Updated: Apr 8, 2021
This is probably one of the easiest campaigns to set up for your book. I’m not saying easy in terms of automatic success but easy as in the least effort ad type to set up with potentially great returns. For my most recent auto campaigns, also feel free to refer to my 2021 Author Amazon Ads Masterclass.
I have been servicing accounts now for around a year. I currently service around 150 accounts and one of the first things I do every time I get into a new account - is to set up an automatic campaign and a product targeting campaign. We'll cover product targeting in another post. Like I said - it’s one of the easiest ways to get a book advertising on Amazon.
Choose your campaign type: Sponsored.
Next you’ll be taken to the ad campaign set up page. Here you’ll use whatever naming convention you use for naming your campaigns. You’ll add it to whatever portfolio you have set up that’s relevant. Set a start and end date. I personally never enter an end date because if a campaign is working, why would you want it to automatically turn it off? There’s better ways to control budget and spend at the portfolio level.
I always set a daily budget at $5 to begin with and then increase as needed. I see this more as just a safety stop where if you spend $5 in an hour and you’re a new advertiser, something is probably not right. This has only happened a couple times and I’ve run over a thousand campaigns. Just better to have and not need.
Click on the Automatic Targeting - that’s the whole point of this blog post. Alright now for the more subjection options.
Dynamic bids - down only.
This means Amazon will use it’s (questionable) intelligence to lower your bid for things that it doesn’t think are relevant.
As a side note, I guess I’m known as referring to Amazon’s back end algorithms that decide what ads to show when as “The Machine” so for the purposes of this blog, I will refer to Amazon’s algorithm as “The Machine”.
The Machines ability to know what’s relevant is oftentimes irrelevant. A human's ability to draw a network between two topics is far superior to Amazon's ability to do so from what I’ve seen. For example, I sold a copy of my book from the word gardening once. My book is about the college to career journey. I'm guessing some parent was searching for gardening stuff and saw my ad and thought it might be a great book for their about-to-go-to college teen. Who knew gardening, something parents who's kids are about to go to college and have time to do, would be a great keyword for my book? If I had left it to The Machine, it would have lowered my bid for the word gardening when I targeted it.
And that’s just it. I come into contact with the academic types who study this stuff in school or are sufficiently educated in Google Ads or Facebook ads all the time and they think Amazon’s advertising works the same way. I will probably write a blog about the inconsistencies in the Amazon Advertising platform that we get to take advantage of. These individuals tell me that my methodologies don’t make any sense but that's because they see The Machine as perfect and all knowing. I know better.
Long story short here - it’s a mixed bag to rely on The Machine for this stuff. Sometimes, like a school teacher favors the front row student who always pays attention and listens, The Machine can be known to favor authors who do what it tells them to do.
I have some friends in the space who only do down only. It’s the best strategy for minimizing costs.
Dynamic bids - up and down
This isn’t just the front row student - it’s the front row student who brings in an apple for the teacher. It’s the bidding strategy that puts the fate of your campaign in the hands of The Machine. The machine is deciding what’s going to be relevant and raises your bid for that, and then it’s going to lower your bid just like the strategy above, for things it finds irrelevant. Now, The Machine wouldn’t be The Machine if it wasn’t effective at its job sometimes. In fact, a majority of my campaigns use this strategy. I’m not saying it’s always the most effective (see my gardening keyword target story above) but it does seem to provide results. You should keep your bid even lower when using this strategy because The Machine can raise your bid up to 100% more than the bid you set. If you bid 30¢ on a keyword, The Machine can raise that up to 60¢. If you’re only selling an ebook for $1.99, that’s 30% of the cost of that ebook. (Also known as advertising cost of sale).
This is the kid that sits in the back of the classroom and shoots spitballs at the teacher. You’re taking full control (Or so you think). This can be great for being intentional about targeting the things you want to target but can also get you in hot water.
If this is the 1950’s the teacher is the supreme ruler (Or so I’ve heard since I was born in 1996). The Machine is the 1950’s kind of teacher who can hit kids with rulers and if you shoot a spitball that manages to hit the teacher between the eyes, the teacher has the capabilities and juridstiction to leave a welt on your skin.
The Machine’s ruler is the ability to show your ad less if you get it wrong. If your ads aren’t converting and there isn’t a high click through rate, Amazon will mark that targeting as a lower chance of converting and stop showing it. This is where, if you chose fixed bids, you may be starting in the bad books of The Machine. That being said, I’ve also run eternally successful, fixed bids campaigns. It’s all about the combination of bid strategy + product + targeting.
From what I recon, you avoid that somewhat with the down only, and up & down strategies since you’re giving The Machine the ability to show ads for things it thinks are going to be relevant. Now, like I said, The Machine can be wrong too, so you still may be hit with a ruler because your teacher thought it was you that shot the spitball when it’s your giggling, troublemaking friend who sits next to you.
The funny thing is that the whole point of automatic advertising is that you’re giving control to The Machine to find keywords it thinks are relevant to your book. If Amazon is only finding keywords that are relevant to your book then it’s probably going to look for the highest possible bid possible from you. In this case, since relevancy is assumed, I think it makes sense to use fixed bids or down only. Up and down would probably mean Amazon going with whatever the highest possible bid is going to be. This doesn’t mean you’ll pay that bid, just that it’ll use that bid when considering who’s ad is to be shown for a customer search. I’ve often run campaigns that have a higher click through rate than the auto campaigns but it can also work the other way.
As a follow up to this piece, I’ll create a case study where I set up each type for various authors and genres and report back. I also have set up a new study on Bid+ Strategies which you can read about here.
This one’s a bit easier - standard ad is my go to. From the studies I’ve run and the opinion of most other advertisers I’ve come into contact with is that the custom text blurb does not really play much of a difference in the campaigns success or failure. Even if it’s marginally better it seems like a better use of time to focus on your bidding strategy and keyword targeting.
I will not leave you with only a seemingly subjective opinion though. I will also create a controlled experiment (or as controlled as can be) on the difference in results of a standard campaign vs. a custom text campaign. Blog on this and the results to come soon.
Select your product. I Always select my ebook and paperback. Amazon changed their reporting methods in 2020 so that they’d only report on the product being selected. This was very frustrating for authors because this meant that a campaign would not show if someone clicked on your ad and then bought another product that wasn’t selected. Amazon counts the kindle and paperback as separate products. Because of this, and despite the fact this means you can optimize for a certain product type, I default to selecting both the paperback and kindle so that you will see the big picture of a campaign and the true sales it brought it. I think this is more valuable than optimizing for a specific kindle or paperback version because if you miss the fact that one of your campaigns sold a paperback, and you were targeting the kindle, you may turn off a campaign that actually sold copies of your book. I find this deeply frustrating but I don’t think Amazon cares (I have tried to tell them) about that so I have accepted and adjusted accordingly.
I always just set a low default bid to start out with and then if it doesn’t perform, I switch over to the suggested bid as long as it’s reasonable. At first, I set a default bid for all of the targeting groups as you can see below. (I wouldn’t use 75¢, that’s just what the default that Amazon put it at when I took a screenshot.
You can also set the bids separately but I haven’t found a good reason to do so yet as each one will work differently for different books. These are pretty self explanatory so I don’t feel the need to go into them here. The logical argument someone might make is that you’d want to bid lower for loose match than close match but if you can’t tell, I seriously doubt The Machines ability to figure out what that is sometimes.
It is possible that bidding lower for loose match might result in a higher CTR for the campaign. I’m yet to figure out, despite the many experts I’ve tried asking, if Amazon measure CTR, and takes it seriously at the Keyword level, Ad group level, or Campaign level. If you have some insight here, I would love to hear it.
So there you have it. A somewhat quick yet detailed run through of setting up an automatic campaign. I have a similar article for those of you wanting to do extensive product campaigns here.
If you just want to get something up and running. My personal opinion is to select Up & Down. Then do default targeting at a bid you’re comfortable with. I’m a fan of low bids so 14¢ or 15ç as a default bid for all targeting groups to start off with should get your product some starting impressions. I’d say about 20% of my books could just run auto targeting and be very profitable. I also have some books that Amazon seams to find totally irrelevant to anything and so I usually push up the bid, or go with the suggested bid, to see if that can convince The Machine to start showing the ad. If there aren’t any impressions, that’s an indication that your bid should be higher if you want The Machine to consider your product with automatic targeting.
As a disclaimer you should always keep an eye on your campaign(s) because it can also be known to wildly spend money with little results.
Overall - automatic targeting is a mixed bag and yet just another tool in your tool belt that might be a good fit for your book.
Ads still not working, check out the top 9 reasons here.