Market Research on a Shoestring

So, I sat down to look for an editor today, and decided that instead of that, I had to give my book a release date. Trust me when I say that the path from point A to point B made perfect sense when I was walking it.

See, one thing I don't want to do is just release my book at any old time. For one, a little time to build a good buzz might be worthwhile. For another, not all release dates are created equal. While doing the review blog (which I will get back to at some point, I promise!) I noticed that there were peaks and troughs: some months I had multiple releases to contend with, others I had time to go back and look at stuff I missed. As a new author, I should probably aim for a trough, when competition will be lighter. If I hit shelves during a peak, I'm liable to wind up lost in the shuffle. I ought to think like a real wolf: pick the prey which is most easily available.

The problem is, how do I figure out where the peaks and troughs are? I'm sure there are numbers on this somewhere, but I've got no clue how to look for them.

So, I decide to get the numbers for myself.

I head on over to Amazon. They may be the bane of brick-and-mortar stores the world over, but they have a huge selection of books, which means plenty of data. After some messing around with the Advanced Search, I figure out how to search by publication date. So, I conduct my own ad-hoc, one-afternoon statistical study. First I do an Advanced Search: keyword "werewolf" (not exactly a genre, but close enough), publication date: Dec. 2011 or earlier. I get 4,293 books, which I ignore. The important thing is the number: 4,293. I record it in a spreadsheet, then punch the back button and do a new search, changing the date criteria from Dec 2011 or earlier to Nov. 2011 or earlier. A simple subtraction gives me the number of werewolf-related books released in December 2011. Go on to the next month, do the same thing, and keep going back to January of 2009.

It's all very duct-tape-and-chewing-gum, but at the end of it all I have a graph that I'm fairly confident in:

Cute, huh? ^_^ Well, aside from not fitting the blog at full resolution...

Anyway, looking at this, (click to enlarge, naturally), we can see a few different things: One is a massive spike in releases in 2011, to the extent that the lowest month of that year almost equals the highest month of the previous year. This makes sense; after all, the Kindle was one of Christmas 2010's most-gifted items, and there's a lot of authors clamboring to fill that market. Even before that, there was a steady increase between '09 and '10, thus indicating that there is still a perceived market for werewolf literature, despite naysayers calling it a fad and haters bashing paranormal for all it's worth.

But this isn't the information I need. What I wanted was trends month-to-month. And some trends do become apparent from this graph. There's a spike in January, followed by a sharp dropoff in February, one of the lowest months of the year. The spring months are middling, with the possibility of a dropoff in May or June. Summer is a busy time, and fall tapers off, with November and December being low months. That spike in December 2011 is something I'm at a loss to explain. Could be an anamoly, a flaw with my statistical methods, or an attempt to capitalize on the Christmas rush.

In any event, it seems the low points are February, May, and November. February is the lowest point, but there's no way I could have the book ready by next month. May, however, looks very promising. Not only is it a trough, but it's a trough leading into the big summer rush, when people have more free time to look for new and interesting books. A strong strategy for me, therefore, is to spend March and April building up buzz, release in May, and then ride the wave over the summer.

Hey, look at that, I can do statistics. ^_^ Truly, an entrepreneur makes friends with many skills.

So then, Bonds of Fenris now has a semi-official release date: May 2012. And holy bajesus, I've got a ton of work to do before then. TTYL, folks!