A fascinating article in Publisher’s Weekly showed up today that’s worth sharing. All of my author friends are familiar with a company called bookscan, that purports to offer unbiased quantified data about book sales throughout all outlets. But do they? The PW article, entitled “As BookScan Grows, So Do Questions”, starts by saying:

When Bill O’Reilly and Al Franken were arguing last month about who
had sold more books, online muckraker Matt Drudge stepped in with the
final word by citing an inviolable source: BookScan.

The point-of-sale data service run by Nielsen and owned by Dutch
conglomerate VNU (a competitor of PW parent Reed Elsevier) is on its
way to becoming a standard-bearer in an industry starved for hard
data. In the space of a few years, it has gone from a sputtering
pipe-dream to a recognized tool, used in everything from settling a
celebrity debate to setting the price of a rights-sale. With Penguin
becoming one of the last of the large houses to sign up, publishers
have firmly embraced the system.

But while the media and the industry increasingly cite BookScan,
questions have cropped up about how accurate a picture it paints – and
how it might be misused…

The article continues…

Agents in particular have been ambivalent,
saying editors too often wield it as a blunt instrument in
negotiations. As publishers become increasingly likely to use it,
anecdotes abound of agents who play down BookScan numbers to them.
“It’s not bad in and of itself,” says Trident Media’s Robert Gottlieb.
“But some editors, especially less experienced ones, over-rely on the
system. Editors shouldn’t be using it unless they’re supervised by
other people in the house with other information.”

So how representative are the service’s numbers? An informal survey of
the top-selling books of 2003 showed some surprising things.

BookScan generally claims to represent between 70% and 75% of sales in
the industry (Wal-Mart and some of the supermarket chains are among
those who decline to report.) But a comparison with in-print figures
supplied by publishers reveals that the numbers are more likely to
represent about 65%, even after deducting for unsold books and
returns.

For BookScan’s top ten nonfiction titles published last year – a list
that include mass-market favorites like Phil McGraw’s diet books as
well as indie hits like Benjamin Franklin: An American Life – no title
had BookScan sales comprise more than 75% of total sales. For some of
the books that had strong special-sales, they ran as low as 25%.

The range of coverage is shown by three Simon & Schuster books. The
service was very close to the numbers reported by S&S for Franklin
(BookScan, 384,000; S&S, 550,000 before returns) Figures for Living
History diverged a bit more (BookScan, 1.1 million; S&S, 1.8
million.). The biggest discrepancy came for The Ultimate Weight
Solution by Phil McGraw (BookScan: 836,000 copies; S&S, 2.5 million).
Showing an even further disparity was The Purpose-Driven Life, which
the publisher said sold more than 11 million copies last year but
charted only 2.4 million. In the same vein, Bill O’Reilly’s Who’s
Looking Out for You sold 430,000 copies on BookScan, while Doubleday
Broadway cites an in-print number of nearly one million.

Publishers say the huge gap is due to the sales of the book outside
BookScan channels, not just at Wal-Mart but at smaller independents,
the religious market and grocery chains. The Purpose-Driven Life, for
instance, sold millions of copies at Christian Booksellers Association
stores, which do not report to BookScan.

Nielsen BookScan’s Tim King said the service is striving to add more
stores. It currently has about 400 independents; it also recently
added Follett. The drug stores, he said, can be a problem because they
don’t always scan according to ISBN but he hoped to also add there. He
does not expect Wal-Mart and Sam’s stores to change their policy of
not sharing data with Nielsen.

The service is proud of its point-of-sale tracking, saying it
eliminates the messiness that can occur when sales are weighted, as
they are in many bestseller lists. But pinpoint accuracy can also be a
disadvantage. Bulk purchases made by an author, for instance, will all
register as separate sales, while when independents buy a hot book
like Harry Potter at a steep discount and then resell it, it can end
up counting twice.

Still, many houses say they are learning to recognize the system’s
limits and work within them. Some say the numbers can sometimes help
in predictive ways, as they did at Rodale, where publisher Amy Rhodes
said that the sell-though numbers from the service gave her the
confidence to repeatedly hit the reprint button on The South Beach
Diet, all the way up to the current five million. “I couldn’t have
managed South Beach without BookScan,” she said. King emphasized that
figuring out the best uses could take time. “We’ve only been on-line a
couple of years,” he said, “There’s a learning curve.”

Most in the industry say BookScan is better than a typical bestseller
list but that it remains a far cry from a royalty statement or a
definitive gauge like companion Soundscan. “It’s not perfect. But it’s
a usable tool. The more numbers you have, the more likely you are to
find the truth between them,” says agent William Clark. In other
words, the only reliable thing you can say about book sales trackers is
that none are fully reliable.

This article was written by the ever-reliable Steven Zeitchik, with Jim Milliot, and if PW had their archives online, I’d point to it there, but it’s important for the writing community to understand the pro’s and con’s of bookscan, I think.