Literary Q&A: How Do I Know if my Keywords are Converting? [Or Optimizing Amazon Keyword Ads] (Series #7)

I did the Amazon ads the way you said with lots of ads and low bids. Now what? I did the keyword ads, but I don’t know ACOS from pizza. How do I make it work better?

Hi everyone! What the author of the question is asking is how to optimize your ads–to make them perform better. It appears they have chosen Sponsored Product ads using keywords as search terms. Okay, so we want to know which keywords are working and which are not.

I do want to make a note here on the “bid low and often” strategy. Make sure you are targeting these ads well. Amazon said not to cross categories in ad campaigns. So make an ad for each category and subcategory you fall into. But we are talking about keywords today.

I have been running ads with keywords with a bid of $0.09 per click. My thinking was that if it takes 10 clicks to make a sale, and my book is $0.99, then I can’t bid over $0.09 a click, because that’s cutting it even–even if it weren’t for taxes, and the fact that I wasn’t getting 1 sale for every 10 clicks. The teen readers didn’t like my covers before–but they LOVE these new ones. Even so, I noticed that once I went down from $0.11 to a $0.09 bid, I started getting substantially fewer impressions and clicks, and only a couple sales.

So before you start anything, [#1] make sure you have a cover the readers like and are clicking on, and [#2] make double sure your description does the job of making the sale. If you are getting impressions, but not clicks, take a look at your cover. Something about it is not convincing readers to click and check out the description. Is your font readable? Do you convey the age group and genre? Is there high contrast, color, or similar attractive features that match the bestsellers in your genre? If you are getting clicks but no sales, the readers are getting to your page, but the description isn’t doing its job–to entice them to buy.

So in that case, go to Fiverr.com and find someone to write the description for you, for about $25. There are cheaper deals, but if you pay anything less, you’ll regret wasting the money. Trust me on this. I hired a few to see what I would get, and the lower the cost, the more I got back a roughly revised version of the summary I’d made up and given them myself. The hired description author can offer all the revisions in the world, but if they can’t craft a terrific description from your notes the first time they offer it to you, it’s doubtable that messing with it will make it better. Don’t waste your time and your money–get it done right the first time.

If you are a new author, you might give yourself some time to settle in and test these things, or you may want to go gung-ho and jump into ads right away. But I suggest you make sure that you’ve got a great, clickable cover and seal-the-deal description first, and at least ten reviews. When you’ve got all that going on, it’s a good time to run ads.

Amazon says you want to check your keyword or ad results and download a report about every two weeks, but since I am an author and I know how hard it is not to check every ten minutes, try to keep yourself away for at least one week at a time, but only to observe. Don’t tweak it until two to four weeks. Tweaking=optimizing. :^D I have found, though, that every week, at least one of my ads has been clicked on at an odd, much more expensive bid, and I have to go into the ad’s targeting and fix it. So keep your eye on the cost-per-click (especially if you have lowered the bid on it), to make sure you aren’t over-paying on a keyword or ad for weeks.

We’re going to be talking about high-performing (or converting) keywords, low-performing keywords, inconsistent keywords, and different possible keyword scenarios. This information is all free on Amazon, I might add, though without my quirky side notes. That’s no fun. Lol. (And sometimes they don’t make a lot of sense.) The same info used on ad reports will show you which ads are converting, just like keywords.

Let’s look at your keywords now. The first thing you want to do is go to your Amazon ad dashboard and from the menu on the left, click “reports.” The keywords are your “targets,” so in the box, type “Sponsored Product Targeting Report,” or “Search Term Report.” You’ll get essentially the same report for these purposes. You can make reports on: targeting, advertised product, campaign, placement, or performance over time.

Click on “Create report.” Then make sure the configuration is the way you want and title the report (you can have it emailed to you, if you choose), then click “Run Report” and it should bring you to a page that says it is “processing.” After a few minutes, the report will be available and you can either find it in your email, or download from there.

When you open your report, it’s going to be a table that shows you each keyword that’s had at least one click during the time period that’s chosen.

The question is, now that you’ve got the info in a spreadsheet, what are you going to do with it? What will it tell you? And how do you optimize it?

First, let’s talk about what you’ve got on the page: the titles of the ads, dates … and under “targeting” is where you’ll find the keywords you seek. But this report has everything you need to know. Now, I was only interested in the keywords that made sales. I consider my sales as income, because they’re trackable–and KENP is gravy for me. (KENP are the pages read through Kindle Unlimited, if you are enrolled in that program)

So, I made a copy of the spreadsheet labeled, “Sales Only” and deleted all the keywords that hadn’t resulted in sales. I ended up making another copy with only the KENP read pages. So between the two, I had a list of keywords that had gotten sales or page reads. Those are the keywords I will compare to see which are converting and which are not.

  • If the keyword is high-performing–add it as an “exact match.”/ Also compare the bid to the CPC (cost per click). More on that coming up.
  • If the keyword is low-performing–add it as a “negative keyword” either as a phrase or exact match./ You may want to decrease the bid, or consider pausing it.
  • If the keyword is inconsistent— move it to a different campaign.
  • If the keyword has low impressions— increase the bid to get more clicks.

But how do I know which of these numbers mean what?

For a high-performing keyword, you want to see a lot of impressions, with a high CTR which is your click-through-rate (It tells the percentage of how many clicks you get out of the impressions you have). Low-performers may have the lowest CTR, or they may have a high CTR with no sales.

The CPC (or cost-per-click) could be on the lower side in a high-performing ad, or not; what you really want to look at, is the relationship between the “spend” category compared to the “14-day sales” category. There should be a gain, meaning you spent less than you made.

The ACOS (or advertising cost of sale) tells how much this keyword (or ad) has cost you, so you want your ACOS to be as low as possible. The high-performing keywords will have the lowest ACOS in relation to the low-performing keywords.

The ROAS (return over ad spend) tells you how much you have left after the ad spending, so you want as high a ROAS as possible.

But how do you KNOW if a keyword is “converting” or “high-performing?” What numbers am I looking for? I hear you. Well, not only did I look stuff up, I got confused and just called Amazon myself. A super nice guy talked to me for about two hours regarding ads and we covered a lot of this.

He said that there is no single metric that will tell you if a keyword is high-performing or not. You have to look at the whole thing–all the category metrics. The Amazon guy said it will depend on your strategy. Meaning, his strategy would be to only have five or so, highly-targeted ads in total. Another would be the strategy I’m using:

Knowing it’s a numbers game and every 1,000 impressions should get you 100 clicks, and every 10 clicks should get you 1 sale, throwing your ad out broad, aimed at your age group, genres, subgenres, competitors, and keywords should–in theory–work out.

A lot of people have luck with this method. And other people with others. If you’ve heard of a better way, let me know. If you want me to test it for you, I’m happy to try it with my own ads and give the results.

Amazon said that you should use all three match types (Broad, Phrase, and Exact) with your keywords. And if the keywords are running with a low ACOS, keep them running and consider raising the bid by 10%–but if they have a high ACOS, consider pausing the keyword.

Remember back there when your keyword was performing well and I said to compare your bid to the CPC (cost-per-click)? Let’s talk about that. If you look at your bid, say it’s $0.21 and your CPC is equal or close to your bid, say $0.19, you’re going to do things differently than if the CPC was much lower than your bid, like $0.09.

Let’s make little tables then, the first one is for when your keyword is converting, and your cost-per-click is equal or close to your bid. (i.e. $0.21 to $0.19)

If the keyword has:

low impressionsbut is convertingincrease your bid to the low end of the suggested bid range
high impressionsand is convertingincrease to the higher end of the suggested bid range
high ACOSwhich isn’t convertingmaintain the bid or pause the ad

This next little table is for when your keyword is converting and you compare the bid and the CPC and this time you find that the cost-per-click is much less than the bid–our $0.21 to $0.09 CPC scenario from above.

If the keyword has:

high impressionsand converts well – with low ACOSmaintain until CPC increases
high impressionsand not converting – with high ACOSdecrease bid to lower than CPC or pause keyword
low impressionsconverting – with low ACOSit says to check for low traffic? and to maintain until CPC increases
low impressionsnot converting – with high ACOScheck for low relevance and consider pausing

So much of that is confusing, I know. What’s the real stuff that we can do to help? Some of us have hundreds of ads, each with hundreds of keywords, and don’t have the time to go through each one comparing ACOS and CPC and whatnot, right? Right. Here’s the last bit, and the part I felt was most helpful.

Scene one: Your keyword, or your ad, has low impressions, what can you do?

  • add relevant keywords, and ASINs of comp books, try category ads, as well
  • use all three match types for keywords
  • increase your bids
  • or use dynamic bidding– with this one, you keep your bid low (like $0.09), and click “dynamic bidding up and down.” The algorithm will raise your bid (up to $0.18 cents in this case, or double your bid max) to get the click, if it thinks you’ll be the better match.

Scene two: Your keyword, or ad, has low clicks. What should you do?

  • first take a look at your cover–ask writer groups for their opinions on your cover. Finding people in your target readers’ age group and genre is the best scenario. Though most readers can recognize an attractive cover, adult male suspense readers are NOT going to have their finger on the pulse of what teen girls want to see in a YA fantasy cover. In fact, I have found these groups to be in direct opposition with belief as to what makes a good cover. Every man I talk to likes best the covers I have that don’t convert, and dislikes the ones I have that are selling.
  • add more, specific keywords
  • use all three match types
  • review your search term report (It’s like your targeting report, but it also gives you the search terms the reader used to find you–which can be extremely helpful) to discover and add keywords that are performing well
  • “A/B test” your creative ad copy on Sponsored Brands campaigns — Sponsored Brands is a new ad platform that makes your ad a banner at the top of the page. You need a minimum of three books to make an ad with Sponsored Brands. This is prime time seating, ladies and gentlemen. Even if the buyer never makes it to the ad at the bottom of the page, they’ll see yours up top for sure! It’s the first thing they’ll see.

Scene three: Your report or your dashboard says that your keyword, or ad, has a high ACOS–or cost of sale–what do you do?

  • this one compares the money you spent with the money you made (ACOS = ad spend ÷ sales). In the case of a high ACOS, you are spending more on your clicks than you are making with your sales. So you need to check your description to make sure that when the reader clicks on your cover, your blurb sells the book. Otherwise you are simply paying for clicks that go nowhere.
  • add negative keywords–take those keywords that have a gazillion impressions and zero clicks, or tons-o-clicks and no sales, and add them as “Negative Keywords” that Amazon won’t even bring up around your ad anymore, like an ex-boyfriend’s name.
  • try using a low CPC and dynamic bids up and down
  • pause the keywords that are not converting (keywords with high ACOS)
  • keywords, or search terms, that have a low ACOS should be added in the “exact” match type.

Scene four: Your report shows your keyword or ad has a low ROAS or return after ad spending. What steps can you take?

(These were much less clear to me and I’ll leave you to decide what they mean or look them up)

  • review product detail page –I think this one refers to making sure your product description, or blurb, sounds good enough to make the sale.
  • create A+ pages –I can only think this refers to having covers that attract, blurbs that fascinate, and ads that entice.
  • create a store and drive your traffic through Sponsored Brands campaigns –this is the one I was talking about that makes your three-book-ad into a banner at the top of the page. I don’t think there’s as much competition with this yet as it just came out on the 18th, so get on while the band wagon’s still rolling out.
  • review the “Keyword and ASIN” strategy –again, we are ending up where we began, with strategy. I don’t know what all the strategy types are, and how many there may be. Let me know if you have interest in this and I can dig a little deeper. Definitely choose your strategy and focus on it, and make all your A+ pages impossible to resist.

I hope this helps some, and good luck with your Amazon Ad campaigns. Let us know how you’re doing and what strategies you want to try, or have me try for you.

Until next week, remember, you can’t always be strong, but you can always be brave.

Step out of your comfort zone and market yourself this week. Don’t beg … offer.

~jenn

Leave a comment