• Welcome to the Fantasy Writing Forums. Register Now to join us!

Amazon Ads

Demesnedenoir

Myth Weaver
So, I'm coming close to now being one month 100% Amazon ads and taking stock of what is happening.

First, how I came to drop FB. For a long time, I was getting crazy high CPC on FB which made it impossible to make a profit. Mind, this is after using FB for years and doing pretty good. Being stubborn, I didn't give up and after a while I got that bad boy running good again, at least as far as CPC goes. Sales? They went friggin' SPLAT! This made it easier to finally switch my tests to 100% Amazon.

Amazon ads, for the first time ever for me, are running close to break even, which means after any normal amount of read-through, the ads will be at profit. This is a shocker because I didn't really modify the ad targeting at all. I just restarted an ad that did okay earlier, targeted at traditional authors and their books, and bang, it did better than FB ads within a few days and started heading toward break even and profit. Huh? I have no idea.

Takeaway #1: Amazon ads really do undercount sales, unless a lot of sales are coming from word of mouth. Part of that is not reporting audio and print, but even beyond that, it undercounts. That's the first time I can say that with confidence, because I don't think I ever had Amazon ads as my sole advertising platform until now. Undercounting could be the cause of assuming losing money on Amazon Ads, if I was falsely attributing sales from it to FB.

Takeaway #2: This is the frustrating part. So, I have one ad close to even and one making a profit, but sticking more money into the ads is pointless because it will not serve my ads with their lower bids. So! I can raise the bids and make it more difficult to reach profitability or expand the keywords.

#2 is the next great adventure. I hold out hope, seeing as I haven't optimized keywords at all. If I learn anything or need to vent, I'll post here, LOL.
 

ThinkerX

Myth Weaver
So, this is why I have not been seeing any 'Eve of Snows' adds on my Facebook.

After dropping a distressingly large amount into FB adds, I finally went and joined a dozen or so book groups and make ordinary posts to them...which is where I believe the majority of my recent sales came from (a grand total of two this month, unless that one promo campaign bears fruit)

My one attempt at an Amazon Ad fizzled.
 

Demesnedenoir

Myth Weaver
Yeah, for a long time FB was effective for me, then it slipped, then it crashed. It fell down and just couldn't get back up, heh heh. I might, might, kick up some FB ads December-February which is when you have lots of Christmas and post-Christmas shopping, but the best I can tell, as they eliminated more and more targeting groups... Splat! It's really annoying, seeing as FB was pretty scalable.

I have no idea why Amazon ads are working well right now because they have not in the past. Or, I was misattributing sales, but even then, I couldn't see a difference when I would end Amazon ads or restart them. So, this is a first for me.

It's also interesting to note, as this is the first time I've been KU for a long time, that the ads dashboard doesn't attribute any KENP pages to the ads, and that's simply inconceivable.

My next trick is figuring out how to find books sitting outside of "top 100's" because a lot of those would have less expensive bidding, heh heh. Books in top 100 have stupid levels of bidding on them for the most part.
 

ThinkerX

Myth Weaver
I guarantee you my books are nowhere near the top 100.

Did get one sale so far from the current promotion...maybe. Could be an outlier.
 

Demesnedenoir

Myth Weaver
Well, you only need to be in the 100k area to be top 100 in a category somewhere, and if you sneak into a tiny category 1 sale might send you pretty high from the rumors, LOL. Top 100 overall? I don't recall where EoS topped out on the overall rating, but it was #1 in Epic fantasy for a brief, brief, brief and beautiful moment with a Bookbub promotion, LMAO. I think it made the top 100 overall, but it's been a while.
 

ThinkerX

Myth Weaver
Lessee here...

Empire: Estate (Book III) is 21,586 in Dark Fantasy and 45,606 in Epic Fantasy.
Empire: Capital (Book II is 761 in Military Fantasy (Kindle) and 3243 in Military Fantasy (books), and 19718 in Dark Fantasy Horror.
Empire: Country (Book I) doesn't have a rank in anything owing to its delisting and re-release in 'skinny' form.
 
My next trick is figuring out how to find books sitting outside of "top 100's" because a lot of those would have less expensive bidding, heh heh. Books in top 100 have stupid levels of bidding on them for the most part.
Maybe through Also Boughts from the top 100 books that give good results?

Amazon ads are tricky, in that it's hard to get them to spend my money (at bids which at least potentially could make them profitable to me). I did have an ad suddenly wake up for a month or so in june, which resulted in one of my best months ever. Of course, then it died out again and my sales dropped back to their regular 1-2 per month. Of course, not being in KU makes it harder to reach break-even on ads.
 

Demesnedenoir

Myth Weaver
I do need to peruse the also boughts. I might be able scrape basic search pages. I'll have to check on that.

Maybe through Also Boughts from the top 100 books that give good results?

Amazon ads are tricky, in that it's hard to get them to spend my money (at bids which at least potentially could make them profitable to me). I did have an ad suddenly wake up for a month or so in june, which resulted in one of my best months ever. Of course, then it died out again and my sales dropped back to their regular 1-2 per month. Of course, not being in KU makes it harder to reach break-even on ads.
 

Demesnedenoir

Myth Weaver
Minor update: Sales from the ad that targets Trad authors and titles are still producing solid ebook and print sales—audio not so much—which is good because in the past I used more typical ads targeting category top 100's dominated by KDP Select books and cash sales plummeted. At the same time, KENP page reads are estimated at 30% of income and this should increase over time, and more importantly, getting an ad or two targeting the usual suspects of the indie KU world.

KU is essentially why I dropped out of wide, because my past experiences suggested that Amazon ads would need KU pages to push into profit.
 
KU is essentially why I dropped out of wide, because my past experiences suggested that Amazon ads would need KU pages to push into profit.
That's my impression as well. At least, if you want to get a decent number of views / clicks. The cost per click is such that you either need page-reads, or long series with good read-through to end up being profitable.
 

Demesnedenoir

Myth Weaver
Ah, the joys of Amazon! I currently have enough ads running shopping for bargain impressions that Amazon could spend $40 per day, but it is only spending around $10 most days and capping out just over $20 here and there, mostly on weekends. This is kind of a cheat, because I have ad sets getting a couple hundred impress ions per day because I am chasing bargains! Most of the spend is still hitting Trad authors and books.

Overall, the first month back on Amazon has been better than expected. I anticipated needing to "build" with read-through and KENP pages to get close to break even then profit, but now that things went better than expected I'm irritated at it not being better, LMAO.

A perk of having had FB ads trash the hell out is that I sold very few copies for the paster few months, so I might be able to get a better feel for read through as time goes on. My assumption is, that after several down months, there's only a smattering of read through from those sales, and they'd probably be semi-balanced by people who bought but haven't even started reading Eve yet. So, read through on books 1-2 is close to 75%, and print is doing god, but the numbers of sale of small enough to be less representative. I figure it'll take a few months to really get a feel for read-through, and it'll get complicated again by a new release in December.

Everything is a building process, a long game. I recall after going wide and dropping out of KDP Select that a year later someone finally got around to reading Eve of Snows, and then, 6 months later, another person. Then aI got an email from someone who'd downloaded it with KU but it got deleted when part way through, and they were wondering if it'd be back on KU. That was well over a year. Build build build. Long game. It's difficult to keep that in mind, heh heh.
 
Congrats. As long as you're in the black then you can keep them running pretty much indefinitely. And each sale now has the potential to make future releases better.

What's your process for finding targets to advertise against like?
 

Demesnedenoir

Myth Weaver
Out of profit by the end of the day (although more sales could be lagging behind all the clicks). Having gotten sales, the Amazon Algo started feeding the impressions like crazy and ramped up clicks big time yesterday like a weekend.

I throw shit against the wall and see what sticks. Yup, that's about right. The ad producing the most visible traction right now is based on simply finding whatever list I could of traditional authors in the (epic) fantasy genre and buklding from there with all their books starting with those authors my readers mention most: Martin, Abercrombie, Jordan, Sanderson, Tolkien, and a few others.

The second best is an old data scrape from hot new releases in Epic Fantasy (I think). Being older with some books that have fallen out of favor, it tends to run fairly cheap but doesn't get a huge # of impressions. It is a bargain shopper.

I got clobbered yesterday by my new data scrape which runs against names from top 100's in a couple of categories, and these are the KU rulers like Peppers where the ads don't come cheap, so I spent $1+ per click yesterday. Not that big of a deal for a few clicks, but when you have no idea what the ad is accomplishing? I dunno.

I would recommend scraping data even if not running ads yet, so let's say you want to run ads in 6 months. You have a pile of not just the big sellers who are expensive but a bunch of middling people and titles for cheaper targeting.

Oddly enough, I do have a couple of sales off Stephen King, not an author I would've targeted on my own. Also, a Jim Butcher series I'd never heard of, LMAO. Funny thing is, it's attribuiting a sale to that series yesterday with only 2 impressions and 0 clicks.

Congrats. As long as you're in the black then you can keep them running pretty much indefinitely. And each sale now has the potential to make future releases better.

What's your process for finding targets to advertise against like?
 
Last edited:

Demesnedenoir

Myth Weaver
Momentum? Well, at least at getting Amazon to fire impressions and get clicks! LMAO. Burnt through the $20 budget on a Thursday, which is a first. $0.50 per click, so 40 clicks. No visible sales yet. I really wish They'd let you know how many KU downloads a book gets, it would make judging success/failure so much easier.
 
Sounds like a sensible approach ;)

So far, I've found three approaches to getting targets:
- just select everything in your genre, and bid very low (35ct at most).
- yours, which is go for a relatively broad range of relevant looking books.
- dig very deep, and select only the most relevant of books.

People have had success with all three, though the first is advised against by many people as well, since Amazon likes relevance.
 

Demesnedenoir

Myth Weaver
I have my suspicions about starting with low bids in general, even though I tend to like the approach because I'm cheap. If you want the ad to connect and get Amazon's attention, relevance or whatever the term one might use, higher bids will pop the ad impressions right off the bat and you and Amazon will learn how relevant the book is quicker. On the other hand... If you start slow and save money, and Amazon discovers relevance, it will start pushing your ad more.

Oh hell, the whole thing is a mess full of overthinking, LMAO.

Sounds like a sensible approach ;)

So far, I've found three approaches to getting targets:
- just select everything in your genre, and bid very low (35ct at most).
- yours, which is go for a relatively broad range of relevant looking books.
- dig very deep, and select only the most relevant of books.

People have had success with all three, though the first is advised against by many people as well, since Amazon likes relevance.
 
Given how software is often grown, I doubt anyone still completely knows exactly how the Amazon algorithm works. Trying different things until you come across something that works is probably one of the best options.

One suggestion I came across which made sense (though I haven't tried it), is to start with high bids. This lets Amazon see how your ad is converting and how relevant it is. Once you've established that relevance, you then decrease the bids to more manageable levels.
 

Demesnedenoir

Myth Weaver
Yeah, starting high makes sense, but at the same time, it can pound your wallet hard and fast, heh heh.
 

skip.knox

toujours gai, archie
Moderator
It's absolutely true no one knows how the algorithm works. Even people who've worked there will say this. If someone claims they do know, they're scamming you.

But even if someone did know exactly, it's only part of the picture. There will always be organic sales. The algorithm is constantly being revised. So even a "correct" answer is correct only for a time.
 
Top