Gaining Traction in the Amazon Ebook Marketplace

By Carolyn McCray, Author | @craftycmc

For all but a select few household-named authors, the days of launching a book, promoting it for six weeks, and then moving on are over. But do not despair! If we embrace this new 24/7/365 paradigm, we can actually increase our bottom line. Yes, the eBook market is extremely crowded with cheap product, however, if you understand how Amazon and the other digital booksellers determine which titles to put into their internal recommendation queues (“Customers Who Bought This Item… etc.), you can leverage that information into greater sales for your eBooks.

This article is broken into three parts.

The first article explores how Amazon et al weigh sales and determine your ranking and the number of times they will put your title out into their internal recommendation queue. The second article discusses “Price Pulsing” and why it is an essential strategy for prolonged digital book sales. The third article discusses “Sales Nodes” and why they are vital to leveraging “Price Pulsing” and all other sales strategies.

Forgive me but for this first article I must discuss some fairly geeky math to help you understand the basics of Amazon’s sales matrix so that you can understand how then to leverage it for greater sales.

First I must state that Amazon will not release the exact matrix it uses to calculate ranking or how it determines which titles to propagate out into the internal recommendation queue, however after nearly a year of analyzing the cause and effect of external advertising campaigns on eBook sales I feel very comfortable making some basic assumptions regarding this vital mechanism.

It appears Amazon’s matrix is looking to pick out the titles that are most likely to sell the most books and puts those titles out into its internal recommendation system the most frequently. Which makes sense. A book selling well now, will sell well if they give it more exposure. The question becomes though, how does it estimate which books are the most likely to sell?

This is where anecdotal research comes in handy. It is widely accepted that Amazon rewards you for hourly, daily, weekly, and monthly sales. It also appears that Amazon weights the monthly sales the heaviest, then weekly, then daily. It weights hourly sales the least.

Clearly the system is set up to protect against flash-in-the-pan type sales. Just because you sell a hundred books in an hour doesn’t guarantee you will even break into the top 1,000, let alone the top 100.

So the first thing we must accept is that the Amazon ranking system is complex, taking in sales from a minimum one month to establish your current ranking and therefore the number of times it puts your book into the internal recommendations queues.

An example. I had two books in a major social media event, the Labor Day Book Blowout, in which all 150 titles were discounted to 99¢ for the Labor Day weekend.

Book 1 –HeartsBlood had been on Amazon for six months and normally hovered in the 5,000-10,000 range. The title normally sells about 8-14 books per day. On the day of the event, HeartsBlood peaked at #1,113 with a total of 39 books sold for the day.

Book 2 – All Hallow’s Eve: The One Night it is BAD to be Good was launched the night before the event and therefore had no ranking or sales before the event. The title sold 69 books the day of the event and peaked at #1,089.

As you can see, two books in exactly the same event ended up with very similar rankings yet sold vastly different numbers of copies (nearly double, in the case of All Hallow’s Eve). It took far fewer copies sold of HeartsBlood to drive her up the rankings.

I could repeat this example over and over again during events. It is clear that previous sales impact hourly rankings. I term what happened with HeartsBlood as “traction.” The more sales a book has over the last month, week, and day provides a built-in number of internal recommendation berths, thereby giving your title some sales traction. This traction is also reflected in higher rankings.

However this traction fades and even vanishes after certain milestone segments of time. Why? Because directly after a large event (book launch, blog tour, sales event) where you have larger than normal sales, Amazon’s matrix sees those sales and seems to say, “Okay, this book is a great seller so I am going to increase the number of times I put this title out into the internal recommendation queues.”

Because of this increase in recommendation queues appearances, your title will continue to sell more books. However usually not as many as you did on during that large event, so the next day when Amazon’s matrix calculates your sales, it realizes maybe you weren’t the incredible seller it thought you were the day before and decreases your internal recommendation berths. Which in turn decreases your sales and ranking.

After that first day drop-off (which is fairly minor), the matrix keeps you in a relatively higher recommendation berth rotation for a week. If your title fails to live up to last week’s major sales burst, after seven days, the matrix will downgrade the number of internal recommendation berths and therefore your sales and your ranking.

This one-week post-high-sales drop-off is palpable. I term it “unlatching.” Your sales and ranking normally take a steep decline after this one-week unlatching, dropping as much as several hundred to several thousand in ranking spots (based on where you were to start).

However, after this drop you will normally hold in the same recommendation rotation for a full thirty days from the event. After that, sadly, if your sales do not live up to that event’s much higher  sales record, the matrix will fully unlatch from your title and you will experience a significant drop in sales and rankings, up to thousands if not tens of thousands of lost ranking spots.

At the one-month mark, many times a title will go into free-fall, dropping hundreds of rankings per hour until it settles to a grossly lower ranking.

For most publishers/authors this free-fall and steep drop-off in sales are a mystery. Why was it doing so well yesterday, and now the eBook is in the 60,000 range? The answer? You lost traction in the rankings (those month-old sales are now stale and not used to calculate your ranking) and Amazon “unlatched” your title from the internal recommendation queue, decreasing your internal sales.

Again, though, no need to despair. Knowing how Amazon calculates sales and rankings allows you to plan ahead and create promotional strategies that maximize this effect.

This is what we will be covering in the next article “Price Pulsing to Maximize Long-Term Sales.”

I also strongly urge everyone to review my other articles but especially regarding “Best Practices for Amazon Sales.” In the next article I will be referencing many of the concepts in that piece.

Again, as always I will be subscribing to the comments below so feel free to leave a question or feedback and I will try to get to it within twenty four hours.

Carolyn McCray is a social media and sales consultant to writers and publishing houses alike. And using the principle laid out in this article, her recent non-fiction book, “Dollars & Sense: The Definitive Guide to Self-publishing Success” debuted at #1 on the Amazon Bestselling list for Study & Teaching and reached #2 on the Authorship Bestselling list beating out such rock stars as JA Konrath and Zoe Winters. Carolyn is also the founder of the Indie Book Collective, an organization dedicated to helping writers utilize social media to sell their books.

26 thoughts on “Gaining Traction in the Amazon Ebook Marketplace

  1. Christopher Wills

    Fascinating stuff; thanks for this; it is perfectly timed as my second book is due out any week now.

    I’d always thought the rankings were a straight hourly sales number; but I suppose if they were the rankings would be all over the place and publishers could manipulate them by mass buying for an hour immediately before everyone goes on the net.

    I look forward to learning how to use this knowledge; presumably it involves periodic promotions of some sort.

  2. Kristie Cook

    This is great info! I made the trip you described: going from 200s of all paid to huge nosedives. At least now I have a better understanding of why. Question: Amazon recently changed its algorithm over the summer. Have you noticed how it might change anything you’ve learned so far?

    1. Carolyn McCray

      No, even though the algorithm changed, it only tightened things down even more. Meaning it feels like you fall faster and maybe a little more quickly and if anything they are weighing toward long term sales even more.

      They have REALLY tightened the bottle neck at the 1,000, 500, and 100 markings. It feels like they do this by not rewarding you with as many “Earned” slots in the internal recommendation queue as you approach these marks.

      Between 5,000 to 1,200 it feels like for every 4-5 sales, you get 1 extra berth in the queue. However once you hit 1,100 it suddenly takes 7-8 sales for 1 extra berth. It feels like it stay there until around 600 then it become 10 to 1 extra berth. At 200? Dear Lordie, about 12-14 to 1.

      This is why when you are high up there, it is hard to keep it there unless Amazon is promoting you internally (indie store, sunshine deals, etc).

      Hope this helps 🙂

  3. Bill Sadler

    I love your articles and have copied them into a separate folder to review before the launch of each book.

    However, your writing is pocked with grammatical mistakes. Everyone who graduated from high school knows that it is \fewer copies\ not \less.\ \Household named authors\ requires a hyphen between the first two words. Etc., etc.,

    You should have an educated friend review your work before you publish. Your superior knowledge loses credibility when you present it as if you were writing a teenage blog.

        1. Allan Wallace

          Thanks for the material, it’s answering many questions. I’ll suffer a few errors gladly rather than have you slow down your output.

          *enjoy your writing*

          As a commenter said on one of my posts, “If we reward great storytelling vs. great grammar, I believe that the great storytellers will eventually improve their grammar. On the other hand, if we reward great grammar, I doubt that great storytelling will follow.” nuestraherencia

    1. Carolyn McCray

      Yes, but more than likely a WEIGHTED moving average. And only Amazon truly knows the exact formulation and they are not likely to give it out.

      What we can make is some educated assumptions that play out consistently on the platform so that we can build strategies (adaptable of course, what works this month may not work next month) to maximize sales 🙂

      1. Tom Ciolli

        Fantastic series about amazon. There is a certain irony in that I discovered it through a PubIt! link from Barnes & Noble. The Pubit links on Facebook are much more useful and marketing oriented than the Amazon Kindle’s posts which try to be kind of frivolous and fun. I think today’s Amazon post was which authors would I put on Mt. Rushmore. Too bad Amazon doesn’t publish more article like yours.

        What’s the easiset way to find your latest articles?

        1. Carolyn McCray

          Funny how that works isn’t it?
          The best way to get my articles? Subscribe to the DBW newsletter! Super easy.
          You can also just search the site under Carolyn McCray and all of my articles will pop up!

  4. Lulu Dean

    When I started reading your article, I must admit that I felt a little confused. I mean, weight? Yes, I’ve put on a few pounds, but do we have to talk about that right now? Price pulsing? Is this the tingly sensation I feel in my pocketbook when they mark down the plus-size jeggings? And sales nodes? Well, I love a good bargain just as much as anyone, so sign me up!

    Seriously, Carolyn, I’m sold on your sage advice. Loved Dollars & Sense and have recommended it to several friends. And now that my collection of humor essays, Her Royal Thighness and the Mannequins of Doom, has launched, I’ll be stopping by to read your columns here on a regular basis!

    As a newcomer, it’s so refreshing to discover talented, passionate and knowledgeable professionals who are willing to share their experience, information and suggestions.

    1. Carolyn McCray

      Ah, thank you so much!

      I am so glad that I could help and entertain you along the way!

      Also if anyone is going to be in Guadalajara next month for the FIL book fair, I am speaking on a panel regarding Meta-Data. It should be a barn burner! LOL

      Again, thank you so much for your kind words and I wish you the best of luck with your book!

  5. Amanda Ball

    This is another great post thanks Carolyn. After following the tips you gave in your other articles and the advice you gave me personally in the comment section, I am now selling Kindle copies of my books in UK and USA every day. Not enough copies to retire just yet but the reviews are building as is my confidence.

    Thanks for more great info, I’m eager for the next articles now 🙂

    1. Carolyn McCray

      I am so glad the advice held up #again 🙂

      And yes, we all start somewhere. I remember when I was just excited to get any royalties at all. Then I was happy they were in the thousands and now they have replaced my salary as a veterinarian.

      Time, persistence and product = a great writing career

  6. Shel Delisle

    Hi Carolyn,

    I’m a newbie to digital publishing and just stumbled across this very helpful article. One question: Can you explain how \berths\ work? Or what exactly a \berth\ is? Maybe that’s covered in another article and I just need a different link? Sorry. That’s actually three questions instead of one, but kind of related. 🙂

    Thanks again. And I just subscribed so hopefully I’ll begin to understand exactly how Amazon works.

    1. Carolyn McCray

      Since readers have limited attention spans, we can only promote so many books to them at a time. In your backmatter (and Amazon Product Description) this usually limits you to at max of 4-5 books. Each spot that a book takes up is considered a berth.

      Let’s say you have 3 books out.
      Inside of your first book, you would promote your other two backlist titles. That would leave you with 2-3 open berths. Those berths would go to other indie authors that have books in your same sub-genre (or if you are a publishing house, it would be 2-3 titles by different authors in your library that are in similar sub-genres).

      And yes, I would recommend that you search under Carolyn McCray on the site to find my archived articles 🙂

      Also, I get asked this a lot so I thought I would just put it down here… I am in the Seattle area now so if anyone is local, I have a Meetup group: join up and join in 😉

  7. Ryan Notch

    Finally, the article I have been looking for. My book dropped off rankings dramatically after climbing for a while and I just could not figure out why. Really I’ve been trying to figure this out for months. I feel like Amazon is the classic black box phenomenon in science, except I don’t even know what inputs are going into the thing. Now I at least have some clues.

    Thanks so much for posting this, you are just awesome. If I may ask, I purchased a system by a person who claimed that Amazon uses backlinks leading to the ebook from around the web to determine its relevance page ranking. But my testing with backlinks doesn’t seem to confirm this. What is your opinion on it?

  8. Bruce McMinn

    This is great information which I hope will position my works for higher discoverability on Amazon. My first novel is doing moderately well but my second which won an award is doing poorly and this information will (hopefully) help.
    You use phrases elsewhere, such as “above the jump” and I have what that means. Maybe you could point us newbies to a place, or article which explains common phrases.
    Thanks for your very useful advice.

    1. Carolyn McCray

      Yes, above or after the jump refers to the row of “alsobot” recommendations (that line of pretty covers).

      While we love being in the alsbot queue, that queue is distracting to the reader so our Product Description must drive them past it to a sale 🙂

      Hopefully that helps!

  9. Susan Fleet

    Carolyn, has anyone else complained that their historical sales data (viewed on their Author Central page) has not updated since Feb 1? My page has this statement at the top
    Sales Rank Updates Delayed: Your historical Sales Rank data has not been updated since February 1, 2013 while we conduct necessary systems changes to improve this service. We apologize for the inconvenience. Once rebuilt, your historic Sales Rank data will include the missing data since February 1, 2013.

    Meanwhile my sales, and sales rankings, for all 3 novels have gone in the toilet. January sales totaled 600 +. Feb sales dropped to 332. March is dismal. Do you have any idea what’s going on, Carolyn?

    1. Carolyn McCray

      As to the data update. I know from my Amazon Pub sales area, there was some “overreporting” on Feb 8th and 9th and they made a correction, however everything is updating fine so I’m not sure what’s going on with yours. Have you emailed KDP to inquire?

      As to your sales falling off, I’m afraid that is simply erosion. As I’ve noted before, there is no steady sales state on Amazon. You are either riding high, treading water, or drowning. Unless you do something to “recharge” your sales numbers into the algorithm, it will downgrade your appearance in the alsobots and your sales will degrade accordingly.

      Hope this is helpful 🙂

  10. Susan Fleet

    Hi Carolyn, thanks for the quick response. It’s not my KDP data that I refer to, it’s the historical data on my Author Central page. Yes, I’ve emailed them 3 times. Each time they apologize and tell me the \engineers\ are aware of the problem and they are working on it. (for 45 days?? and there’s still a problem?) The last query I sent they pretty much told me not to bother them with the problem, they’ll let me know when it’s fixed. Of course, I will know instantly when it’s fixed because I’m checking the data every day now.

    BTW, I want to thank you for your previous excellent articles. Based upon your advice, I updated my book descriptions and now that I have 3 books in a series, I switched out of one category and put them all in the \series\ category.

    Question: isn’t \historical sales data\ part of the algorithm for the Sales Rank?

    1. Carolyn McCray

      it feels like the historical sales data the algorithm pulls from is directly from KDP. If your book sales are logging into your KDP dashboard/reports then your ranking is probably fine. Trust me it if wasn’t you would have fallen into the abyss pretty quickly 🙂
      Hopefully your Author Central page will catch up.



Your email address will not be published. Required fields are marked *