Amazon Ads

[pmmg]

Sounds like a confusopoly.

 

[Demesnedenoir]

I think the official term is a "hot mess" heh heh. Or maybe FUBAR. At least for books. One big issue, as spoken of by someone who was once involved in the guts of the Amazon world, is that the system is not designed to be optimized for book sales, despite Amazon's origins.

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.

 

[skip.knox]

Sounds like a confusopoly.

I don't think so. Rather, I'd say human beings are complex and inconsistent creatures whose arbitrary and even contradictory buying habits Amazon is trying to track reliably. It's a Labor of Hercules to be sure, one that makes mapping the human genome look like a school exercise.

 

[Demesnedenoir]

Just a weird little update... For November, paperbacks are out earning ebooks. This isn't a total apples to apples situation as EoS make more money on PB than E, but on the other hand, E makes more profit on books 2 & 3. But no matter how I slice it, PB sales have been strong since going straight Amazon ads, but they don't show as sales in the ads dashboard, just to make things more confusing.

Double checked, ebook and paperback are tied for actual sales, not including the mystery of KDP Select downloads. On the one hand, lower in ebook sales than normal (but normal to my brain was before people could download on KU) and on the other hand, high for paperbacks.

 

[ThinkerX]

Just a weird little update... For November, paperbacks are out earning ebooks. This isn't a total apples to apples situation as EoS make more money on PB than E, but on the other hand, E makes more profit on books 2 & 3. But no matter how I slice it, PB sales have been strong since going straight Amazon ads, but they don't show as sales in the ads dashboard, just to make things more confusing.

Double checked, ebook and paperback are tied for actual sales, not including the mystery of KDP Select downloads. On the one hand, lower in ebook sales than normal (but normal to my brain was before people could download on KU) and on the other hand, high for paperbacks.

That has been the case with me since the start. Right up until the $0.99 Books Butterfly promo, the paperback sales have been outselling eBooks by 2 to 1. Best as I can tell, ALL of the sales for book 3 - 'Empire: Estate' have been for print books. (Granted, 'Estate' has sold a mere dozen copies, but still...)

 

[Demesnedenoir]

Well, a slow month when I had things going was 100 ebooks per month, with months in the hundreds. Now, it's more like 50 after the FB explosion that murdered my sales. If I came close to those numbers in print, I'd be making money, LOL. Today is a bit of a unicorn, 5 paperbacks have shipped today. I have no idea why.

 

[ThinkerX]

I am currently at fourteen sales for November, making this my second-best month ever. I figure with luck, and the release of 'Empire: Metropolis,' I might hit thirty sales for this month.

 

[Prince of Spires]

Has either of you ever tried Twitter ads? I'm wondering with the current Twitter implosion and advertisers fleeing the platform if maybe it's a cheap place to get clicks. Just don't have an idea on where to start (having never actually used twitter...).

 

[ThinkerX]

Has either of you ever tried Twitter ads? I'm wondering with the current Twitter implosion and advertisers fleeing the platform if maybe it's a cheap place to get clicks. Just don't have an idea on where to start (having never actually used twitter...).

A couple of the promo outfits I signed up with say they promote on Twitter.

 

[Demesnedenoir]

Twitter needs to shake out, I suspect, but I have had the same thought. My impression of Twitter has been that it was like FB only with waaayyyy more bots. So, I never bothered to advertise there. From what I gathered from many folks, Twitter wasn't that good for book ads. Not saying it can't work, but several writers I've spoken to gave up after a short time and stuck to Fb and Amazon. Privacy issues are killing FB's targeting so I'd suspect Twitter to be similar. But that's a guess.

 

[Prince of Spires]

I think due to the type of content people post there that Twitter doesn't know nearly as much as Facebook about people's interest. Which makes it a lot harder to do the very specific type of targeting people say is needed to make Facebook and Amazon ads a succes. So I can definitely see both click-through rate and conversion rate being lower.

The question then becomes, what is the cost-per-click like. If it's similar to FB or AMS, then it's not worth it. However, if it's much lower, then it may very well be more profitable than any of the other two. And given the current chaos on Twitter there might very well be some bargain cost-per-clicks to be had. Maybe I'll check it out. See if I can find any information on advertising on Twitter. If I do I'll let you guys know how it went.

 

[Demesnedenoir]

Minor update: Print sales are satisfactory purely on Amazon ads, with digital to print being about 60/40, which is a big change from the FB days of 90%+ digital sales. One bugaboo now is reaching the Kindle Unlimited market, which seems to be escaping my targeting.

The other bugaboo is scaling up. If $10 in ads = X in sales, $20 in ads does not equal 2X, more like X+2, LOL. Some of this comes from sales say in Canada, Uk, and Australia where I have no ads running but still make the rare sale, or in countries unnamed who are shopping through Amazon US and UK. I seem to have found a small audience somewhere in a country that does not have a specific Amazon site, which I thought was kind of cool.

Bugaboo #3 is that Amazon is botching my book rankings by categories and denying me exposure on the top 50 and 100 lists of smaller categories, and when I contact them, they seem to either not get what I'm saying or ignore me. Rather annoying.

 

[Prince of Spires]

One bugaboo now is reaching the Kindle Unlimited market, which seems to be escaping my targeting.

I seem to remember that you mainly target trad-published authors with your ads. Maybe that explains part of it, since they probably aren't in KU.

Bugaboo #3 is that Amazon is botching my book rankings by categories and denying me exposure on the top 50 and 100 lists of smaller categories, and when I contact them, they seem to either not get what I'm saying or ignore me. Rather annoying.

Yeah, they changed their policy somewhere in the last few months where they only rank you for 2 or 3 categories (which they magically decide), instead of for each category you're in.

Part of that makes sense, since otherwise the top 50 of most fantasy categories are dominated by the same 50 books. But because Amazon places your book into a few categories besides the ones you ask for, and then you only rank for 2 or 3 of them, you can end up ranking for some weird categories...

Has either of you ever tried Twitter ads? I'm wondering with the current Twitter implosion and advertisers fleeing the platform if maybe it's a cheap place to get clicks. Just don't have an idea on where to start (having never actually used twitter...).

I recently experimented with Twitter ads. Attempt 1 saw my account banned from running ads again. Turns out you need a header image, and a website link in your bio. But instead of simply telling the user that they just block the account... Anyway, finally got that fixed and managed to run an ad.

They're harder to target from what I can see than Facebook ads. As in, you have fewer options. It's more about targeting followers of specific Twitter users than anything else. And so far I've had pretty expensive clicks (is in, 80ct per click), with low conversion. I don't doubt that it can be optimized and be useful. But at first glance it's not an untapped goldmine.

 

[Demesnedenoir]

I expected Twitter ads to suck. The good ol' days of privacy invasion with FB seem to be fading. Because people didn't get they were chit-chatting on an ad platform, LOL.

I seem to remember that you mainly target trad-published authors with your ads. Maybe that explains part of it, since they probably aren't in KU.


Yeah, they changed their policy somewhere in the last few months where they only rank you for 2 or 3 categories (which they magically decide), instead of for each category you're in.

Part of that makes sense, since otherwise the top 50 of most fantasy categories are dominated by the same 50 books. But because Amazon places your book into a few categories besides the ones you ask for, and then you only rank for 2 or 3 of them, you can end up ranking for some weird categories...


I recently experimented with Twitter ads. Attempt 1 saw my account banned from running ads again. Turns out you need a header image, and a website link in your bio. But instead of simply telling the user that they just block the account... Anyway, finally got that fixed and managed to run an ad.

They're harder to target from what I can see than Facebook ads. As in, you have fewer options. It's more about targeting followers of specific Twitter users than anything else. And so far I've had pretty expensive clicks (is in, 80ct per click), with low conversion. I don't doubt that it can be optimized and be useful. But at first glance it's not an untapped goldmine.

 

[Demesnedenoir]

Everything was rolling until about a week before Christmas when the Amazon ads just stopped feeding. My assumption is that people had the bidding cranked up for the holiday last-minute shoppers. This has more or less remained in place until yesterday. Most of my serves on Amazon were coming from indie authors and my better ad with named authors had pretty much dumped. It's now coming back. The CTR was about 25% of normal, meaning it wasn't feeding at the top of the page with the bidding structure. Now, things appear to be returning to normal and already beginning to see sales. Hopefully that's not a coincidence, heh heh.

 

[Prince of Spires]

interesting observation. I'm curious who did the christmas bidding. If it was trad-publishers or just indies. Or of course a combination of both. I guess we'll never know At least good to see things returning back to normal.

 

[Demesnedenoir]

Yeah, it's all tea leaves, but targeting Trad/Name Authors was the one clobbered the worst, although both were damaged.

interesting observation. I'm curious who did the christmas bidding. If it was trad-publishers or just indies. Or of course a combination of both. I guess we'll never know At least good to see things returning back to normal.

 

[Demesnedenoir]

Well, I have been digging deeper into more people's opinions on Amazon, FB, and Bookbub ads, and have a few thoughts to smash out. Much of what I thought has been confirmed, the big one being:

Making a profit on Bookbub ads is damned near impossible, heh heh.

Concept: However, it was pointed out that BB ads can be useful in manipulating also boughts on Amazon, which in reality means manipulating the data that the Amazon algorithm is working with... Hmmmm, me thinks. I kinda sorta tried this once but lost patience and quit. So, one can take a short-term loss on BB ads to tickle the Amazon algorithm. I am curious, but...

Conclusion: This might be tricky with a book like EoS that has been out for 4 years and sold thousands of copies. However, with my pending release, this might make sense.

On Amazon, I must dig into how to use their attribution system. I've been meaning to, but it's now a must.

FB ads I still have a minimum of confidence in despite their being my go-to in the past. However, once again, they can still be used to prime the Amazon algorithm's feedback loop if I'm understanding things correctly. What I'm basically piecing together when it comes to "unattributed" sales with Amazon ads firing well and FB ads getting clicks, and Amazon's ad platform telling me I'm not getting many sales, is that those sales are coming from Amazon's "recommended" emails they send out. I've suspected this for a long time, but I'm getting additional info to confirm this. So! The ultimate goal then is to juice the Amazon algorithm to recommend your book (step 1) but also to recommend your book to the right readers, which means you want your also boughts and comparable authors/titles aligned, or you will waste success in step 1.

If you're following the meaning of my rambling, this is how I totally cluster-flubbered myself way back in 2018 when I released Eve of Snows and it was relatively on fire at times, but most of the emails I got recommending EoS were combined with a random shit-show of books that were also-boughts because of something no one should Ever Ever Ever do! Which is to run a Goodreads giveaway. That F'd my also boughts and all the emails getting sent out right up the wazoo.

So, all that jibber-jabber aside, I'm now designing both the ad campaigns for my next release while also taking some different approaches to re-firing the algorithm for EoS and seeing what happens.

 

[Demesnedenoir]

Oh sheeeeit. I am connecting some serious dots and understanding not just that I F'd up, but more details on the why my F'ups F'd me up, heh heh. I had the Amazon algorithm by the nuts twice and let go both times, combined with my Goodreads promo fiasco.

Another duh! When doing a $0.99 promo target the shit out of books/authors you want the algorithm to use as comps for you and your title.

 

[Demesnedenoir]

Keyword performance oddities.

Such weird things... I didn't realize that I had Neil Gaiman targeted into ads. In one, it's a complete disaster. In the other, it's at profit. Ummmmmm. Huh. I don't even think of Gaiman as a comp, really.

Brandon Sanderson? Loser. Mistborn, Big Loser! Well of Ascension? Big Winner. Ummmmmmmm. Huh.

Oh, a ha ha! for those who know my opinion of Name of the Wind... Name of the Wind is a big loser! Patrick Rothfuss is a winner. Ummmmmmm. Huh.

I could go on, but those are the two weirdest I picked out.