I don't know if there's anything to learn here but this is a recent improvement for me this past week. If there's any method to the madness, I'm not seeing it but I'll share it if it'll help anyone, and if anyone can figure out what caused the improvement, all the better.
I've been running my campaigns since August. Up until 3 weeks ago, I had 2 campaigns running on the same book. One I target books and authors that are right on genre, the other are cross-over genres. I did this primarily to not use up the 1000 KW spaces, and to have one campaign that is more laser focus. The Main On target campaign, I set spend limit at $6, and the other one $2. 3 weeks ago I started a third campaign with $1 spend limit. My goal was to move all the KWs from the Main Campaign for books that rank in in pit bottoms to what I'll call the Secondary Genre Targetted campaign. So I got this:
1. Main Genre Targetted Campaign -- Books that sell well, rank at least below 100K at $6 spend limit
2. X-genre Campaign -- not directly on genre but good possibility of interesting some readers, at $2 spend limit
3. Secondary Genre Targetted Campaign -- Same genre books that rank 200+K to Millions, at $1 spend limit.
For the Secondary Campaign, I figured that I could get my book visibility in case of a strand reader here and there, but it wouldn't cost me anything because the click rates would be low, due to low traffic to those pages.
Things went well for months but for inexplicable reason, sales began to decline. I periodically tweak KWs, upping or lowering prices depending on the Sponsored Ads situation on the KWs books; also adding new KWs. For some reason, sales were still kind of falling. Not dire, but falling.
Then last Thursday, I ran a 99c ENT promo which jumped my sales. During the ENT promo, I paused all the ads. When the ENT promo was over, I only restarted the Main Genre Targetted Campaign, but this time upping the spend limit to $10.
Why I did this? I have an author friend in the exact same genre. Our results should be very similar, but I noticed that she was able to keep her sales high. I know she put in a higher spend limit than I do (I'm pretty sure it's $10), so I wanted to experiment a bit. She said all she does are AMS ads and tweet, and I take her word for it.
Since then, I've been able to maintain the sales in the low 5 digits. (Ok that might be considered failure to some of you but for my genre and what I do, that's pretty good.) In any case, the point is, the changes of upping the spend limit and runing only one campaign appear to be helping. At first I attributed it to being an ENT tail, but I'm not so sure about that. ENT was great, but my last few promos didn't give me a sticky tail like this. I think what ENT did was that it sparked the algorithm, and from there on, with the combination of the changes in my campaign, sales for now are sticking. In fact, slightly higher than my best AMS ads performance before all this happens.
I feel a bit worried that I left all the KWS from the Secondary Campaign and X-genre Campaign un-used. But things are working for now so why fix what's not broken.
I don't know what to make of this but thought I'd share anyway. I just listened to Mark Dawson's podcast on AMS ads. He's recommending the "brute force" option, which is to keep the spend limit low ($1) on each campaign and run many campaigns. His reason for it is that Amazon refuses to spend all your money when you set a high spend limit (and that is true). But I think I'd have to disagree with him there. My own observation is that Amazon only show your ad to a small audience when your spend limit is $1. And this is something I observed even back in Sept, before I upped the spend limit to $6. I have hunch that Amazon show your ad to more pages when you have a higher spend limit, and the best thing is, they won't use up your spend limit anyway so you don't actually spend $10/day. That said, if your budget doesn't allow a potential $10/day, it can be iffy and scary.
My conclusion at this point: it's all voodoo.