Did Nielsen Report US eBook Sales Were Down in 2014, or Volume?
When I reported yesterday on Nielsen’s estimates for ebook "sales" in the US in 2014, I questioned whether the report was correct. I’m still waiting for Nielsen to get back to me on this, but based on my conversations on Twitter this morning it now appears that I was right to question the report.
Update: Never mind. The Nielsen Pubtrack figures are completely bogus.
Unfortunately for me, I didn’t ask the right question.
Yesterday I picked up a story in which The Bookseller reported that Nielsen said the US ebook sales dropped by 6% in 2014 from 2013. I doubted that report because the AAP offered contradictory revenue data, but now it appears that The Bookseller may have been the source of the problem, and not Nielsen.
According to several people who attended the BEA 2015 conference session where Nielsen shared the data, the Nielsen rep did not say that sales were down. Instead, they said that volume (or sales volume) was down in 2014, and that is not the same thing.
What’s the difference?
Well, I can’t give you a simple answer, but the least complicated version is that Nielsen reported on how many ebooks (units) were sold, and not on how much was spent to buy those units. This was not clear in The Bookseller’s story, which said that ebook sales were down but really meant that ebook sales volume was down.
Volume, or sales volume, is measured in the number of units sold, while sales (or more often revenue) are measured in currency.
To give you an example, the AAP releases sales or revenue data every month, while Nielsen shared volume estimates last week at BEA 2015 (or so I’m told).
If you would like other examples, try googling the phrase "sales down volume up" and see how many financial reports you find which make this distinction.
In short, I have to call out my own error. I was wrong to use the AAP to cast doubt on Nielsen’s numbers. They did not contradict each other so much as they described different things.
Or at least that is what I think is going on. I’m going to have to wait for Nielsen to explain this better before I know for sure what really happened here.
Update: I was right.
More than anything, this served as a reminder of how one should be careful in choosing the right words. As someone who regularly flubs word choice, I can’t fault The Bookseller, although I do wish they had been more clear.
P.S. And while we’re on the topic of sales volume, Nielsen’s data is nothing to write home about. They exclude market segments which are going to see huge activity in 2015:
@thDigitalReader @philipdsjones Nielsen was very clear that promotions, special sales, subs and the like were NOT included in unit numbers
— Andrew Rhomberg (@arhomberg) June 2, 2015
At last report, Kindle Unlimited accounted for over 7 million ebook loans in April 2014. By the end of 2015 KU loans could make up 20% of the sales volume, and to exclude it would mean ignoring a huge chunk of the market.
image by garysan97
Smart Debut Author June 2, 2015 um 11:13 am
Caveat emptor when it comes to "official" industry stats — even the AAP data only measures half of the ebook market. And it really is roughly half: the Big Five and the other medium-size publishers that make up the AAP comprise no more than 40% of the paid unit sales volume and 60% of the dollar sales.
Don’t forget, something like 30% of all paid ebook unit sales in the US are of books without ISBNs.
You mentioned KU as an uncounted ebook segment, but you didn’t mention the other non-KU indie books, which make up the bulk of that "shadow" 30%. Or Amazon’s own publishing imprints, which account for 7-8% of the entire US ebook market right there.
Once you factor those into the equation, it becomes impossible to conclude anything about the overall ebook market from the AAP’s numbers, let alone Nielsen’s PubTrack stuff.
Any decline reported by the AAP et al. could just as easily be measuring a loss of market share by the Big Five and medium traditional publishers to indies, smaller presses, and Amazon imprints ebooks, while ebooks continue to grow.
Nate Hoffelder June 2, 2015 um 12:17 pm
I mentioned KU because it let me point to the list of excluded segments and use Nielsen’s own details to shoot it down. I thought that was a stronger argument than the usual point about the limitations of what is and is not being tracked.
And I left out the point about ISBNs because I’m not sure whether they matter in this case (I don’t know). But I agree, indies are probably being ignored or discounted.
Maria (BearMountainBooks) June 2, 2015 um 3:07 pm
It can’t be very accurate if they leave out promotions. How would one account for sales via kobo that use coupons? The price shows the same and unless Kobo shares certain sales data a coupon sale will look like a regular sale. (And some of the coupon amount can/does/sometimes comes from the author or the publisher. Other times it doesn’t.) So it sounds like they really are just grabbing any old bits and pieces and then trying to compare it to possibly different bits and pieces from the year before.
Not that stats matter all that much to me unless several sources show a trend. Given the data we have now and the ways books can be sold/borrowed/given away, even trends may not be real.
Ventes d’ebooks USA 2014 : ils rapportent toujours plus d’argent June 3, 2015 um 5:31 am
[…] récupérés par l’AAP ne compte que pour les moitiés des vraies ventes d’ebooks (lire ici). Pour obtenir les chiffres, ils doivent donc extrapoler en multipliant par deux les données […]