People do buy ads. Split from: Also-boughts are back!

[Anarchist]

People do buy from ads.

[TimothyEllis]

People do buy from ads.

Ok, so now you have to explain how you do that. Because I've never seen anything like that result.

And even seeing it, I dont believe it.

According to that, every single click sold a book. And for only 9c each. Back in 2016 I'd have believed it, but not now.

I seriously don't believe that is real. If it is, you need to explain it.

[Anarchist]

People do buy from ads.

Ok, so now you have to explain how you do that. Because I've never seen anything like that result.

And even seeing it, I dont believe it.

According to that, every single click sold a book. And for only 9c each. Back in 2016 I'd have believed it, but not now.

I seriously don't believe that is real. If it is, you need to explain it.

I'm fine with people disbelieving it. In fact, that's a sensible response.

I'm running thousands of campaigns. The above is just one keyword in one campaign. Here are similar ones:








This is the tip of the iceberg. I've been doing PPC and email for a long time. That's why I focus my time and capital there.

I mentioned awhile back (I don't remember the sub-forum or thread) that people might be stunned by my AMS numbers. But I don't expect anyone to believe them.

[Joe Vasicek]

People do buy from ads.

Ok, so now you have to explain how you do that.

Seconded.

[Bill Hiatt]

People do buy from ads.

Ok, so now you have to explain how you do that. Because I've never seen anything like that result.

And even seeing it, I dont believe it.

According to that, every single click sold a book. And for only 9c each. Back in 2016 I'd have believed it, but not now.

I seriously don't believe that is real. If it is, you need to explain it.

I'm fine with people disbelieving it. In fact, that's a sensible response.

I'm running thousands of campaigns. The above is just one keyword in one campaign. Here are similar ones:








This is the tip of the iceberg. I've been doing PPC and email for a long time. That's why I focus my time and capital there.

I mentioned awhile back (I don't remember the sub-forum or thread) that people might be stunned by my AMS numbers. But I don't expect anyone to believe them.

I believe. I'm sure you have better things to do than fire up Photoshop and create fake AMS results. Assuming you're not trying to sell a course or something, there'd be no point in that, anyway.

I don't expect you to give away all your trade secrets, but it would be nice to know, in general terms, how you manage to get that result. Mine looks kind of like yours--but with the the spend and sales columns reversed.

Of course, it's possible your books are amazingly better than mine and/or that you have an enormously bigger fan base. The latter is almost certainly true, though existing fans would probably be buying without clicking on AMS ads, at least for the most part.

We know that AMS ads should be effective because the people already shopping on Amazon would be more motivated to buy on Amazon than people browsing Facebook, for example. But for most of us, they've never quite lived up to their early promise. I spend time writing good ad copy and picking appropriate keywords, and I get a few sales, but not many. I think I get a somewhat larger number of KU reads. (I've experimented with stopping ads, and reads always go down when I do and back up when I start again.) At times, AMS success almost seems to be based on some kind of magic unavailable to ordinary mortals.

[Bill Hiatt]

It would have been better for everyone--Amazon, authors, readers--if they'd never allowed sponsored books at all, or at least kept them below and perhaps smaller than the real also-boughts.

Some may see this as heresy for even suggesting it, but frankly, if Amazon wanted to make more money off KDP, they'd do better by (say) lowering the royalty rate to 65%, or charging a "per-month, per-book" fee to be on KDP--if that would keep the neutral recommendation engine in place over the pay-to-be-seen. The heavier the influence of ads, the more of a Wild West it becomes and the less the customer benefits, IMO.

"Money changes everything." --Cyndi Lauper/Tom Gray

I don't see it as heresy.

The problem is that we don't know how much Amazon makes from book sales vs. how much it makes from ads. More precisely, we don't know how much it makes from KDP books vs. how much it makes from ads. (I see some trad ads on book pages, but not that many compared to all the KDP AMS ads I see.) If Amazon is making an enormously greater amount on ads than on book sales, then making ads prominent is a reasonable business strategy.

Of course, the best strategy would be to find a balance that maximizes income from both. That would seem to mean, at minimum, that ads shouldn't interfere with Also Boughts, which should retain a prominent position. Right now, Amazon seems to be heading back to Also Boughts above all the sponsored product stuff.

I've expressed skepticism before about how much Amazon can possibly make from CPC ads in two huge carousels. I checked the page for my first book, and the carousel right below the Also Boughts has 14 pages. At 13 books a page, that's about 182 (the last page isn't full). The bottom carousel has 62 pages. That means another 806 ads, for a total of 988 ads on one page. Do people actually page through all those ads? My guess has always been that what clicks there were occurred on the first page of the first carousel. Yet I must be wrong, since Amazon continues to put out that many ads.

Despite all I read about ad blindness, which would suggest that floods of advertising gradually lose their effectiveness, Amazon is certainly not the only company to do this. For a long time, every search result on Google has sponsored ads at the top and sometimes other places as well. That's just one example, of course. Even relatively small blogs sometimes have both left and right sidebars filled with ads, leaving the actual content in a narrow ribbon between them. The blogs that are more restrained seem to have sponsored ads every two paragraphs or so. That trend doesn't seem to be diminishing, either.

Maybe when ad blindness really kicks in in a big way, those trends will change. Until then, I understand Amazon's desire to monetize the page space, even though it still annoys me.

[Anarchist]

I don't expect you to give away all your trade secrets, but it would be nice to know, in general terms, how you manage to get that result. Mine looks kind of like yours--but with the the spend and sales columns reversed.

My AMS strategy has evolved over the last several months. These days, I keep my cards close to my vest. The only info I'd be willing to share these days would be so rudimentary as to be useless.

We know that AMS ads should be effective because the people already shopping on Amazon would be more motivated to buy on Amazon than people browsing Facebook, for example. But for most of us, they've never quite lived up to their early promise. I spend time writing good ad copy and picking appropriate keywords, and I get a few sales, but not many. I think I get a somewhat larger number of KU reads. (I've experimented with stopping ads, and reads always go down when I do and back up when I start again.) At times, AMS success almost seems to be based on some kind of magic unavailable to ordinary mortals.

One thing that's not evident from the above screen shots is that I have plenty of dogs, too. That's part of the process. The key is to prune 'em quickly. I only posted the above results to express that many people do buy from ads.

Also, I bid on lots of stuff and let the data sort it out. You wouldn't believe some of the odd keywords that convert for me. I won't reveal them, of course. But it's akin to advertising a book on how to fix a car's transmission, bidding on "Gordon Ramsey" and seeing results like those above.

It makes no sense. But again, I just follow the data.

I think a lot of people just load up titles and authors in their genres and that's it. Everyone does the same thing. That's why bids skyrocket, and advertisers complain that ROI is minimal or negative.

My approach is different. It's complicated. And it may stop working tomorrow. Who knows? But right now, it working fine. And if things change tomorrow, I'll adapt. Nothing lasts forever.


In the meantime, this might really blow some people's minds...





I have a lot of keywords like this one. And while I'm at a loss to explain what's going on, I have a theory (which I prefer to keep to myself).

[David VanDyke]

I've seen some results somewhat like this--not so dramatic, but odd--that makes me wonder if Amazon isn't gaming their own system. In other words: are they reporting the data we think they're reporting?

All it would take was to report sales that do not come directly from ads.

Say, for example, that they reported sales of your books that had ads running, but where that sale had no relation to an ad. No impression, no click, etc. That would skew Amazon's data in favor of looking like the ad got you that sale (and therefore a better ACoS), when the sale would have happened anyway.

[Anarchist]

I've seen some results somewhat like this--not so dramatic, but odd--that makes me wonder if Amazon isn't gaming their own system. In other words: are they reporting the data we think they're reporting?

All it would take was to report sales that do not come directly from ads.

Say, for example, that they reported sales of your books that had ads running, but where that sale had no relation to an ad. No impression, no click, etc. That would skew Amazon's data in favor of looking like the ad got you that sale (and therefore a better ACoS), when the sale would have happened anyway.

That's possible. There's no way to know for certain.

[Bill Hiatt]

I've seen some results somewhat like this--not so dramatic, but odd--that makes me wonder if Amazon isn't gaming their own system. In other words: are they reporting the data we think they're reporting?

All it would take was to report sales that do not come directly from ads.

Say, for example, that they reported sales of your books that had ads running, but where that sale had no relation to an ad. No impression, no click, etc. That would skew Amazon's data in favor of looking like the ad got you that sale (and therefore a better ACoS), when the sale would have happened anyway.

What an interesting thought! I have a number of sales that aren't associated with ads, but perhaps Amazon doesn't bother to do that with prawns.

However, if Amazon were that interested in making the ads look productive, wouldn't it have been easy to associated ads with the number of KU borrows generated? It's true that it couldn't assign a dollar value to the borrows, but people could check their pages read stats and do that for themselves. What they can't do is equate any of that activity to the ads without some information from Amazon. Amazon's been asked repeatedly for this data and keeps not providing it. This doesn't look like a company trying to show off its ads in the best light. (It's possible AMS doesn't have that much impact on KU, but it certainly seems to for me.)

On the other hand, Anarchist's most recent example certainly looks as if something is seriously wrong somewhere. Over 700 bucks in sales from one click? I suppose if an author had a big enough catalog, and the clicker went along and bought every single title, something like that might happen, but how often would anybody behave that way? I usually only purchase multiple titles when I'm already familiar with the author. No matter how good an ad, I don't go out and buy the whole catalog.

[Joe Vasicek]

In the meantime, this might really blow some people's minds...





I have a lot of keywords like this one. And while I'm at a loss to explain what's going on, I have a theory (which I prefer to keep to myself).

It doesn't blow my mind so much as it makes me skeptical about all the "data" that AMS claims to give. Even if that one person bought all of the books in your catalog, then used your affiliate link to buy a big ticket item and the affiliate income somehow got attributed back to your ad, I doubt it would be that high.

OTOH, if that one click got your book over the tipping point for a particularly lucrative also-bought slot, and AMS for some reason counted all of the sales from the also-bought toward the ad, then that might explain it. Just what is the relationship between ads and also boughts, anyway?

[Anarchist]

In the meantime, this might really blow some people's minds...





I have a lot of keywords like this one. And while I'm at a loss to explain what's going on, I have a theory (which I prefer to keep to myself).

It doesn't blow my mind so much as it makes me skeptical about all the "data" that AMS claims to give.

If my theory is correct (and my KDP dash numbers suggest it might be), it's plausible that Amazon is reporting accurate data. Unfortunately, AMS makes it difficult to definitively associate inputs and outputs, cause and effect.

We're not flying blind exactly. It's like flying in a heavy fog: limited visibility.

[Bill Hiatt]

In the meantime, this might really blow some people's minds...





I have a lot of keywords like this one. And while I'm at a loss to explain what's going on, I have a theory (which I prefer to keep to myself).

It doesn't blow my mind so much as it makes me skeptical about all the "data" that AMS claims to give.

If my theory is correct (and my KDP dash numbers suggest it might be), it's plausible that Amazon is reporting accurate data. Unfortunately, AMS makes it difficult to definitively associate inputs and outputs, cause and effect.

We're not flying blind exactly. It's like flying in a heavy fog: limited visibility.

In other words, your KDP dashboard attributes enough sales to your books during the right period of time that the ad could conceivably have generated that kind of avalanche? (Or perhaps we should say, advalanche.)

The flying in heavy fog analogy is very apt. Long delays in reporting the ad data make it very hard to coordinate with sales data.

One of the things that makes me crazy is that AMS lumps all kinds of sales together (different formats, for example). It may or may not lump sales of different products coming from the same ad click together. (I hear conflicting reports on that.)

The one thing I will say is that Amazon has at times issued corrections, usually with the effect of making the ads look less appealing. If it were trying to deceive us, I doubt it would do that. Still, the system, like most of Amazon, could be much more transparent than it is.

[Anarchist]

In the meantime, this might really blow some people's minds...





I have a lot of keywords like this one. And while I'm at a loss to explain what's going on, I have a theory (which I prefer to keep to myself).

It doesn't blow my mind so much as it makes me skeptical about all the "data" that AMS claims to give.

If my theory is correct (and my KDP dash numbers suggest it might be), it's plausible that Amazon is reporting accurate data. Unfortunately, AMS makes it difficult to definitively associate inputs and outputs, cause and effect.

We're not flying blind exactly. It's like flying in a heavy fog: limited visibility.

In other words, your KDP dashboard attributes enough sales to your books during the right period of time that the ad could conceivably have generated that kind of avalanche? (Or perhaps we should say, advalanche.)

That's correct.

It's difficult to put the pieces together due to delayed stats. But when I time my audits right (luck mostly), I've been able to isolate factors that hint at causation.

[Lu Kudzoza]

I have a lot of keywords like this one. And while I'm at a loss to explain what's going on, I have a theory (which I prefer to keep to myself).

I'll theorize for you.    Amazon gives the ad credit for other books bought from your catalog in the same session as the ad was clicked. For example: Reader clicks your ad, doesn't even buy the book in the ad, but buys a few other books from your catalog. If you're using amazon advantage and you're selling other products besides books the ad gets credit for that too. So, you have a book about fixing cars, person clicks your ad, sees some auto parts that you sell on your page and decides to go buy that, presto changeo you make $700 on one 35 cent click.

I'm guessing you sell non-fiction too. Am I correct?

[Anarchist]

I have a lot of keywords like this one. And while I'm at a loss to explain what's going on, I have a theory (which I prefer to keep to myself).

I'll theorize for you.    Amazon gives the ad credit for other books bought from your catalog in the same session as the ad was clicked. For example: Reader clicks your ad, doesn't even buy the book in the ad, but buys a few other books from your catalog. If you're using amazon advantage and you're selling other products besides books the ad gets credit for that too. So, you have a book about fixing cars, person clicks your ad, sees some auto parts that you sell on your page and decides to go buy that, presto changeo you make $700 on one 35 cent click.

I'm guessing you sell non-fiction too. Am I correct?

I don't use Advantage for advertising books in the U.S.

I do sell non-fiction.

[notthatamanda]

I have a lot of keywords like this one. And while I'm at a loss to explain what's going on, I have a theory (which I prefer to keep to myself).

I'll theorize for you.    Amazon gives the ad credit for other books bought from your catalog in the same session as the ad was clicked. For example: Reader clicks your ad, doesn't even buy the book in the ad, but buys a few other books from your catalog. If you're using amazon advantage and you're selling other products besides books the ad gets credit for that too. So, you have a book about fixing cars, person clicks your ad, sees some auto parts that you sell on your page and decides to go buy that, presto changeo you make $700 on one 35 cent click.

I'm guessing you sell non-fiction too. Am I correct?

I don't use Advantage for advertising books in the U.S.

I do sell non-fiction.

Sorry, I don't know if this thread is dead, but I'm trying to understand AMS as well.  Can I ask Anarchist if you sell high ticket items on your page?  I wasn't even aware you could do that, not that I'm planning on it.
I've thought that it's possible for sales not to be hooked to the ads.  Buyer sees ad, is interested, but doesn't want to click on it and lose the page they are on now.  So they open another tab and search on it.   (I do this all the time, but maybe I'm weird.) Why would the sale be hooked to the ad then?  I have one sale credited to a keyword that has no clicks.  Is the scenario I just described how that happens?  Does Amazon track what the user is seeing but not clicking on?
I'm just trying to sort all this out before I commit more $ to the cause.
For the record, I think the suggested bids are very inflated, if my own author name is any indication.  I'm not worth 45 cents, I promise you.