After a couple more weeks of advertising Star Riders on Amazon, I’ve noticed my sales dropped off even more. And after some investigation, I believe it’s because of something very non-intuitive.
Back on October 11, I improved my blurb, expecting it to convert more clicks into sales. But instead, I found sales dropped off. At first, I thought maybe I just need to wait a bit, but even after a few more days, still the sales were anemic. Even the Kindle Unlimited page reads dropped to zero.
My first thought was to change the blurb back. And then I thought maybe I should pause advertising with Amazon, and wait until I have more reviews. Then I decided to dive into the numbers. After all, if there’s one thing I know, it’s math. And Excel. OK, that’s two things, and a math guy really should know how to count.
I’ve been keeping track of the AMS statistics in a spreadsheet, so I looked into the numbers, and began analyzing the trends. What I found was that in addition to sales, my impressions had also dropped way down:
How did that happen? I didn’t decrease my bids. Did Amazon users suddenly decrease their searching? Was this some odd effect of changing my blurb? And then I recalled an odd fact from the Kindlepreneur class on AMS. The instructor had said that Amazon doesn’t just give primo impressions to whoever bids the highest. They also take into account sales.
You see, Amazon isn’t going to just give me a better chance at a click than, say, The Martian by Andy Weir. Because they know Andy will do better with the click than this weak-selling newcomer. And even through Amazon may get less from his publisher for the click, if they’re more sure he’ll make the sale (and Amazon profits from the sale, too, remember), they’ll go with their balance sheet and let Andy slide in there ahead of me.
So it seems that Amazon wasn’t impressed by my sales the first few weeks, and around the time I changed my blurb (apparently coincidentally), they decided to back a faster horse. As usual, in the game of capitalism (which I strongly endorse, by the way), the big guy wins out over the little guy. For now.
I was left wondering though, if my new blurb had made any difference, so I went back to the numbers. I still had sales, after all, just not as many, and while the numbers are too low to be statistically reliable, I think I can say the blurb was an improvement.
What I did was, I looked at how many sales I was making per click. Not the Click-Thru Rate (CTR), which is the ratio of clicks to impressions. That was fairly steady. That would be a function of the ad, to see how many of the impressions turn into clicks.
What I wanted to see were how likely I was to make a sale once they clicked on the ad, and went to my book’s page. That’s where the blurb was. So I went back to the spreadsheet and inserted a column for sales per click. And it told me that prior to the blurb change, I was selling a book about once every 17 clicks. After the blurb change, if was once every 9 clicks. That’s a pretty good improvement.
For now, I plan to stick with the AMS ads. I may not be getting as many impressions as before, but it also means they’re not charging me as much for clicks, and I do seem to be converting clicks to sales better. So I’ll stick with it for now, and continue to collect more data.
In the meantime, I really need to increase my reviews. I did contact a few book bloggers, and one agreed to read Star Riders and print a review. He told me to expect it sometime in November. His website is http://ricksreviews.blogspot.com. Hopefully, that will generate a little more exposure, from people who read his blog.
I’ve also gotten a little exposure locally. The local newspaper did a write-up (basically my words, it’s not like they read the book or anything), and I have seen that result in at least one visitor to this blog. I’m also participating in a multi-author event at my local library on November 4.
But I’m now thinking that to get more reviews, I’m going to have to try a free giveaway promotion. I gave away 300 free e-books on the launch weekend, and that probably netted me a few reviews. But there are services out there that can greatly expand that number. I’ll write about that in my next post.
Update Oct 26:
Right after posting this, my impressions seemed to take an uptick, even more than the earlier period of about 8,000/day, to over 10,000/day:
I have no idea why this is happening. I’m getting a few more clicks per day, but not in proportion to the increased number of impressions. And a few sales:
On the other hand, Kindle Unlimited has fallen off a cliff. After a fairly decent period, I’ve gone a week with zero page reads.
Well, I’ll just have to keep tracking, and see what happens.