Note: super-wonky book marketing post coming, likely only of interest to independent authors or publishers. You were warned.
During the process of preparing The Reintegrators for publication, I did more research into things like keyword selection than I had with my previous, shorter works. Since the info on how to choose keywords online is somewhat fragmented/incomplete/not targeted toward authors, I thought I’d put together a little to what I did, so that others can benefit or tell me what they think I’m doing wrong.
Just to make it clear: I don’t claim any proven results from this method. The below is just what I chose to do based on some logical inferences, using the best tools I can find. All feedback is welcomed.
What is an Amazon Keyword?
When you publish a book on KDP, Amazon lets you specify two categories and seven keywords to help readers find your book. Keywords can be a single word or more than one (called a “phrase”). These keywords are used when a user searches amazon–to oversimplify a bit, if a keyword in the user’s search matches one of your book’s keywords (or your title), your book will appear in the search results. There are three main criteria for what constitutes a “good” keyword:
- Volume: How often is a keyword searched for? If no one is looking for books with a particular keyword, it won’t do you any good to use it.
- Competition: How many other books use the same keyword, and will they be ranked above you in searches? Of particular importance is whether your book will show up in the first page of results, since this will exponentially increase how many people see it.
- Relevance: Is your book going to appeal to people who are searching for this keyword? It’s nice to appear in the #1 position in a search for romance novels, but if your book is a gritty post-apocalyptic space marine bloodfest, you’re probably not going to see much in the way of revenues from it.
Aside: Categories
Choices of categories and keywords are intrinsically linked with each other, as the below content will make clear. I would encourage anyone reading this guide to also read Michael J. Sullivan’s category guide, and make sure you take the time to really understand it. You will notice he mentions that some categories are “easier” than others, i.e. your book doesn’t need to be ranked as high in them to appear in a bestseller list. One thing to keep in mind is that this concept is only useful if your book is already selling well; in other words, lets say I have two categories available:
- Category A’s #10 book is ranked 3,000 and my book fits into it really well
- Category B’s #10 book is ranked 50,000 and my book isn’t a perfect fit for it.
If my book is ranked above 50,000, then B may be the better choice, because it will show up when readers browse the bestseller list for that category. However, unless you already have an established audience, odds are when you first start out your books will start ranked somewhere around 800k or higher. In this case, category A is a better choice, because then people who find your book via keyword search in that category will find your work more relevant, and you aren’t going to show up in the top 10 for either category anyway. Once you’ve done the legwork and sold enough copies of your book to rise above the 50k mark, you can always go back and change the category to take advantage of your new status.
So, go ahead and browse Amazon and make sure you’ve found the best two categories for your book before continuing on with this guide.
Step 1: Brainstorm
Create a new spreadsheet. In the first column, start entering all the keywords and phrases you can think of to describe your book. To begin with, go back to the beginning and remember what makes your book special; the elements you thought of that made you want to write it in the first place. Write a synopsis, and go through it chapter by chapter looking for ideas. Mine settings, characters and plot devices. The trick here is not to worry yet whether a keyword is good or bad–this will only slow you down. The next two steps will take care of whittling down the list, so just let ideas flow out of your brain unimpeded for now. Some tips on keywords:
- As mentioned above, any word in your book title automatically becomes a keyword, without counting against the seven allotted to you by Amazon. This is why you often see books with titles like “The Arwen Chronicles – Urban Fantasy Romance.” This is a clever trick to expand your keyword count, although perhaps it makes your listing look a little cheesy, and Amazon’s guidelines actually forbid it.
- Conversely, your book description is *not* used in keyword searches, so any exciting elements from your description need to be captured separately in a keyword.
- Don’t be afraid to consider common two or three (or more?) word phrases. These so-called “long tail” keywords are often under-utilized.
- Make sure you consider variant spellings and different ways of saying the same thing (werewolf, werewolves, wolves, wolfkin, lycanthrope, etc.), though Google will help with this (see step 2).
- Generic keywords like “fantasy” are generally useless because there is too much competition, but if in doubt, just include it; if a keyword is too generic then it will be weeded out in step 3.
Step 2: Look up Volume
The 100,000 pound gorilla of the keyword research game is undoubtedly the Google Adwords Keyword Lookup Tool. This handy tool is meant to evaluate and suggest keywords for Adwords campaigns, but it’s enormously useful to see estimated volumes for Amazon keywords as well. Note the “estimated,” here; we’re making somewhat of a leap in faith that the number of people searching for a keyword on google is approximately proportionate to that same keyword on Amazon. Whether or not this generally holds true in real life is debatable, but since the Adwords data is the best we have, we’ll go ahead and use it anyway, secure in the knowledge that at least we’re not just leaving things to blind chance.
So here’s what to do:
- In the “Word or phrase” box, copy and paste your keywords/phrases, one per line.
- Leave “Website” blank. Set “Category” to “Books & Literature” to improve the quality of keyword suggestions.
- Click Search.
You’ll be given two tables of data; the first contains the monthly search volumes for your keywords, and the second contains additional keyword suggestions, again with volume data. Most of the suggestions will be crap, but sometimes you’ll find some good ones in there. Ignore the “Competition” column, it tells you how many people are bidding on an Adwords keyword and is of no use to you. That leaves two more columns of data: Global Monthly Searches and Local Monthly Searches. The “local” column means searches only from your selected region (the USA by default). I tend to use the global data instead since people buy Amazon books from all of the world, but to each his own. Whichever you choose, copy the numbers from that column into your spreadsheet, along with any additional keywords you want to use.
Step 3: Look up Competition
ort your spreadsheet by volume number, descending. Again, don’t take the actual numbers here too seriously. For our purposes, they’re essentially fictitious, except for their values *relative* to each other. So, if one keyword has a volume of 42,000, and another has 46,000, you may as well just consider them equal. But if another has a volume of 460,000, that’s perhaps reason to consider it as a better option.
Now comes the tedious part. This process could be automated using the Amazon APIs, but from my reading, doing so would violate their ToS, so I’ll just continue as if we’re doing this by hand:
- Open amazon.com and in the dropdown next to the seach box, select “Kindle Store.”
- Enter your first keyword and search.
- On the search results page, change “Sort by” from “Relevance” to “New and Popular” (note, this is probably not indicative of what an average reader will do, but it is more conducive to estimating how popular your book needs to be for a given keyword. As your Amazon rank rises, so will your position in the “Relevance” search as well (or so I assume), just not in as predictable a manner.
- Scroll down to the bottom of the page, and open the last book listed on the first page of results in a new window.
- Scroll to or search for “Amazon Best Sellers Rank.” It will say something like “#XXXX Paid in Kindle Store.” That is the rank you need to beat to come up on the first page of global results. Note it in your spreadsheet in a new column. If you like, you can also note the number of results of the search (at the top of the result page), but this number is not nearly as important; if a search only returns 300 results, it’s still not much good to you if they are the 300 most popular books on Amazon!
- Back on the result page, under “Departments” on the left hand side, find your book’s first category and click it. If your category is not listed, click “Kindle eBooks” and select it from the new list that comes up. If it’s still not listed, then no other books in your category use that keyword–this *might* be a good thing, but it might also indicate that the keyword in question is not very popular with readers of your genre.
- If the category is present, select it and then repeat steps 4-5 above. Note the category-specific competition rank in yet another spreadsheet column. Then rinse and repeat for category #2, followed by the rest of your keywords with enough volume to consider bothering.
- Depending on your dedication level and how popular your categories are, you may also want to repeat this process for the parents and grandparents of your categories, since your book will also show up there as well. Doing this exhaustively could suck up a lot of time, but in general you’ll find after few searches the results are pretty predictable for how much a rank will go up for each child level. In my case, one of my categories is Science Fiction & Fantasy -> Science Fiction -> Adventure. I spent most of my time looking around in the top level and in Science Fiction & Fantasy -> Science Fiction, while also dipping into the Adventure sub-category for some keywords I was particularly interested in. But there’s no reason to check every single category level for every single keyword.
During this process, you should also be thinking about relevance. In general, you want the list of books that come up to be at least somewhat similar to yours. Of course, sometimes you may have a cross-genre book, like a space opera with elements of a political thriller, that you’re comfortable being listed with a bunch of political thrillers. But other times, you want to avoid attracting readers that aren’t looking for what you’re offering. For example, in The Reintegrators, there is a lot of fictional stuff about mathematics, some of the characters are mathematicians, etc. The problem is, while some fiction readers will be interested in those elements, its obvious that the vast majority of searches for “math” or “mathematics” will be for math textbooks, so the volume numbers for these keywords should be considered to be seriously inflated at best.
Step 4: Choose
This is the easy part, only made difficult because Amazon limits you to seven options. But now you have all the data in front of you: (approximately) how many people are searching for each keyword, followed by (approximately) how popular your book has to be to show up in both a global and category-specific search. Generally speaking, you’re looking for keywords that provide “high” numbers in all columns (keeping in mind that books numbered higher on amazon are lower ranked), but which ones you pick depends once again on your current standing and goals. With a little salesmanship and dedication, any book can break the top 200k; in this case, a medium-volume keyword where the first page contains 300k ranked books is gold. On the other hand, if you think you have the wherewithal to break 50k or even 10k with your book, then you’ll want to target higher-volume, more generic and more competitive searches. And of course, don’t be afraid to go back and repeat this process later if your book grows in popularity.
Aside: How important are keywords?
Very important, and also not important at all.
Keywords work as a form of “passive” promotion; you spend a few hours one time setting them up, and they will continue to pay dividends in the form of a steady trickle of exposure without you having to do anything. For a working writer, this is enormously useful, since it frees your time to write your next book instead of wasting it on active promotion. Chosen correctly, a good set of keywords can provide a recurring income that grows and grows with each new book you add to the market.
On the other hand, all a keyword can do is put your book in a list of search results. In order to actually sell a copy, your cover must be better than any of the other options available (although if you look at as many Amazon searches as I have, you’ll realize quickly this isn’t such a tall order), your blurb must draw the reader in, and your sample and reviews must seal the deal. If any of these elements are lacking, the best keyword in the world isn’t going to help you at all. So, while marketing can be interesting in itself, don’t lose sight of what’s important: putting out the best product you possibly can.
That’s it. Good luck!
-W.W.
Follow @metanautics
I really appreciate your breakdown on how to choose key words. What confuses me is the book description affect. I thought words from that & the title act as key words too. Under that assumption, I planned to put words/phrases that didn’t work in my description into my key words. From what you wrote though, it sounds like I should repeat it again in key words. Can you clarify? Thank you.
Tracy — If you want to verify what I said about book descriptions not being used for keywords, just search on Amazon for a somewhat unique phrase from your description (which isn’t in your title or keywords) and see if your book shows up.
I can’t because I haven’t published my book yet. I am in the process of choosing key words and 7 just doesn’t seem like enough. Amazon states that they do use the description in their search function, but other sites state that this isn’t the case. I’m wondering if I should just play it safe and make sure my key words say it all even if they are redundant alongside my description.
Very interesting article about keywords. I’ve been doing a bit of research myself and agree with you about Amazon not using the description.
There was one part you had backwards. I’ve studied a lot of SEO, and the big thing you want is to be a big fish in a small pond. In a lot of your points here, you’ve pointed that out, but you sort of missed on one point, and it’s a point people generally get wrong. You might want to have a few keywords that have a lot of people using them (sort of like being in a category that fits you very well, but you probably will never place well), but in general you want ones that are lower. More specific, fewer people using them. So, if you have one keyword that 500k people used, you’ll end up being a tiny fish in an ocean. It would be like using a search term like “fantasy.” You’ll never place. This is the advantage of multi-word keywords, and why you want to pick keywords that people use moderately. These tend to be keywords used less frequently not only by readers, but also by competitors. You’ll be more likely to place in a meaningful way.
Your goal, as a writer, is not to reach as many eyes as possible, but to reach the RIGHT eyes. Categories and keywords that are too vague or just plain wrong are more likely to hurt you than help. A broad net is never the right way to go. Period. You’re more likely to have readers get spiteful, negatively impacting you business.
Now, other than that, I wanted to thank you for your thoughtful post. I’m going to be saving this. This is a great post, and will help me when it comes time to organize my thoughts and figure out who the heck would actually want to read my crap. I may understand SEO, but it doesn’t mean I good at figuring out keywords…
There’s something I just noticed. When sorting by “New and Popular,” it’s not sorted by sales ranking. I just looked up something where the last one of the first page was ranked over 3k, and the last one of the second page was ranked at less than a thousand, like 400, I think.
Amazon gives a significant search priority to newer books ( < 30 days old), so this may be what you're seeing. Unfortunately nothing about this is exact science, so sometimes the numbers come out a bit fudged. And I agree with you about the less popular keywords as well, especially if your book targets a more niche audience.
Well, any book can cater to a more niche audience (it’s just about finding the perfect keywords that both fit your book and your audience). As a new author (or a new book), I feel that should always be your priority, at least when it comes to categories and keywords on Amazon. There are just too many books out there to try to appeal to a broader audience on Amazon with your books, unless, of course, you’re a bestseller.
This is something I’ve been struggling with, and something I think I’m starting to get straight in my head. I’ve always seen my books in such narrow terms, which means I’ve never been so great at coming up with categories or search terms.
I’m just happy that I think I’ve figured out what categories will be best for my upcoming book. A while back I discovered the dark fantasy category, and I just recently figured out paranormal thriller/suspense. I’ve always thought of thriller as just shy of a horror movie, but all my books fall into that category, really. I think that’s probably the hardest thing, rethinking how you see your stories.
Hi, I value your post as it is systematic and easy to follow. When you are saying “in this case, a medium-volume-keyword where the first page contains 300k ranked books is gold”….. I wonder what does medium-volume-keyword mean to you in terms of monthly searches ? (using Google keyword planner).. Thanks, Mark
I think it depends on the choice of keywords you have available to you to accurately describe your work. I try not to focus on numbers in google because they don’t really mean anything; rather I just pick the best keywords I can think of, then look at the numbers relative to each other. The sales rank on Amazon is a better number to focus on because this can be influenced and understood directly as a function of your sales.