Skip to content Skip to navigation

The Re-Boot: Numbers Geekery

Wednesday, May 31, 2017 - 22:00

So the question all my regular readers are wondering about (let’s pretend for a moment, ok?) is whether my month of “book release re-boot” blogging has, in fact, had an effect on sales of Mother of Souls. The simple fact is that I have no idea. (Please note: it may seem in this post as if I'm obsessing over numbers. And some may feel compelled to suggest to me that this only leads to despair. Obsessing over numbers is how I make a living in my day-job. Crunching data is one of my primary self-soothing mechanisms. Analyzing data makes me happy. Please don't harsh my mellow.)

I alternated between linking the re-boot posts to the Bella Books site and to my own website’s book listing (which includes buy links to both Bella and Amazon). But as the sales data I get from Bella is lumped into half-year increments, there will be no way to tell whether there was any bump in May. I can get a very rough estimate of sales through Amazon via their Author Central features which gives me access to Book Scan numbers for paperbacks. I can also do a rough estimate of Kindle sales by tracking sales ranking changes. Comparing the May 2017 sales in each of those categories for each of the three books to the monthly data for the last year (or in the case of Mother of Souls, the last 5 months), all sales were within one standard deviation of the mean. So: no measureable effect.

But since I’m playing with numbers at the moment, how are the books doing in comparison with each other relative to release date? Looking only at the Amazon data on a monthly basis (and combining Book Scan and the Kindle estimate), The Mystic Marriage is consistently selling about 65% of what Daughter of Mystery did at the equivalent date. And Mother of Souls is selling about 40% of what Daughter of Mystery did. Is this a good retention rate? I have no idea.

I can do a similar comparison on a half-yearly basis for the overall sales totals from my royalty statements, though the figure for Mother of Souls only represents the first two months and so isn’t comparable. (Comparing the release period data is tricky because it depends on when in the 6-month cycle the book is released. Daughter of Mystery was in month 1, The Mystic Marriage and Mother of Souls were both in month 5.)

In this data set, The Mystic Marriage is selling about 80% of what Daughter of Mystery did in the same period, but the percentage has been dropping with each period (from 87% to 68%). The initial data for Mother of Souls shows it at 20% of Daughter of Mystery, but perhaps a better comparison is between the release-period sales of MoS and MM, which reflect the same 2-months data. Here Mother of Souls is selling 130% of what The Mystic Marriage did in the same period (although final numbers may be different once returns are factored in--just comparing first-period raw numbers at this point).

There's also an unknown factor in the form of the Storybundle ebook sales of Daughter of Mystery in the second half of 2016. Once the other books get to an equivalent date, I'll have to figure out how to factor that in. The Storybundle offer came close to doubling the number of ebooks of Daughter of Mystery out in the world, but that doesn't necessarily translate to series interest, as people buying the Storybundle may not have any specific interest in my book. And there was no evidence of a corresponding sales bump (in the Amazon data) for The Mystic Marriage although there was definitely a bump in Goodreads reviews for DoM after the Storybundle. But review data is a different topic.

What this tentatively suggests is that as the series goes on, buyers are more likely to buy from non-Amazon sources. (Technically, the Book Scan numbers aren’t just Amazon, but the whole set of bookstore sales that report through Book Scan. Realistically, though, as my books aren’t carried in chain bookstores, the hard copy sales reported through Amazon are probably primarily direct Amazon sales.

What will be interesting to see is whether that 80% series retention rate holds up over time. At the rate it’s falling, the Amazon-based 65% retention rate may end up being more accurate. And keep in mind that when the first book came out, it was selling in a vacuum. Whereas books two and three are presumably selling to pre-existing readers of the series. In which case, a better metric might be comparing the sales rate to the total sales at that date for the previous book(s).

In theory, shouldn’t returning fans of the series snap the book up immediately when it comes out, with sales then falling off to a more gradual trickle of brand new readers? Well, for that to happen, those returning fans of the series need to actually hear that a new book in the series is out. And that’s where the lack of a “buzz machine” really hurts. So far, Mother of Souls has received precisely three review-like-objects outside of Amazon and Goodreads (that I’ve been able to find). What normally drives readers to new releases in their favorite series is a lot of publicity buzz in advance of the release, or at the very least a bolus of professional-level reviews right around release. Lots of review copies went out, but only the reviewers have the power to convert those into reviews.

Major category: