How Amazon and Barnes & Noble Price Changes Affected Sales of HarperCollins Ebooks

Expert publishing blog opinions are solely those of the blogger and not necessarily endorsed by DBW.

On Sept. 11 and 12, new contracts went into place between HarperCollins and ebook retailers like Amazon and Barnes & Noble. The new contracts, instigated by a settlement between HarperCollins and the Department of Justice over the issue of ebook price-fixing, allowed Amazon and others to set the prices on ebooks.

The moment it was possible, discounting began on HarperCollins ebooks. Now that a few months have passed, we’ve taken a look at the data to see what, if any, effect the price changes had.

In the following five examples, I found that price decreases didn’t have much effect on sales ranking. If an ebook’s price is lowered from $10 to $8, for example, it needs to sell about 25% more copies to maintain the same level of revenue. I didn’t observe any sales rank bumps that would indicate a 25% bump in sales. Ideally, a title has more than a 25% increase in sales, justifying a price change as better for business overall.

All data below is provided by Iobyte Solutions and represent sales ranking data and not actual sales (although we are very confident that sales rankings are a suitable proxy for sales). The sales rankings for the below are all on Barnes & Noble’s Nook platform, which I chose to use because more data was available than on other platforms. I chose the titles below because they are exemplary of much of the changes I saw when exploring sales ranking data from Iobyte. Iobyte Solutions has much more data on these titles and others.

In the cases below, Amazon’s price changes prompted Barnes & Noble to change its own prices, I believe.

I’ll be presenting the data below and much more during a webcast with the Book Industry Study Group. Click here to learn more and register.

Similarly, we will be diving deeply into best-seller list insights from this data at Digital Book World 2013.

In the data below, I discuss how a X% decrease in price would require a Y% increase in unit sales to result in the same revenue. Just to be clear, the Y% increase needs to be above what sales would have been without a price decrease. So, it’s largely a theoretical discussion — but one that has impact on the bottom line.

The Rise of Nine by Pittacus Lore (HarperCollins, on sale 8/21/12)

After debuting at No. 6 on the Nook ebook best-seller list, The Rise of Nine had two price decreases which both seemed to have minimal impact on sales ranking. The first was a decrease of about 20% and the second was a 10% decrease. After the first price decrease, a 25% bump in sales would result in the same amount of revenue and after the second, a roughly 11% increase in sales would even everything out.

Though it should go without saying, on the chart below a lower number is better (it is better to be ranked No. 1 on a best-seller list than No. 20); this is true of all the charts below.

The Ugly Duchess, Eloisa James (HarperCollins, on sale 8/28/12)

After debuting at No. 11 on the Nook ebook best-seller list, The Ugly Duchess‘s sales rank degraded typically for a new title. Its price was lowered by about 14%, which would require about a 16.5% increase in sales to result in the same amount of revenue.

The Fallen Angel by Daniel Silva (HarperCollins, on sale 7/17/12)

After debuting at No. 8 on the Nook best-seller list, sales rank for The Fallen Angel degraded typically for most new titles. Its price was dropped in late July when it hit the New York Times best-seller list — a stipulation in the Apple agency pricing contracts. The title’s sales ranking continued to degrade, for the most part.

The first price drop was from $14.99 to $12.99, roughly a 13% decrease — requiring a 15.5% increase in sales to validate.

The second was from $12.99 to $9.99, roughly a 23% decrease — requiring a 30% increase in sales.

Divergent by Veronica Roth (HarperCollins, on sale 5/3/11)

Unlike the titles above, Divergent wasn’t released over the summer. Nevertheless, its price was dropped by 27% in early Sept. and then again by about 4.5% in early Oct. The first price drop had some effect on sales rank. The second seems to have had little effect.

State of Wonder by Ann Patchett (HarperCollins, on sale 6/7/11)

For this last title, the price was dropped rather precipitously to $9.35 from $12.99 (28%) and it seems to have resulted in a rather severe sales ranking improvement. However, the difference in unit sales between a book ranked around 900 and one ranked around 400 is not huge. While the title has fluctuated greatly in the period since, it’s in a zone where we can reasonably assume that sales have basically held steady.



12 thoughts on “How Amazon and Barnes & Noble Price Changes Affected Sales of HarperCollins Ebooks

  1. Jeremy Greenfield

    The above are B&N sales ranking charts and we believe sales ranking is a fairly strong correlative for actual sales. Obviously, there’s some “gut” guess-work here but we feel the data and insights, though imperfect, are still telling.

    Thanks for the comment, Nate!

  2. Jim Fallone

    This looks good because there are a lot of charts but it is important to understand the metric we are tracking before we make assumptions about trend.

    \Barnes & ranks books daily based on rolling six months of sales data. Sales ranking will change as new books are added to the database. If there are no sales during a particular amount of time, or if other titles have more sales, a title can lose its ranking.\

    Factors such as \other books sales volume\ and \total number of titles on site \ have a direct impact on the title in question’s sales ranking. As well, coop marketing and promotion such as paid placement of title on the site or links from advertising and promotion outside BN.Com also can greatly impact sales rank.

    Sales rank is a god metric for determining a general sense of a titles individual performance and popularity but because sales rank is based on an inconsistent and variable base it would be dangerous to rely on it for more complex trend analysis. The overall ranking is based on and aggregate of rolling 6 months sales so the impact of issues such as price change are too diluted as to be significantly attributable to a single weeks movement in rank.

    I would also note that the first titles you cite are newer and the drop in sales rank also reflects the natural bell curve of a new release. The last few titles are older and the natural bell curve of the first three months sales of sales does not appear on the chart making the graphs when compared to the newer titles misleading. Showing the data for the exact same period of the products life cycle is important when using a inconsistent metric so as not to inadvertently mislead the viewer by showing sales at different magnification.

    I would be extremely cautious of relying too much on trends based on ranking and only use it for a general gauge of individual success.

    1. Jeremy Greenfield

      Great comment, Jim. I endorse wholeheartedly everything you say here. I would just add that this should be taken for what it is: anecdata. It’s anecdata with charts and numbers, but anecdata nonetheless.

      I think what it suggests is what I write about above. What it SAYS (different than suggests) is that these price decreases didn’t result in a HUGE sales bump for these titles. Of course, theoretically, if other titles also received similarly huge sales bumps, sales rank wouldn’t change all that much. We believe internally that that is unlikely; however, one should take that into account when judging the information above.

      Still, those without the benefit of unit sales knowledge of the above (read: other publishers considering pricing) should, I would say, consider that price decreases need to result in unit sales increases to make sense financially.

  3. Michael Croy

    It is very interesting to see the lack of impact down pricing has on ranking. It’s important to consider the motives for down pricing from a publisher perspective, was it in reaction to favorable review? or extended media attention, Often both – it’s also an effort to sustain sales momentum in the face of new releases. Since today’s life cycle of a book is so short – sustaining momentum through discounting is more an opportunity to maintain the sales level than an effort to affect sales ranking.

  4. Jim Fallone

    Thank you Jeremy for my new favorite word – anecdata. I think it is this years turthiness. It certainly explains how the media and the Republicans were so surprised over the election results.

    I appreciate and welcome the data above and think it is valuable but with so many new people now self publishing without background or experience in the book trade I think its important to sometimes say the obvious out loud.

    I look forward to your follow up work.

  5. Rae Summers

    Who sets the cost of cereals or toothpaste – the manufacturer or the supermarket? Because surely the supermarket pays the manufacturer a set price for the cereal, regardless of the price they eventually sell it?

    Why should books be any different? Why should the author/publisher (the manufacturer) earn only what the retailer (Amazon) feels like charging the customer this week?

    Would Nestle, Unilever or Procter & Gamble still be in business if all industries worked the way publishing did?

  6. Elizabeth Burton

    I appreciate the effort, but this is meaningless. It’s a given a book is going to sell the majority of its copies within a few weeks of release. By the time the titles listed here were tracked, the avid fans had already paid the premium price and moved on. So, to suggest that discounting the price would cause another surge of sales is wishful thinking.

    In other words, this all ignores the established book-buying patterns of readers, so the date, while interesting, really tells us nothing about the efficacy of discounts. We would have to wait until each of these authors produces a new title then compare the first few weeks’ sales of those to the equivalent period for the ones listed.

  7. KathrynEdgar

    The biggest problem you have here is you analyzed nothing that has dropped below a 5.00 price point. As a writer I want to charge the maximum I can for my work and that is my right. The problem with most of the 9.00 + ebooks is the writer is still seeing pennies to the dollar royalty wise and the emarket has changed books. Blame freebies. Blame whatever you want but ebook buyers don’t have the same advantages as a normal book and they want compensated by saving money on their purchases.
    As a buyer I can go in to the bookstore and buy a new hardcover for 19-25.00 approximately. I can read the book, let my husband read it, share it with my sister, then my mother in law, then the neighbor, then the neighbor’s dog. I have in my hands a tangible item I can give away, trade for other books or turn in for cash at a locally owned store.
    As an ebuyer I have something I MIGHT be able to loan to ONE person. Nobody will ever give me cash for it. Ever. It’s already decreased the value of the purchase. It’s already worth less than 50% of the tangible product. But the ease is worth something – just not 20 bucks. In fact, I as a buyer and believe me when I say I’m a huge book buyer and avid reader I hate paying more than 5.00 for them. They are like the Big Mac at the steak house in books.
    My Kindle has about 500 books on it at the moment and I got it 11 months ago. 200 maybe were freebies – only 2 of them were 10.00 + purchases and both were big name authors who were auto buys for me as a reader – everything else almost 300 books were purchased at less than 5.00.
    Even as I say that I know this means the mourning kiss of death for authors like myself who are Indie publishing because once the big houses catch on to our market and market to their pockets we are going to hurt. It is already happening with some of the romance authors related to Harlequin.

  8. KathrynEdgar

    Rae that’s a somewhat invalid point. The maker of a product such as toothpaste has a wholesale price point. The supermarket, Walmart, family store all do set their price on the item. Some choose to take a loss and host a sale on a product. Others buy during a special deal for pennies on the dollar so they can run the sale without a loss. But make no mistake that Unilever, P&G or anyone else distributing to the stores has no control over what the shelf price is of the product. They don’t care. It’s why a rural grocery store can charge 10.00 for a tube of toothpaste knowing you would have to drive an hour to pay the right price for it.

  9. Kelly McClymer

    Interesting. I would be curious to see a bestseller’s rankings over the same time period under the agency model, to see if the price drops kept the book selling more strongly than typical.

    I really look forward to seeing a year’s worth of data, though. That should be fascinating. We’re all still getting used to the long tail model, and seeing the overall effects of combining price incentives with holiday/seasonal sales on titles that are not newly published will be very informative.

  10. Kevin O. McLaughlin

    Keep in mind that this surveys only major bestsellers, which are among the LEAST price sensitive sort of books. In other words, being $2.99 instead of $9.99 will often help a book by a lesser known writer. It might even make that writer a bestseller, at least briefly. But it is unlikely to help the work of a writer who is already a major bestseller. The fans of writers already selling millions of copies are used to paying $15-25 for a copy. A couple of dollars difference could in theory impact where readers buy the book, but thanks to DRM readers are much less likely to “shop around” and much more likely to lock in to a single ebook retail platform.

    What pricing drop does is not so much boost these bestsellers’ sales as overall enhance the impression that ebooks are cheaper, therefore worth the expense of a device, increasing the rate of reader switchover to ebooks and hurrying along the date of physical bookstore collapse.



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