How (Not) To Build Amazon Ads
Get a glimpse into my introductory Amazon Search Ad strategy. In this post I share my ad planning and creation process and the performance from my first - failed - campaigns on the platform.
As a marketer by trade, I started considering what campaigns I would run to promote my book well before I ever finished writing it. I was really looking forward to testing some new campaigns and strategies – specifically paid ad campaigns – that I didn’t have a lot of prior experience with. You know, publish a book and become a more well-rounded marketer in the process. #goals
If I’m being honest, I thought it would be fairly easy. Boy, was I wrong.
Turns out I had set my expectations unreasonably high (it’s a character flaw of mine to be sure) and even though I poured my decade of marketing experience into following Dave Chesson’s free Amazon Ads Course to the letter, here I sit 4 months later, several hundred dollars poorer, and with very little to show for my efforts – except maybe a little humility, some great keyword research, and an even greater drive to get this right.
At this point, you might be thinking this is a bit of an overshare and are wondering why I would bother to publish this post. But there is a method to my madness. I want to:
Share my successes and failures with other self-published authors so that they might have an easier time than me
Track my strategies in a consistent way so I don’t screw it all up so badly next time
Solicit advice from other authors who have already screwed this up and are willing to help
So here we go.
What Are Amazon Ads?
For self-published authors who choose Amazon as a publishing platform, there’s little doubt that Amazon’s native advertisement system (formerly known as AMS) is one of the best ways to get discovered in a very crowded market.
Amazon allows you to choose where to place your ads – either in search results (much like Google) or alongside other books or products on Amazon that are related to yours. The platform also has the added benefit of exclusively reaching people who are already shopping for books (unlike Google or social media). Finally, Amazon’s ad platform allows you to set your own CPC (cost per click), offering close control of your costs and in a way that is basically unheard of on every other ad platform available.
It seemed like as good a place as any to get started, so I began my first ad campaign just a few days after launching my debut novel, Daughter of the Antediluvian World: An Atlantis Origin Story.
How I Got Started
Although I’ve been in digital marketing for longer than I care to admit, I don’t have a ton of experience with ads, so this was a whole new experience for me. I started by absorbing everything in Kindlepreneur’s free Amazon Ads Course and as far as I can tell it was pretty good information, I just haven’t yet figured out how to implement it properly.
I decided to start with general search ads (as opposed to product ads) – ads that would show up any time a related keyword was searched in Amazon by a buyer. Of course, how often you appear depends on about a zillion algorithmic factors, but overall it seemed like the widest net to cast to get started. I also decided to focus on two primary book categories that I thought most closely reflected Daughter of the Antediluvian World – Science Fiction and Action-Adventure. Here’s how I created my first campaigns:
Step 1: Identifying Keywords
This is arguably the most important step in any marketing campaign. Selecting the right keywords to represent your novel is critical to reaching the right audience. I did a lot of this leg work when I was selecting keywords for my book listing in Amazon prior to launch and had a ton of help from Publisher Rocket – an online tool that allows you to scour Amazon for keywords, categories, and book information at the click of a button. (I swear I’m not getting paid by Kindlepreneur for these links, they’re just a really great resource for self-published authors and I highly recommend you check them out.)
When I tell you I obsessed over choosing the right keywords…
I started by identifying about 15 books that were very similar to mine. I reviewed their Amazon rankings, publish dates, plot lines, cover designs, publishing houses, competition and search stats, average monthly earnings, and every other metric I could just to select my allotted book keywords on Amazon (which I ended up changing 4 weeks later). Of course, that wasn’t nearly enough keywords to start an ad campaign (you need 200-300 keywords per campaign to get started) so I repeated that process what felt like 100 times and fell down several rabbit holes to get my keyword list where I wanted it to be.
I also scoured:
Amazon search results for related/suggested keywords and phrases
“Customers Also Bought” recommendations
Top-selling books in my desired categories (whether I was ranking there or not)
Hot new releases
My Magic 8 Ball
Hours and hours of keyword research left with me several spreadsheets that look a bit like this:
Step 2: Writing the Ad Copy
Writing a book is easier than writing ad copy. There I said it. It’s not just jotting down what sounds good and running with it. You need to capture not only the shopper’s attention but also the very essence of your book, in 30 words or less.
I started by grouping my keywords into broad categories to help direct my messaging to certain groups of readers. All my ad research said I should group my keywords based on the psychology of the people searching (uh huh…) as well as group by common themes. Well, that was another whole series of rabbit holes. I finally ended up grouping my keywords into the following categories:
From there I tried to write unique ad copy for each keyword grouping that incorporated what I thought would be the most profitable, lowest competition, and highest searched keywords. Oh, and I had to fit it into 150 characters. That’s no easy task for someone verbose enough to write a 137,000-word novel. And just a tip, the character limit is more like 125 characters, so I ended up making edits to my carefully crafted copy on the fly just to get my ads running. All of my painstaking research culminated with a list of ads based on the book categories I was targeting on Amazon:
Step 3: Building the Campaign
At this point, I thought most of the heavy lifting was done, but it turns out there are about three dozen additional decisions to make when creating your ad campaigns in Amazon Ads. I won’t go into the details of each of those decisions here, but I will share the settings I selected for my ads:
Bidding Strategy: Dynamic Bids Down (This allows Amazon to lower your bid whenever there is an opportunity to and still allow your ad to show – aka a strategy for people on a tight budget)
Ad Format: Custom Text (This is required in order to use your own copy as painstakingly outlined above)
Targeting: By Keyword (Indicates that you want your ads to show up in search by providing the keywords and phrases you want your book to show up for – as opposed to Product Targeting, which is a whole different animal that I eventually attempted and also have so far failed to tame)
Default Bid: $0.51 (Bidding in this way tells Amazon to bid on every single keyword at the amount I selected, unless they could charge less and still be seen, thanks to the Dynamic Bid- Down selection made previously)
Filter: Broad (As opposed to Phrase or Exact, this gives Amazon the best chance to show your book)
The (Un)Finished Product
Weeks of research and several hours messing around with settings in the Amazon Ads platform later I had three seemingly beautiful search ad campaigns running. One targeting Sci Fi readers, one for Action-Adventure readers, and one General.
General Ad: A Lesson Learned
My “general” died a swift and expensive death, lasting only 2 ½ weeks. I would caution anyone interested in running ads against trying to reach everyone with a generic ad. Because an ad meant to be all things to all people ends up being nothing to everyone. I had 783 keywords built into the campaign and ran it for $5.00 per day with a bid of $0.51. Here’s what I ended up within two and a half weeks:
My reach on this ad was phenomenal. I included every keyword I could possibly think of related to my book and tried to appeal to the masses with my ad copy. And it worked! Everyone saw it. But no one bought it. Why? Because I wasn’t speaking directly to anyone, more like shouting into the void.
In two weeks, I spent $70 to make $10. Keep in mind that during this time my book was at launch pricing of $0.99 so the sales aren’t as abysmal as they look, but obviously, this was not a sustainable campaign.
The ACOS – or Advertising Cost of Sales – shows the percentage of your sales that you are spending to advertise. The lower this number is the better, and ideally, you want it to be under 50%. 666% is a bad number for so many reasons. So, I killed the campaign before I hemorrhaged anymore.
In hindsight I now know that I make more in KENP Royalties – the per page read rate I make by being a part of Amazon’s KDP Select Program – than I do in book sales, and this read rate was actually pretty promising (and my royalties from this ad were actually a bit higher than they originally appeared.) But this was still not a campaign I could afford to keep testing.
Sci-Fi Ad: A Tale of Two Tests
I probably let this ad run much longer than I should have, but I did use it as an opportunity to test several different settings in my ads to see what – if anything – would make a difference. I let this ad run for about a month for $5.00 day at a $0.51 bid and ended up with this:
Not great, but at least it wasn’t costing me anything. With a promising CTR and ACOS, I left the ad copy and keywords along and decided to test my budget strategy. In week 6 I increased my bid to .76. (I was told to never use a “normal” number, by bidding an odd value you have a chance of winning over everyone bidding the expected even – in this case $0.75.) When I didn’t see a lot of improvement, I tried increasing the budget on individual keywords that were performing well within the campaign and even updated my bidding settings to Dynamic Up & Down – which allowed Amazon to increase my allotted budget slightly in order to win a placement while still staying within my daily budget restrictions. After Month 2 my performance looked like this:
Boosting my bid made a big difference in how many people I reached, but it didn’t help with conversion rate. My click through rate dropped and my overall ACOS doubled. By now I knew to watch my KENP royalties, but they still weren’t nearly making up for what I was spending.
If throwing money at it wasn’t going to help, maybe updating the ad copy would! In Month 3 I made some edits to the ad copy and added several hundred new keywords to the campaign. I started tweaking keywords weekly based on what was converting and what was costing too much. By the end of Month 3 this was the progress I’d made:
Ultimately, I ended up turning this one off too. It felt like I was making meager improvements as I ran my tests, but I wasn’t willing to keep spending the money I was for the returns I was getting – and for an ACOS in the triple digits. No bueno.
Action-Adventure Ad: Quantity Over Quality
Much like my Sci Fi campaign, I let this ad run for 4 weeks untouched to see how it would do. Like the others I also started it at a $0.51 bid and a $5.00/day budget.
This campaign definitely started off better than my Sci-Fi campaign – which leads me to believe that overall the copy and keywords were slightly better optimized or the audience was a bit closer to my ideal reader – but it still wasn’t performing great. Because of what I discovered about well-performing keywords I actually spun this ad off into a separate product ad, which is another tale of woe that perhaps I’ll highlight in another post.
Rather than testing a different approach than I did with the Sci-Fi ad (rookie mistake) I tried adjusting the budget for this ad as well to see if I could improve performance. Like the Sci-Fi ad, I changed my bid price to $0.76 and adjusted the strategy to Dynamic – Up and Down. However, I also decided to pump up my keyword list at the same time. The result was that my campaign ran out of money and didn’t even run the full week after I made the changes. So, I ended up adjusting my bid strategy back to Dynamic – Down Only and let it run for another three weeks.
What a dumpster fire. Simultaneously increasing budget and keywords inflated by impressions significantly but did nothing for conversions. My ACOS was the highest I’d seen, and I immediately turned the ad off – effectively giving up on search ads (and focusing on product ads) until I could do more research.
The short version of all of this is that I followed best practices in structuring my first search ads, allowed my ads to run for an appropriate amount of time to test their effectiveness and tested various components on each to no avail. In three months, I managed to spend a hefty amount of money and time with little to show for it. So now what?
I’ve missed something. I’m not sure what it is but now I’m focusing on more education, testing and research to try to figure out what it is. The one variable I have yet to alter – mostly because I just don’t want to – is the description of my book on Amazon. According to everything I’ve read – when all else fails it’s likely your book description that is causing the problems. But I’m kind of attached to my description – I think it’s an exciting and accurate overview of my book – so I’m hesitant to make such a dramatic change just yet.
If it’s not my book description the more alarming consideration is that I’ve miscalculated the genre of my book altogether. This would be a bit embarrassing considering I, you know, wrote it but it’s definitely a possibility. Daughter of the Antediluvian World is a genre-bender, and I’m finding that Sci Fi readers are generally not very agnostic. So, my next step will be to take a closer look at my Amazon categories and possibly to restructure my ads for a completely different set of readers. Wouldn’t that be fun?
Stay tuned – I hope to have an update in another few weeks that shines a better light on my ad management skills. Until then I welcome any advice you, my dear reader, might have for me. Want to stay up to date on my progress? Subscribe for monthly updates from Hooke'd! Otherwise, I wish you happy reading, and hope that you will never, ever stop looking for your lost city.