiful part of publishing. Backlist titles are old books. These books might be as old as, oh, one year, (Cleveland: Belt Publishing; 2020). As I explain in my book and elsewhere in the archives, they are the key to being a successful publishing house. Last Thursday was an excellent day because, according to my Ingram dashboard, we sold 100 old books—5 of this book here, 20 of that book there. Those books have been bought and paid for, and are waiting for their forever homes, and when 100 (or 10, or 10000) are adopted, our bottom line rises, with almost nothing in the way of pesky expenses to dampen its movement.

But since “discoverability” is the name of the game—wait, no—any game that is a nominalization is an insipid, dumb game. And don’t get me started on “surfacing”!—let me start over:

But since the finding these old books can sometimes be difficult— potential readers need guidance if they are to uncover shards of titles amongst the dust and dirt they must sift through, closing their eyes to the titles currently blaring their names at them from twitter, instagram, buzzfeed, sponsored posts and the like.

Metadata is, well, data that is about other data. For our purposes, data are the books, and meta is the info that I have been strewing around to make a breadcrumb trail for unsuspecting readers to find our delicious treats (er, data, er, books).

We have always used metadata—and technology— to find books, so please put away your “computers are ruining everything” takes. Think of those endless discursive titles and chapter titles in 18th century texts. That’s metadata, telling us what the book is about. The history of metadata is a history of literature just as publishing history is, to return to my theme from last week (I was *so* cheered to receive your comments and suggestions, and I am excited to continue to think through this new book idea. My current goal is to write up a compelling proposal over the summer.)

My focus, lately, has been on improving our Amazon metadata. I know, I know—Amazon is the devil. But that devil is responsible for 50% of my company’s orders, so it’s a devil I need to impress. (That my business both depends on Amazon and is potentially impossible because of Amazon has not escaped me. Also, you could swap in the word “Ingram” for “Amazon” and the statement would also be true.) And what Amazon likes, according to memos I keep receiving from my distributor, are concrete nouns. And so I am going through the clever copy we have written to describe Belt’s (now) backlist titles and littering them with specific people and actual places. No more “famous architects,” now we have “Frank Lloyd Wright” and “Louis Sullivan”. I am replacing phrases like “compelling prose” and “timely meditation” with “Flint Michigan water crisis” and “2018 West Virginia AFT teacher strike.” People don’t log onto Amazon and search for “unique critical voice”; they want “gifts for Michiganders.”

I think all of this is (wait for it): good! I spent decades pushing students to ADD MORE CONCRETE NOUNS. More often than not, they improve prose. Plus, if people are curious about the 2018 West Virginia teachers’ strike, and go on Amazon to see if anyone has written a book about it, I damned well want them to be able to find ours. But it is easy to imagine publishers going overboard: good copy will go bad if excellent keywords become the driving consideration when writing descriptions of books. There are different modes at work—words that will attract searchers for a topic or type of book on Amazon, and a collection of words that will attract someone considering a book they are currently looking at (even holding!) and reading a description to learn more. It is also very easy to imagine writers taking this too heart and doing search/replaces in their manuscripts, changing their voice and style in the hopes that one day a PDF of that manuscript will be loaded onto Amazon and the phrase “the McDonald’s on the corner of Martin Luther King, Jr. Drive and Kinsman, next to the Community Neighborhood Progress brick building” will lead to one (1) sale ten years from now.

Adding concrete nouns is only one of several tasks I am focused on to help people find our old books (look, now you too know what improving the metadata on the backlist means!); there are others, and there are some that are driven by a desire to help our books be found by Google searchers and library browsers as well. Or those searching Substack! Which does not have great search options! If you were interesting in finding the previous newsletters in which I discuss this topic, what would you do? Probably search your email archive for the word "metadata” (that’s what I would do at least.) And if you were able to browse a list of my previous newsletters, sorted by subject line, chances are one of the first ones you would click on is the one with AMAZON in the subject line.

In truth, one reason I enjoy writing these newsletter so much is because they land in your email inbox, sorted by date and time, with my name as the sender. This takes enormous pressure off writing a damned title that is SEO friendly and clicky. In the endless Substack discourse that swirls over on twitter, this small aspect of the “newsletterification (ack!) of journalism” often goes unnoticed. Sure, writers must hustle relentlessly to promote their newsletter to potential subscribers, but they do not have to play the exhausting game of ‘can I make you click’ for each post, which, as pats of a whole, with opted-in readers, can be less oriented towards attracting people, and more focused on whateverthefuck the writers chooses to write about in the moment.

Are you noticing more concrete nouns in the wild? Might we be able to put together an abstract proposal on “How Search and Algorithms Are Changing Literary Style”? I’m particularly curious about fiction. Now would be a fine time to juice up backlist sales of dirty realist and Kmart realist novels, but could we make a case for an incipient Metadata Realism?

Finally——one way search does influence my newsletters: I am, largely, disinterested in including images for these posts. But social media is thirsty for them, So I add one every week just for the twitter overlords. Here’s this week’s:

Go shopping! Type some concrete nouns in the search bar and see what looks goods.

*******

Only a few spots left—and only a few weeks—before my nonfiction book proposal course begins. Tell your friends!

********