Saturday, May 12, 2012

Day 1 Post-Free Results

Well, the first 24 hours coming off free teaches us one thing: It's going to take more time to draw any real conclusions. (See yesterday's post, if you missed it, to get caught up on today's sales discussion.)

Let's get right to the data. Below the chart, we'll discuss what you're seeing. (Click on the chart to see it larger.)



One of the interesting things I noted after coming off free was that the majority of these sales came within the first 8 hours. Sales have tapered off since.

The titles are in the same order as they are in the chart we saw yesterday, which showed best free rankings and number of downloads. If the book showed up in a category's Top 100 bestsellers list, I've noted its rank in that list in a column on the left.

On the right side of the chart are sales by region: US, UK and all the others lumped together as sales are minimal outside of the US and UK. There's a column for total sales across all the regions, a column for total number of borrows, and a column showing total returns/refunds across all regions. It's pretty clear that coming off free in the middle of the day confused a number of customers who thought the books were still free. A few obviously went on to borrow the book after returning a purchased copy. And some of those returns could well be freebies that were returned. How Amazon is counting them I won't know until tomorrow's weekly summary report.

The far right column shows sales in the US for the week prior to the free run. Most of our books were in a bit of a sales lull and sales immediately post-free in most cases surpassed each book's prior week's total. Vet Tech Tales is the only outlier here; historically it seems to sell consistently whether it's been on a free run recently or not.

The column with the popularity rank is one we will want to pay particular attention to. The recent algorithm changes mainly affected these lists, and books coming off successful free runs are not rising as high on the lists as they did before May 1 (when 3 different display lists became 1) or March 19 (when the single popularity list that was Select-friendly was split into 3 different display lists that were randomly tested on customers).

It's clear that the Science Fiction and Thriller categories have updated since yesterday. SECTOR C was further down the ranks in both the SF and Thriller related categories on Friday. Its ranks on the new popularity lists track well with its ranks on the bestseller lists. The only discrepancy is with the Medical Thriller list. That list, however, is a Books list and not a Kindle list, and I'm pretty sure that one simply hasn't updated yet.

In the past, given the rank and number of downloads SECTOR C achieved, it would have hit the popularity list at  #1 in both Technothriller and High Tech SF and possibly hit at #1 or #2 in overall Science Fiction and wound up for a third time on the Kindle store homepage as the representative title for the SF category. The overall Thriller category is much more competitive, and SECTOR C would likely have landed somewhere between #22 and #30 on the Thriller popularity list and been in the Top 100 for the broader Mystery & Thriller category as well.

I'll be particularly interested to see whether the other books track as closely to their bestseller ranks. SECTOR C has the advantage of bringing last month's sales during its Top 100 run to the algorithm table today since it appears the algos weight sales heaviest over the last 30 days.

Likewise, where our books have Top 100 bestseller ranks in categories where they're not showing up in the first 250 items on the popularity lists, I'm guessing it's because the daily updates haven't taken place yet. In the past, we've seen it take anywhere from 24 to 48 hours for these lists to update, contributing to what was affectionately known as the 3-day sales bump.

In addition, considering the discrepancy (though not so wide) between Vet Tech Tales' performance in the two lists, it looks like its popularity lists haven't updated yet either. It was indeed selling well enough before going free that it hadn't dropped completely off the radar on those lists.

So is there anything yet that we can tell from the data here? Not so much. There have been several folk who've run a title free since May 1 when the algorithms changed and who've reported only a handful of sales over the 3 or 4 days following the free run. Here, we see good sales for 5 of our Steel Magnolia Press books post-free, moderate sales for 3 of them, and meh sales for 2, so sales immediately after a free run seem variable across the board. What we need to do now is wait to see where the books land on the popularity lists once those lists recrunch and then see if their new ranks help drive additional sales.

The problem? Will sales that are tailing off now because the books aren't currently appearing high on a list affect where they ultimately wind up on that list? We'll take a look at that together first thing on Monday morning. Tomorrow, we have a query to critique!

No comments: