Amazon Ads for Authors: Two Case Studies Showing They Do Work
Amazon ads have been the talk of indie author town since Amazon opened its AMS platform to all KDP users in 2016. While Facebook ads become increasingly competitive, as Mark Dawson already predicted on this blog back in 2015, Amazon have made huge improvements to their advertising platform, and are progressively becoming one of the most favoured channels for marketing a book.
Like any other advertising platform, the AMS platform comes with its own challenges and learning curve. And like any other platform, success largely depends on data analysis and iteration. In this post, we’ll analyze two case studies (one fiction, one nonfiction) and try to understand what these authors are doing right.
A brief intro to Amazon AMS ads
Before we get into the case studies, it’s important to have a look at what Amazon ads do. If you’re already familiar with them — or have taken the Reedsy Learning course on Amazon ads for authors, or Dave Chesson’s video training — you can skip this part.
When judging any advertising platform, you should think about four crucial things:
- Placement: where do the ads show up?
- Targeting: how can I decide whom the ads get shown to?
- Cost: when am I being charged? What is the minimum I need to invest?
- Analytics: how do I know if my ads are working or not?
There are two types of ads you can try on Amazon:
- Sponsored Product: these ads show up on search result listings and other books’ product pages (below the “also boughts”).
- Product Display: these ads show up right next to the book description on a product page, and on the Kindle homescreen and screensaver.
For Sponsored Product ads
Amazon will let you choose between two options for Sponsored Product ads:
- Automatic targeting: you let Amazon choose where to display your ad.
- Manual targeting: you choose keywords and Amazon will run your ad when people search for those keywords.
For Product Display ads
There are also two targeting options for Product Display ads: either by interest, or by product.
“Interest” is pretty simple. You just select one or several Amazon categories or subcategories.
However, with “product” you can have more fun! You can select which exact titles you want your ad to show up for. Do so by searching for them by name, by genre, or by ASIN (Amazon identification number).
While Facebook generally charges you per 1000 impressions, on AMS you only pay when you get a click. This means that if your advert is shown to 20,000 people, and no one clicks on it, you won’t be charged a thing.
For Sponsored Product ads, you can start advertising for as low as $2 a day. For Product Display ads, you need to set up a “lifetime cost” per campaign (the minimum is $100).
Note: these are budgets only. It is extremely likely that Amazon will actually only spend less than 50% of the budget you put in.
Amazon’s AMS dashboard is fairly simple (too simple, even). It shows your campaign’s budget, ad spent, impressions, clicks, cost-per-click, sales, and crucially, the ACoS.
“What the hell is an ACoS?” Good question. It’s an acronym you don’t come ACroSs (hehe) on platforms like Facebook Manager or Adwords. It stands for “Advertising Cost of Sale,” and it’s a ratio that measures the amount you spent on the campaign relative to the sales you earned directly related to that campaign.
Basically, ACoS = Spend / Sales.
It’s a wonderful little metric that tells you, at a glance, whether your campaign is making or losing you money. Note that sales are calculated by Amazon based on the retail price of your book, not your royalties.
- If your book is priced between $2.99 and $9.99, you’re making 70% off of each Amazon sale, so your ad is making you money if ACoS < 70%;
- If your book is priced below $2.99 or above $9.99, you’re earning 35% royalties, so you’re only making money if ACoS < 35%.
All in all, Amazon Marketing Services offers a relatively straightforward platform with clear reporting. Its main advantage, though, is that it lets you advertise your books in a place where readers are actually searching for and looking to purchase books.
Which is exactly why these two authors gave it a try. Let’s dive into their case studies!
Amazon Ads Fiction Case Study: Annelie Wendeberg
Annelie Wendeberg is an award-winning author of historical and climate fiction, and has been using AMS ads on three of her books in the past few months: The Devil’s Grin (Victorian mystery/thriller), Keeper of Pleas (Victorian mystery / police procedural), and 1/2986 (Climate Fiction / Hard Science Fiction). All three are the first books in their respective series.
The results? Pretty good.
“I just started the ad for the science fiction, so I can’t say anything about performance yet. The mysteries are doing pretty well, with an ROI (return on investment) of around 500% (includes paperback sales and sales of all books in the series).”
Here’s a first pointer for you series authors: advertising the first book in the series means that the readers you’ll acquire through the ads can then purchase and make their way through the rest of their series.
Another good pointer is the type of ads Annelie went for:
“All my ads are sponsored product ads running with manual keywords. I didn’t try Display ads yet, because I find the placement odd — kind of far away from where I would look when I’m interested in purchasing a book.”
Now, the main issue when running Sponsored Product ads with manual keyword targeting is, well… finding the right keywords. Here’s how she goes about it:
“I try to find authors and books that are similar to me and my books, and then I test those keywords. Anything that performs well is then again tested for similar keywords. And so on.”
You’ll probably find that you need to test a lot of keywords before you find some that really work for you. A good way to find more keywords is to go through your book’s “also boughts,” and then those books’ “also boughts,” and so on.
Annelie also says:
“There are 4 key elements to successful advertisement: (1) Copy, (2) Image, (3) Target audience, and (4) Data. Number crunching is immensely important, otherwise you’ll have no clue what’s working and what isn’t. How often is the ad shown? How many people find it interesting enough to click? And how many of those people purchase the product? Why is one keyword working and another isn’t? Spend time on analyzing your data, and do it daily when starting a new ad.”
“For example, when you look at the data in the screenshot I took of one of my ad’s performance, you’ll see that some keywords have been shown more frequently than others. “Rhys Bowen” has been shown more than 20,000 times, giving me a large enough sample size to see that the percentage of people purchasing the book is statistically significant. I’ll keep that keyword, because it performs well. Another keyword looks like it’s performing well — “Tracy Grant” — but the number of impressions is a little too low to be certain (less than 2000). All keywords with a number of impressions around 1000 to 3000 are scrutinized the most. If they get too few clicks / purchases after 3000 impressions, I cull them.”
Takeaways for fiction authors:
- Advertise the first book(s) in your series;
- Try Sponsored Product ads first;
- Use manual keyword targeting and test keywords patiently;
- If a keyword gets too few clicks/purchases after 3,000+ impressions, pause it.
Amazon Ads Nonfiction Case Study: Joseph Alexander
Joseph Alexander started writing books about guitar five years ago (in 2012), and in that time has gone from being ‘just one guy writing books’ to becoming a small company that turns over around $500,000 a year.
“Back then, I didn’t have the advantage that AMS gives new authors. In fact, it was only in November 2016, when AMS became available to non-KDP-Select authors that I experimented with AMS.”
Let’s see his current results and then look at a few specific examples.
The first thing to notice is that he too has 100% gone with sponsored products.
“I tried display advertising but I couldn’t get it to work for me. I’ll give it another go in the future, but most people I’ve spoken to have had similar results.”
Next, let’s take a look at the average daily spend.
“I set this to $5 or £10 but unfortunately I’m never able to spend my budget. I wish I could, because look at that conversion rate! $5 a day (which I’m unable to reach) is an extremely cheap way to find out if your advert is working. The lesson here is to experiment as much as you can. It’s very, very cheap to find out what works.”
Now, remember what we were saying about being charged only per click? If you look at the third line in his dashboard, it’s not the one making him the most money, but it’s driven 1.5M impressions so far. In terms of brand awareness, that’s pretty “impressive”, eh?
OK, let’s look in a little more detail at Joseph’s actual adverts and setup. Here’s his top campaign:
The main thing to note is that his targeting is automatic.
“Here’s my logic: Amazon knows! It’s creepy, but Amazon knows what you’re shopping for, what you’ve been looking at, what your interests are and how best to deliver the right advert at the right time.”
For more insights into Joseph’s campaigns, take our free course on Amazon ads! Two of the 10 lessons are written by him and he shares everything about his process. On top of that, Joseph’s a freelance book marketer on Reedsy. If you’re a non-fiction author looking to self-publish or to dabble with AMS ads, you can get in touch with him and seek his help.
Takeaways for nonfiction authors:
- Try Sponsored Product ads first;
- If your book’s metadata is well-optimized, go for automatic targeting;
- Even if the ACoS isn’t the best, consider the number of impressions you’re getting and the impact in brand awareness.
What’s your experience with the AMS ad platform been like? Share your insights with us below. If you have any questions for Annelie and Joseph, feel free to ask in the comments as well!