Home » Advertising-Promotion-Marketing, Amazon, Self-Publishing » Half of Amazon Book Sales are Planned Purchases

Half of Amazon Book Sales are Planned Purchases

21 February 2013

From Forbes Blogs:

The Media Briefing’s Digital Media Strategies conference is going on in London this week.

. . . .

[T]his single statistic, that nearly half of Amazon’s book sales come from people who already know what they want and are simply using Amazon as a way to get it, has huge implications.

It’s well known that book lovers are quite happy to spend time in a bricks-and-mortar bookshop to see what sort of interesting titles they find and then go home to buy on Amazon where the prices are often cheaper.

. . . .

Self-publishers often think themselves unaffected by ‘planned search and purchase’, not least because for most of us the idea of having our books in a bookshop feels like a distant dream. The assumption that self-publishers work to is that the key to cracking Amazon is to rank highly in their recommendation algorithm.

. . . .

But McCabe’s statistics show that only a piddling 10 percent of Amazon book choices are made because of its ‘bought this/also bought’ recommendation engine. Bestseller and top 100 lists influence 17 percent of book choices, with 12 percent down to promotions, deals, or low prices. Only 3 percent came through browsing categories. Planned search by author or topic, however, makes up a whopping 48 percent of all book choices.

. . . .

Self-published authors have limited resources for promotion and these figures show that you should focus not on trying to woo Amazon’s algorithm, but on building awareness outside of Amazon.

Link to the rest at Forbes Blogs and thanks to Bridget for the tip.

Advertising-Promotion-Marketing, Amazon, Self-Publishing

39 Comments to “Half of Amazon Book Sales are Planned Purchases”

  1. How would McCabe know?

    • “How would McCabe know?”

      At least one part of this, that a large number of people are using B&M for discovery and then going home to buy online instead of the store is a claim I saw elsewhere a couple weeks ago in this story ( http://www.thebookseller.com/news/barnsley-bookshops-could-charge-browsing.htm ), which originated from a BBC 4 radio show. In it the CEO of HarperCollins UK and International claimed that B&N had announced that 40% of the people who walked in their store did this. The numbers are different and it looks like both these quotes originated in the UK which is interesting. If this claim originates from B&N I would expect there would be other stories out there. I haven’t seen them, but that doesn’t mean they aren’t there.

      • And I see if I would have read the linked article first that it quotes the same person I was talking about. 🙂

        • It’s all anecdotal, though. B&N doesn’t surely know how many of its customers plan to purchase online after visiting the store. And short of customer’s using Amazon’s price-check app, even Amazon doesn’t know how many book orders were from people that had visited bricks and mortar store first prior to ordering online. Amazon can make a fairly accurte guess that a purchase was an impulse buy versus a planned buy by the customer’s behavior at the website, but I don’t think Amazon is sharing that information. How would McCabe even begin to know how many people purchased an “also bought”, which is Amaonz data also?

          • Polling the consumer.

            Furthermore, “also bought” and other internal data is not a very good measure – because it doesn’t actually measure the level of spontaneity. Most of the books in “also bought” lists tend to be known quantities. So, for instance, if I go to Amazon, planning to buy up several Agatha Christie books for my summer vacation, all of the books I planned to buy are likely to be in the “also boughts” making it quicker for me to them through those links. (I’ve done this many times with many authors.)

            Furthermore, if I wasn’t planning on buying a book and came across it in an external link, that would appear, by internal data, to be a planned purchase when it isn’t.

            The only way to know the difference between planned buying and spontaneous purchases is via consumer survey.

            And MacCabe’s data analysis company specializes in seeking out and analyzing multiple studies of this sort. (And you can believe Amazon is using data like theirs as well as internal data in driving their own business.)

      • That’s how it worked with me.

        I would “discover” that the book I was looking for wasn’t in Barnes & Noble.

        So I would go home and order it from Amazon, and it would arrive two days later.

  2. Wow,

    This really pokes a lot of holes in the various schemes to game the Amazon (and everything else) system. I fully agree with this writer’s analysis. Discovery is still an issue for the self-published writer, and trying to manipulate the Amazon algorithms is not necessarily the answer. It may even be contributing to the problem. It’s going to be hard to argue with her numbers, but I’m sure many will try.

    If you haven’t read the comments on the original post you really should.

    “Post hoc ergo proctor hoc.”


  3. Planned search by author OR TOPIC.

    Holy loopholes, Batman. Apparently when I go into B&N the fact that I gravitate to three specific places is just browsing, but if I search by topic on Amazon I’m a deadly Book Ninja who strikes and then withdraws.

    On the plus side, I guess this makes the fact that I sat there and put sixty promo cards in sixty promo books to hand out at a convention seem somewhat less pointless.

  4. From the article: “The title of McCabe’s slide is “Amazon – only the end of the funnel, so far?”, and it’s an important point for self-publishers to take to heart. Amazon is a destination for purchase, the place you funnel your fans to, not a discovery mechanism in and of itself. ”
    Note that somehow “a destination for purchase” got conflated with “the place you funnel your fans” – an indefinite became definite. Those of us without Kindles will not fit through that funnel, and may not become fans who purchase. I have to really want a book before I go search it out from a non-Amazon source it there isn’t a link to an epub version in your blog. And all of Amazon’s discovery tools won’t help me find your book. I don’t look for books at Amazon.

    • That’s a very good point. Not everybody has a Kindle and other e-readers are dominant in other parts of the world. And very few readers are willing to jump through hoops and convert via Calibre. So unless an author is exclusive with Amazon, I don’t understand why they don’t link to all major retailers where their book is available on their site.

      • One problem with linking to every site is the list of links will be longer than the blurb for your book: seven or eight for amazon, forty or fifty for apple, etc. Or do you only link to the US sites and leave the readers in the rest of the world fend for themselves?

  5. Add another point made for the “branding” argument: be prolific and multiply (your sales).

  6. Planned purchases, eh? Well, not long ago I was in town and because I’m ever hopeful, I dropped into B&N to see if they had a particular book of clip art I wanted. B&N used to carry a fab selection of Dover clip art books. This time they had two Dover books. Only two, and they weren’t even clip art books. I asked a clerk who happened to be passing by what had happened to the Dover books. He’d never heard of Dover books and treated me as if I were making stuff up just to annoy him. The customer service kiosk was less than twenty feet away. He could have said, “I can look that up for you.” Instead he made it clear he had more important things to do than mess around with me about a mythical publisher. By then I was so irritated by the wasted trip, I didn’t bother to browse. I left.

    I went home and ordered the book from Amazon. While I was there, I saw two more clip art books that looked very interesting and ordered them, too, and as long as I was shopping, I bought one of the books off my wish list. While I was doing that, I spotted an interesting looking title in the “c’mon, you know you want it” scroll. That one turned out to be not quite right, but in another scroll there was one that looked really good, so I bought it, too.

    Hey, book sellers, here’s a novel idea. Give me something to discover. I’ll discover it, trust me.

    • It’s probably be too dangerous then to tell you about mythical Dover’s own mythical website ( http://www.doverpublications.com ) where you can e-browse the interiors of hundreds of their clip art books (and e-clipart CD ROMS) and see if a specific volume has what you’re looking for. You might spend your way into clip art heaven (or at least into the poor house). 🙂

    • Jaye, this reminds me of my most-recent bookstore experience. I went to a local B&N looking for 4 items (3 books and a CD). The store only had ONE of these items. This particularly surprised me because 2 of the books they did NOT have in stock had been released within the past year and were selling reasonably well, so it hadn’t occurred to me that they WOULDN’T be available in a bookstore with 30,000 sq ft of space.

      Anyhow, in the case of the 3 items I sought which the store did NOT have, I had an alternate item in mind for each one (I was buying gifts), equally acceptable to me. But the store didn’t have any of those, either. Since I needed all the items for that weekend and had left my shopping until the last minute, I soldiered on until I found enough items to buy. But I left the store fully satisfied with only ONE of the FOUR items I’d bought, and I’d spent more money than expected because nothing was discounted, AND I’d lost at least 30 minutes of my life I’ll never get back.

      I realized this is -exactly- why I’ve mostly become an online shopper in the past decade, -especially- in circumstances where I know what I want. (And a lot of people seems to have made the same choice; the store had very few customers while I was there.) I was annoyed with myself for not planning ahead (i.e. leaving enough time for shipping), because if I had shopped online, then (I know from consistent experience) I could have gotten (a) exactly what I wanted, (b) at a much better price, (c) with far less time and effort expended.

      So one thing not being explored in this article is that online shopping seems to be the better choice when you KNOW what you want, because you know you’ll get it, whereas going to bookstores when you know what you want is more of a time-consuming crap-shoot. Bookstores are good for browsing to see what’s there that you hadn’t expected to find and would like to try, but online shopping is much, much better when you know what you’re after and just want to get it.

  7. For me, nearly 100% of my book purchases on Amazon are planned purchases. I never go there just to browse. My TBR stack is always overloaded, so there’s no need.

    One thing that has changed for me in the last 6 months is that I now look at reader reviews on Goodreads rather than Amazon. Goodreads has so much more information. An example: Code Name Verity, a book I read recently, has 147 reviews on Amazon, but over TWO THOUSAND on Goodreads. In addition to checking out reviews before I buy, I also like to look at reviews after I’ve read the book, just to see if other people reacted to it the way I did. I used to do this on Amazon, but now I do it on Goodreads.

  8. This is something I’ve been saying for a very long time. People buy books they’ve already heard of by authors they’ve already heard of.

    The way they hear of them is word of mouth. In many cases, they don’t even realize how they’ve heard of them. That is, word of mouth isn’t just friends says “Wow, you’ve got to read this book!” It’s also friends just having the physical book on the coffee table in their house, and strangers reading it on the subway. It’s bloggers mentioning the book, maybe even in the context of talking about a different book. Small things accumulate to bring a book and author into a reader’s consciousness.

    And the most powerful tool for that kind of exposure is Google. Google makes connections and puts related items in front of every searcher:

    Here’s an interesting thing I discovered the other day. I went and searched on my name on Google images. I not only found a bunch of my book covers and pictures I’ve posted on my blog, there tons of pictures that had no direct relation to me. For instance, pictures from blogs where I had commented — even though my name was nowhere to be seen on the page where that picture was posted.

    So that means that those blogs and people who are associated with me, benefit in some small way from any attention I get. (And maybe benefit from any attention family members of mine get — because if you’re associated with me, you’re associated with “LaGuire”)

    But you don’t get that by flogging for it. It’s something that happens naturally and organically. It takes time.

  9. Every day I look at Amazon’s digital deals. Every day there are digital books marked down to $1.99 or $.99 and while many days when I read the sample chapters I am not always impressed. Still I have bought many books this way and have a great collection of books on my iphone and laptop (which is where I mostly read my ebooks.)

    As far as Barnes and Noble goes, I can buy on line from their store from Amazon with my iphone, but I seldom bother to go to B&N any more since Amazon digital deals have given me an archive of close to 500 ebooks. Further my wife and family have pushed me to give away almost all my physical books since I seldom read them any more. I also have a nook account, Calibre software on my laptop and ibooks on my iphone. But mostly I stick with Amazon ebooks.

    • I do this too, although I do it via a Pixel of Ink subscription. I don’t use the Pixel of Ink emails to browse for new books. I use them to spot sales on books I already know I want to buy. Sometimes this means picking up an inexpensive digital copy of an especially-loved book I already own in print, or it’ll be a book or author I’ve been hearing good things about for a while and have been ready to try out.

      I think the downside for publishers re: all these on-sale ebooks is that I pick up so many inexpensive ebooks that I seldom need to pay full price for anything. My Kindle is already loaded with more books than I have time to read. But I do pay full price when one of my must-read authors releases a new book.

  10. Thank you Josh!

    Amazon does not release figures, so McCabe has no ground to stand on here.

    I’m going to copy/paste that statement so I don’t have to keep re-typing it for all these articles that ‘consultants’ write without any actual data. What they are really offering is an opinion.

    There is also a subtle anti-Amazon slant here that I think Indies should watch out for. The idea is that Amazon’s lists are not necessary, and Amazon doesn’t add anything of value.

    And what is that last line about giving Amazon the two fingers….?

    As for what sells – well, I have no data, but I would guess there must be some significant percentage of people buying from lists and recommendations or Amazon would not be putting all that energy into it.

    And – either way as an indie author – I think it makes sense to spread your marketing efforts and not rely on just one factor.

    But again, I state – the article is not based on actual fact. It’s just an opinion. An anti-Amazon opinion.

    • We’re not talking about some publishing pundit here, or uninformed journalist (or worse, some indie publishing guru who only half understands the headlines he reads and then propounds “Truth” from them).

      This is the COO of a data analysis company. It’s their business to gather data. They do hundreds of consumer studies — talking to the user, who is the only one who really knows what drives them to the site.

      As we learned in the last election, don’t diss the number crunchers. Instead, turn around and question your assumptions.

      • Camille, I have no assumptions here at all. I am pointing out the fact that if Amazon has not released the data, then no one has the data.

        Consumer surveys are not a reliable or accurate way to gather information. The company can act like this is valuable data, but it is not. Job titles and corporate ‘missions’ do not give legitimacy to data.

        They can have a ‘guess’ based on anecdotal evidence but that is all they can have until if/when Amazon releases their data. Otherwise it is just an guess.

      • It’s reasonable to ask about the study methodology no matter who he is.

    • Mira wrote:
      “As for what sells – well, I have no data, but I would guess there must be some significant percentage of people buying from lists and recommendations or Amazon would not be putting all that energy into it.”

      This is exactly right.
      If you want to know what works, just watch and see what Amazon puts their energy towards.
      I can’t count how many times I’ve read that free books hurt sales, and yet, Amazon makes it possible for many authors to offer their books for free, particularly the first book in a series.
      Ignore the haters and the “experts” and keep a close eye on what Amazon does. If they’re doing something, they’re doing it because it works and makes them money.

      • I agree, Terrence.

        Donald – exactly. Amazon is brilliant. If they are putting huge amounts of energy into something, it’s probably worth it.

        • Not to put words in your mouth Mira, but the last two words on the end of that sentence should be “for Amazon.”

          Another commenter said he had over 500 titles on his Kindle thanks to the Kindle Daily Deal. I’m sure Amazon loves that too. I would say he’s well anchored to them, which is their goal all along. Not necessarily a bad thing IF the relationship remains symbiotic. But if that should change…

          If there is something that Amazon is putting a lot of energy into (read money/time) and its working out in favor of the indie author that’s great. I’m all for it. But you should never forget that it always works out for Amazon FIRST.

          Amazons end goal and your end goal are not always the same.

  11. I read this, and I’m scratching my head, because the conclusions reached by the Forbes writer don’t conform to my direct personal experience. I think she is misinterpreting the statistics. Please bear with me while I elaborate:

    When I published my debut thriller, HUNTER, in June 2011, I already had something of a “brand,” having been a nonfiction writer for decades. I also tried to enhance my brand by expanding my online “platforms.” I launched a catchy blog, “The Vigilante Author,” meant to position me in the minds of thriller readers who like that kind of fiction. I did a lot of social media: Facebook, interviews with other bloggers, etc. HUNTER received several nice, if temporary, sales surges when a few big blogs mentioned it.

    But I also did promotions internally on Amazon. I encouraged happy readers to leave customer reviews there and on various online sales sites. And I worked to craft a compelling product description, so that once the curious went to its product page, they’d get a strong pitch to close the sale.

    Overall, the novel did pretty well — though not spectacularly — for the first five months, selling 4,000 copies at an ebook price of $3.99. But it wasn’t only my external efforts bringing it attention. HUNTER ranked high on a couple of Amazon subcategory bestseller lists, and right near the top of the “customer rating” lists. It also quickly garnered a lot of very positive “customer reviews.”

    Yet all my external promotional efforts resulted in utterly paltry sales on Barne & Noble’s Nook, on Smashwords, and on iTunes. If promotion external to them all was a critical sales factor, then why weren’t the sales spread more evenly among all the online retailers?

    Then Amazon entered it into a week-long, post-Thanksgiving sales promotion, along with scores of other books, lowering the price temporarily (with my consent) from $3.99 to $1.99. Simultaneously, from those scores of titles, they also selected HUNTER as one of their eighteen top “Editors’ Picks.”

    As a result, the book was showcased boldly and prominently at the very top of about 4-5 different Amazon Kindle pages. You couldn’t go to the Kindle main home page, the “Editors’ Pick” page, the sales promo page, the “Movers and Shakers” page, or various category “recommendation” pages, without my book’s cover staring you in the face. It also began appearing on a lot of “also bought” recommendation lists for other bestselling books in the same categories.

    Overnight, sales exploded, from a few dozen per day to thousands per day. By the last day of the sale, it was the #4 bestselling product on Kindle, and hit #1 in the “Mysteries and Thrillers” category (and thus numerous subcategories), as well as #1 in “Romantic Suspense.”

    It’s interesting to note what happened in the days, weeks, and months afterward. Once the price went back up to $3.99 and HUNTER was no longer featured prominently on those various promotional pages, sales began to fall. However, it took a month for the book to fall out of the “Top 100.” During those four weeks, it was still highly positioned (and thus “discoverable”) to millions of customers browsing various Amazon category bestseller lists, and it continued to sell hundreds of copies per day.

    In the months following, HUNTER sales continued to slide slowly back to its pre-promotional levels, similar to what it had been selling right after I published it. But by then, it was far less “visible” — thus discoverable — on Amazon…even though by then my outside promotions and name-recognition had never been better.

    Face it: Those sudden, massive sales (over 50,000 copies in just 35 days) could not have occurred because most purchasers were going to Amazon specifically looking for my book. The huge, months-long sales spike came entirely from what was happening internally on Amazon: visibility generated via its promotions, bestseller-list placements, recommendation system, and algorithms.

    I know for a fact that other authors participating in that same promotion also saw significant spikes in sales — though not as great as mine. And the only difference among us was that their works were not featured as prominently, or on as many Amazon pages. We see this same phenomenon occur every single day, with the “Kindle Daily Deals,” where a product/book is singled out for spotlight attention and soars right to the top of the Kindle bestseller list.

    So, I’m not sure what McCabe’s “percentage” statistics are measuring. I would be willing to believe that most book titles, which suffer in relative obscurity, may be purchased roughly in the proportions and for the reasons that his pie chart suggests. And, based on personal experience, I do believe that an author’s “outside” promotional efforts — if done prudently, applying sound marketing principles — can generate solid, ongoing sales. Yes, for sales success, I think it’s very important for an author to distinguish himself from the pack by establishing a unique “brand” by marketing methods external to Amazon.

    But I think the WRONG message to take away from McCabe’s statistics is: Amazon promos, bestseller lists, staff picks, “also bought” recommendations, and categories don’t really matter as much as does an author’s outside promotions. Or, as the Forbes writer puts it: “People are simply not browsing for books based on Amazon’s recommendations, not in any significant numbers.” Huh? The pie chart shows that 52% of Amazon book buyers are making their selections precisely because of Amazon’s internal recommendation processes. The writer’s error lies in how she interprets the pie chart percentages.

    The fact that 48% of Amazon’s book customers go there with a planned purchase in mind, does not mean that 48% of a given author’s buyers will have gone to Amazon specifically to buy his or her book — or that only 12% of his sales will have come from Amazon promotions, etc. As every bestselling Amazon author knows, a huge percentage of his or her sales can come from Amazon’s internal promotional methods; external promotions may have only kickstarted the process of getting under Amazon’s spotlight. So, if your book is selected for special promotion to “only” the 12% of Amazon’s book buyers who pay attention to such promotions…well, 12% of Amazon’s millions of book buyers is an enormous market.

    Given a choice between having a Facebook presence, or having my next book selected as a “Kindle Daily Deal,” it wouldn’t be a contest for me. Anyone disagree?

    • Here’s the other thing to remember: the audience for books is HUGE. You can do very very well catering to just a small portion of the audience — such as those who use promotions and best seller lists to browse for books.

      That doesn’t negate the fact that this is a subset of the overall audience. It’s just that the larger audience is harder to reach.

    • That’s a very good analysis. However, may I point out that it is not possible to control Amazon’s editors choice to make one’s book an Editor’s Choice or to bestow upon one’s book some other editorial honor–that’s beyond an author’s control.

      • …and as soon as it happens the price of the book gets lowered to impulse-buy territory and the whole thing makes you an outlier. Not that that’s a bad thing, but it is what it is.

        Regardless, there are never enough numbers to make the argument one wishs to make. I do enjoy seeing everyones own spin on these articles though.

  12. As a reader I buy probably 90% of books as a planned purchase. The others are an “also bought” or similar recommendation that looks interesting enough for me to click on and read about and then buy on an impulse. I keep a running list of authors and/or books I want to check out, and so when I’m in the mood to read or have a bit of book spending money, I’ll pick up several. For pub-house authors, I keep an eye out for sales and grab those books up on promo, whether it’s a publisher-driven promo or a daily deal or monthly deal. In my experience discovery happens elsewhere than the point of sale most of the time, for the kind of reader that I am. I see a review on a blog, hear a friend talking about it, read a comparison to the book, encounter an author talking about what they’re doing, or someone else talking about what they’re doing, and if it interests me I note the name, and go look at it when i’m in the mood for that kind of book. I agree with the point above about many data points and references being key to perception of your name and ‘brand.’

  13. Wow. Talk about completely missing the point.

    These statistics don’t mean at all what this article suggests. Let’s start with the title of the slide in question:

    “Book choosing methods on Amazon.com, May 2012”

    Evidently, we’re dealing with data from a snapshot survey of Amazon.com customers from May 2012. I have a lot of questions about this, but for the sake of this argument, I am going to assume that these figures are valid. Come, let us reason together about what the data means.

    Issue #1. The ecological fallacy.

    In the simplest terms and for the question at hand, this means making inferences about individual book titles based on the aggregate statistics. The original poster makes this mistake when she says:

    Self-published authors have limited resources for promotion and these figures show that you should focus not on trying to woo Amazon’s algorithm, but on building awareness outside of Amazon.

    Whether or not that is good advice, there is simply no support for that statement in the data. The figures most certainly do not show that. Think it through. This survey almost certainly asked people if they bought a book from Amazon.com in May 2012. If they answer yes, the follow-up question(s) are about the last book they purchased through Amazon. That’s standard practice in this type of survey.

    With that sort of question, the survey most certainly did not get a representative sample of all titles sold at Amazon. It was trying to get a representative sample of all books sold at Amazon. The difference is critical. Generally, the top 10 bestsellers make up anywhere between 50% and 90% of the sales for any month (and it is usually towards the high end). This survey mostly tells us how people find bestsellers at Amazon. In that respect, the “planned search” + “best seller lists” categories seem a bit low (65% of the total).

    For the authors here, the question I have for you is:

    Does it help you in any way to know how people find best sellers?

  14. Issue #2. The Amazing Amazon Algorithm

    If 10% of Amazon’s total book sales come from the also-bot (i.e. the algorithm that recommends items based on what other the buying habits of other customers who bought the book you just bought), that is biggest publishing industry news I’ve ever heard.

    I happened to find book sales stats for May, 2012 (http://www.mediabistro.com/galleycat/category/sales-stats )

    Leaving aside the University press ebooks, the other categories add up to a total sales figure of $479.8 million. In May 2012, Amazon had somewhere between 30 and 50% of the total sales in those categories.

    Which, if you do the math, means that the alsobot generated at least somewhere between $7 million and $12 million of sales in a single month.

  15. FYI, the data cited in the Forbes piece (and thank you PG for posting it!) has prompted a long and somewhat contentious discussion over on Kindle Boards: http://www.kindleboards.com/index.php/topic,143088.0.html

    A lot of the points raised there have been raised here in the comments, as well. I can add that the data in question apparently come from the Codex Group, a research firm specializing in book discoverability — if you look down at the very bottom right corner of McCabe’s slide, you’ll see the attribution.

    Personally, I’m sort of disturbed that the Forbes writer didn’t get a comment from either McCabe or the Codex Group for this article. I think we need some help in understanding these data at a more fine-grained level. (For instance, how much of the 48% is made up of “author” searches and how much of “topic” searches. And what about title searches? Are they in there as well?) I’d also like to know how the data were collected and whether anything’s changed since May of last year.

    • Becca – good questions!

      It’s really important, I think, for authors to question things. Without that, they are open to being manipulated.

Sorry, the comment form is closed at this time.