The Real Keyword and Category Research Behind One Picture Book
One shelf needs 19 sales to hit number one. Another needs 1,073. Same book. Here’s my full keyword and category research, numbers and all. – The KidLit Creator’s Chronicle – Issue #63
One of the categories my book fits in needs 19 sales to hit number one. Another one needs 1,073!
In Issue #61, What Testing Your Book Idea Can Tell You, and What It Can’t, I promised I’d show you my actual keyword and category research, with real numbers. So here it is.
I’ll walk you through the three checks from Issue #61, done on Lilou’s Lost Rainbow, step by step. Then I’ll run a very successful self-published book through some of the same steps, so you can see what it looks like when it’s all working.
I’ve done a lot of this kind of research over the years, helping authors publish. And one thing I’ve learned is that you can never guess keywords, because you simply don’t know how well each keyword will perform on Amazon without doing the research. And categories are always changing, so you need to check them out regularly to choose the best ones.
Why Metadata Research Is So Important
Your categories and your keywords are a big part of what decides whether your book gets found at all. Get them right, and readers who are browsing or searching can actually discover your book. Get them wrong, and nobody finds your book.
Categories on Amazon are the shelves your book sits on, the same way a physical bookshop has a shelf for picture books and another one for cookbooks. Obviously, you need to be on the right shelf to be discovered. No one looks for picture books on the adult horror shelf.
But Amazon’s shelves are huge, containing anything from about 500 to 100,000s of books each. So you also need to be visible on your chosen shelves and not buried under other books. And if you rank number one in a category, you get that Amazon bestseller badge, which is wonderful social proof.
No one wants to be on page 400 of the books in this one category…
Keywords determine if your book comes up in searches. To choose the right keywords, you need to know which ones people are actually searching for, and if you can rank for them. So, in the same way as categories, you need suitable keywords where your book won’t be buried under other books.
So, the research helps you find the best shelves and search terms and to see how competitive these are to rank in.
To do this right, you need actual numbers. For me, I only use Publisher Rocket and think it’s the most incredible tool. It enables you to get the real numbers you can’t get otherwise, and you can do the bulk of it in an afternoon.
The Walkthrough
Categories first
I always search for my categories before my keywords, because once I know which categories I’m using, I use a keyword field or two to anchor those categories. This means to tell Amazon “yes, these shelves are right for my book.” I’ll come back to anchoring in a moment, but it means I need my categories first.
You get three categories per format, so three for your ebook and three for your paperback or hardcover. You used to be able to email KDP to request up to ten categories, but that’s no longer the case. You pick three per format, and that’s it. So every slot counts.
Inside Publisher Rocket, I go to the category search and start with children’s books, which covers the paperback and hardcover. (You can’t yet publish hardcover picture books on KDP, just to avoid any confusion. So in this case, it’s just paperback.)
Here’s what I do:
Filter to “selectable, excluding ghosts.” Amazon has “ghost” categories that readers can’t find, so if your book is in such a category, it won’t help it to be discovered. So, of course, you want to filter those out in Publisher Rocket, since you don’t want to select any of them.
Watch for duplicates. A duplicate is a second category with a different name that puts your book in the same category as its duplicate. Publisher Rocket flags these for you. You can pick one, but don’t pick both, since then you use two of your categories but only get one.
Sort by “sales to number one.” This brings the easiest categories to rank in to the top. Then I look through and see which ones actually suit my book.
You can search, but I prefer to scroll through all of them to make sure I don’t miss anything.
The rule is, go into the deepest subcategory possible, because that means less competition, plus you are still put on the parent shelf too.
I look through a lot of them, clicking on the actual category itself whenever I’m unsure if my book would fit well. This opens the category on Amazon, and sometimes it becomes obvious my book wouldn’t fit well there. Any that could work, I copy and paste into a document, and I add the sales to #1 and to #10. This tells me if I could hit #1, and if I could rank at least in the top few books. Of course, there are some where my book would fit perfectly, but they’re far too competitive to be worth it.
So, I keep going and put all the good and fairly good fits into my document, until I have ten or more. Then I can choose the best ones from my list.
How to choose your categories
Once I had a bunch of categories to choose from, I kept the three categories that truly match Lilou’s Lost Rainbow and with the lowest competition. Books > Children’s Books > Science Fiction & Fantasy > Fantasy & Magic > Visionary & Metaphysical was the easiest choice, since it’s not too competitive and my book really fits there.
To give you a couple of other examples from my own search, my book has fantasy and magic in it, so the children’s Science Fiction & Fantasy > Fantasy > Magic category is a great fit, but it takes 111 sales to reach number one there, up against Harry Potter and Earthsea. The chance of ranking above books like that is basically zero. But in the end, you also have to weigh how appropriate each category is, and you may have to select one where you won’t rank well. Of course, we try to avoid that.
In my case, most of the ones that fit my book fell under the subcategories Religious Fiction > Christian, like Emotions & Feelings, Values & Virtues, and Action & Adventure. My book is not religious, but it fits the smallest subcategory in each of these, and they are less competitive. Some books in those categories are also not religious, but most are. This is where a judgement call comes in, “Do I prefer to fit perfectly, or will I take the chance to choose a category because it’s a fairly good fit and less competitive?”
First, it’s worth knowing Amazon can remove your book from a category if the algorithms decide it’s not a good fit. But there are other risks. So, I chose not to take the chance, for a few reasons:
- Choosing the Christian subcategories means that a browser who finds my book is less likely to buy, because it’s not what they wanted. They were searching under Christian. Amazon pays attention, and poor conversion is a signal to Amazon to show my book to fewer people.
- Kindlepreneur claims the longer a book sits in the wrong category, the more performance data Amazon builds around that wrong fit, making positioning harder to recover even after you switch.
- I already have the Books > Children’s Books > Science Fiction & Fantasy > Fantasy & Magic > Visionary & Metaphysical category, where I only need 19 sales to hit #1, and 2 sales to be in the top 10, so choosing a category that isn’t such a good fit (and has higher competition) isn’t worth it.
So, I ended up choosing Books > Children’s Books > Science Fiction & Fantasy > Fantasy > Magic. Even though I won’t rank so well, it sends the right signals to Amazon (and any buyers who may happen to find my book in this category). Being in a suitable category helps with Amazon also-boughts or recommendation, because it tells Amazon which other titles your book is similar to.
So, the value isn’t only ranking. It’s showing up next to the right books, in front of the right readers.
The only downside here is that two of my categories fall under the same parent category: Science Fiction & Fantasy. But since my only other option is Books > Children’s Books > Action & Adventure, which is much more competitive and a poorer fit, the Magic subcategory is still the best option.
Ebook Categories
After Books, I moved over to Ebooks, which are almost always easier to rank on than print (for children’s books). Almost immediately, I found the colours category, which was near perfect for my book about rainbow colours, and it needed just one sale to reach number one. A completely different situation to the print books categories.
I found seven ebook categories and then stopped, since I was happy with all of them and could easily select three.
Anchoring Categories
Earlier, I mentioned you use keywords to anchor your categories. So what does that mean?
Anchoring means using a keyword or two to tell Amazon your chosen categories really do fit your book. This helps to keep you in your chosen categories.
Here’s how I do it. For each category I’ve picked, you can get Publisher Rocket to show the keywords that connect to it.
I copy those into a list, then delete all the ones that don’t fit my book, and keep the rest. For instance, under the Colors category, there was “colour mixing”. That doesn’t fit my book. But “colours of the rainbow” fits perfectly.
I then go back to Publisher Rocket Keywords, and check the ones that fit. Here I found “colours of the rainbow” has a search volume of 246 and a low competitive score of only 11, so that’s a keyword I can use to anchor my colours category.
Keywords: you can’t guess, you have to check
After choosing anchor keywords that will fill the first one or two keyword fields, it’s time to choose the rest. You get seven fields with 50 characters each, and it’s best to use the characters available, rather than using just a short keyword and wasting the rest.
You can put more than one keyword per keyword field.
I start with what my book is actually about. I put my logline, blurb and description in front of me, because they’re full of keyword ideas. My logline is this: “When a gust of wind steals Lilou’s magical rainbow, she sets off on a brave journey to get her colours back. Along the way, she discovers that the magic she was searching for was inside her all along.”
Straight away that gives me things to search: magic, magical rainbow, brave journey, and so on.
From there, I brainstorm in buckets, which helps me think of more possibilities. I use these buckets:
- Theme: what’s your book about? Resilience, courage, colours.
- Character and story type: is it a girl, a boy, a bear? An adventure, a bedtime story?
- Age and use: mine’s for 5 to 8, though I can test 4 to 8 as well. Is it a read-aloud, a bedtime story, one for the classroom?
- Comparable books or themes: The Wizard of Oz has similarities, a girl on a journey, growing as she goes.
- Gifts and occasions: always worth adding one or two, like “gift for girls ages 5 to 8,” because it can help you get found from a completely different angle.
These are a starting point. I will vary the phrases as I go, and also follow possible leads and ideas I spot inside Publisher Rocket as I go. Also, even though a lot of my brainstorming ideas contain “book” or “picture book”, I research the keywords without those terms too, because KDP prefers if you don’t use “book” (because they already know your book is a book), and it wastes characters to repeat “book” and “picture book” over and over in my keyword fields.
I did end up choosing some with “book” and “picture book” in them, because they are the exact words people search for and they had great numbers. But mostly, I left out the “book” and just used the key idea instead.
Once I have my list, I go into keyword search in Publisher Rocket and work through my ideas, looking for two things: a competitive score that isn’t too high, and a decent number of searches.
I do the same as with categories, where I copy and paste all the good options into a document, and I put the searches and competitive score next to each, as well as the character count in parentheses.
It’s easy to search through hundreds or thousands of keywords this way. It can become very time-consuming, though, so I make a point of finding a bunch of good ones and then stopping.
The most important thing is, you have to actually check, because the results won’t always be what you expect. You can’t guess.
For example, I searched variations around finding yourself and inner strength. “Finding yourself” came back at a competitive score of 4, and “inner strength children’s books” at 12, both very workable. But phrases that sound just as good, like “picture books about confidence,” came in at 36. I’d only know which was which by checking each one.
By finding enough, you’ll find the ones that will really help your book. For example, “kids book about journey” turned out to be a strong keyword, with a competitive score of 5 and 832 searches.
And sometimes a tiny reword changes everything. “Empowering books for young girls” is a good match for my book, but it has 650 searches and a competitive score of 51. 51 is too high. So I changed it to “empowering girl book,” and suddenly it was 1,597 searches at a score of 38. Same idea, much better odds, just from moving the words around.
So, a small change in wording can give you a completely different result. Try it out!
Fitting the keywords together
As a reminder, you get seven keyword fields, and each one holds up to 50 characters.
Most keywords are shorter than 50 characters, so I combine keywords to use more of the available characters. While it’s not public knowledge exactly how Amazon uses the keywords, it does combine and recombine all your entries. So “introduce concepts like colours of the rainbow” gives Amazon “introduce concepts,” “colours of the rainbow,” “concepts like colours,” and several more, all from one field.
Amazon views the keywords or phrases in one field as more closely connected than keywords in different fields. So if you put three keywords in field 1, then Amazon will view them as more closely connected to each other than to the keywords in other fields.
So, group phrases that go together and make sense side by side, and try not to waste characters repeating the same word over and over. It’s a little like building a puzzle. You slot phrases together, find you’ve gone over the character count, take some out, and keep playing with it until it works. There isn’t one perfect answer.
Let’s Look At a Bestseller
Let me show you the same approach on a book that’s doing very well.
I chose This Book Is Perfect by Ron Keres, from the Finn the Frog collection. It’s self-published, which is part of why I picked it, and it has over 4,000 reviews on Amazon.
First, its categories. You can see these right on the Amazon page, under product details. On the hardcover, it’s ranking in three: children’s disaster preparedness, children’s frog and toad books, and children’s general humour books. You’ll notice these are specific subcategories, not a broad category like “children’s books.” It’s gone right down into a sub-sub-subcategory, where there’s far less competition and it’s easier to rank.
You can find out more in Publisher Rocket’s Competition Analyzer. I searched for the exact book there, and it showed me even more categories.
Taking these categories and looking them up in Publisher Rocket’s Category Search showed me that Books > Children’s Books > Growing Up & Facts of Life > Difficult Discussions > Disaster Preparedness needs 21 sales to #1, and 6 sales to #10.
Books > Children’s Books > Animals > Frogs & Toads needs 82 sales to #1, and 11 sales to #10.
So, for both of these, it’s quite easy to stay in the top 10 if you’re making consistent sales, which means this book stays discoverable.
Some other categories were similar. This shows you the author chose his categories well. They’re not only very suitable, but he can rank in them.
Reverse ASIN lookup
This is an excellent tool for finding keywords. You grab the book’s ASIN, which is under product details on Amazon, put it into Publisher Rocket’s reverse ASIN lookup, and it brings up pages and pages of keywords that book has ranked for. Some won’t suit your book, like the title or “frog book” if your book is about a llama. But if you’re patient and look through, you’ll find real, usable keywords, with their search volume and competition, that you can add to your own list.
And I know you may be wondering where the pages and pages of keywords come from. Amazon uses the keywords you put into the 7 KDP fields, but a book also ranks for search terms because of everything else Amazon knows about it, like the title, description, age ranges and the rest of its metadata.
In a few minutes, this enabled me to see that this book uses “this book is just for fun” with a search volume of 832, and competition score of 25, “the monster at the end of this book” with a search volume of 276 and competition score 1, and “children book age 6-8” with a search volume of 348 and competition score of 33. These are already three very strong keywords, out of many.
For this edition, I’m showing you why this book is successful. Having loads of keywords that it can rank for is a part of its success.
For your own research, this can be valuable to find more keywords for yourself if you have a similar book. This is really what the tool is for.
Competition Analyzer
Another helpful tool in Publisher Rocket is the Competition Analyzer. It shows you the top 10 books ranking for any specific keyword, with their page counts, prices and sales. So you can see your competition at a glance, and whether you could realistically make it into the top 10 for that keyword, which is where most of the sales are.
You can also do Reverse ASIN lookups on these books, to find more keywords you might never have thought of.
Over to you
If you have a book idea, draft, or manuscript, try one of the checks I shared in this edition.
Then tell me how it went!
I asked this last time, and I’d love more of your answers. Have you ever done this kind of research for one of your books? How did it go? And if you haven’t, what’s held you back?
Let me know! Your answers really do help me work out what’s most useful to cover next.
If you don’t have Publisher Rocket yet, I highly recommend you get it. It really is an invaluable tool.
If you look back at Issue #61, I decided I was going to write Lilou’s Lost Rainbow whatever the research said, and I did. So, I was okay even if I found I didn’t make the best strategic decisions for marketing, specifically for categories. But the research showed me that while I may not be able to grab the easiest categories, there are plenty of good categories and keywords for my book.
I’m pretty sure that if you do the research, there will also be great ones for your book!
Whatever your goal, whether you’re after loads of sales or you simply want your story into the right hands, having the right categories and keywords makes a big difference. And it only takes an afternoon.
