The Best and the Rest
How Methodology Makes the Bestseller
It is a truth universally acknowledged that every writer wants to become the author of a bestseller.
That’s not to say everyone wants to write a cookie-cutter story purpose-made to appear on a list, but rather that every writer hopes their book will be recognized for its merits, read by many, and successful in the marketplace. I can’t imagine a writer saying, “I hope my book will not sell well enough to become a bestseller.” People have declined the Nobel Prize in Literature, the Pulitzer Prize, and the National Book Award, for various reasons, but nobody turns down unit sales.
Well, almost nobody. The only book I can think of that actively tried to not become a bestseller is Steal This Book, which unfortunately sold more than 200,000 copies when it came out in 1971, deeply embarrassing Abbie Hoffman.
Hilariously, right-wing publisher Regnery decided a few years back to stop marketing its books as “New York Times Best-Sellers,” and requires its authors not to advertise their own books as such, due to their perception of the NYT’s bias against conservative authors (read about it here).
Everyone wants their book to be a bestseller but in my experience many folks don’t know what it means for a book to be a bestseller. That’s not anyone’s fault, because the term bestseller doesn’t actually mean anything.
In the most pedantic sense—the all-words-mean-something sense—a bestseller is any book that sells more units (or generates more sales dollars, depending on how you’re counting) than most other books. There is no specific level of sales that a book must reach to become a bestseller, the way a record “goes gold” when the RIAA certifies that it has sold 500,000 copies. There’s no one governing body that ratifies bestselling books. There’s no agreed-upon methodology for calculating bestsellers, no one source of data that every list uses. So when I hear about a new book coming out from a “bestselling author,” I become suspicious. “Bestselling author” can mean anything.
The Lists
There are four major publications in the United States that produce regular “bestsellers lists”: The New York Times, the Washington Post, USA Today, and Publishers Weekly. In addition to these four, Amazon is very happy to show you its bestselling books list (its bestselling anything by category, in fact), a section of the site that updates hourly based on sales data.
All these lists publish on different schedules and organize content in different ways. Their methodologies are proprietary trade secrets. Outside of the people who work for these companies, no one knows for sure exactly which data they’re using and how they’re using it to arrive at their lists.
New York Times
This is generally regarded as the most prestigious bestseller list and it’s been around forever. NYT splits its bestsellers into eleven separate weekly lists and another seven monthly lists. The weekly lists are: combined print and e-book fiction; hardcover fiction; trade paperback fiction; combined print and e-book nonfiction; hardcover nonfiction; paperback nonfiction; advice, how-to, miscellaneous nonfiction; children’s middle grade hardcover; children’s picture books; children’s series; and young adult hardcover. The monthly lists are audio fiction; audio nonfiction; business; graphic novels and manga; mass market paperbacks; middle grade paperbacks; and young adult paperbacks.
The weekly lists include fifteen books each, so each week the NYT names 165 bestsellers. To put that in perspective, there are somewhere between 5,000 to 6,000 new books published each week in the United States by traditional publishers. So that’s, I think, around 2 to 3 percent of books that become bestsellers according to the NYT (keep in mind that plenty of books remain on the list for more than one week so it’s not 165 unique titles each week).
What does it take to be one of those 165 books per week? You have to sell the most copies of all the books that were published that same week, right?
No. First of all, they don’t track every book. They exclude several broad categories of books, like perennial sellers (eg, The Bible), textbooks, periodicals, and so on.
Next, they don’t track every sale. NYT gets their data from a specific number of “reporting vendors” around the country. Their reporting vendors include indie and big-box bookstores (but specific ones; so “Barnes and Noble” isn’t a reporting vendor but “the Barnes and Noble on Telephone Rd in Ventura, CA,” could be) as well as wholesalers. NYT says that they’re tracking actual consumer transactions—they’re counting units sold to people, not units sold to Costco—as reported by their participating sample of vendors. (That really makes one wonder what they’re using the wholesaler data for.)
There are flaws with this method of determining bestsellers, one of the major ones being that if an author or publisher can figure out even a handful of those reporting vendors they can turn around and rig the whole system. This happened in 2017 when a publisher contacted a number of NYT reporting vendors and ordered large quantities of their newly published title for “local author events” they claimed they were going to be hosting, skewing the sales numbers and landing a completely unknown title in the #1 spot. This isn’t a new tactic; Jacqueline Susann did the same thing with Valley of the Dolls in 1966.
Washington Post
The Washington Post publishes a list of bestsellers each week—paperbacks one week, hardcovers the next, then back to paperbacks, and so on. The paperbacks are divided into fiction, nonfiction, and mass market; hardcovers are divided into fiction and nonfiction. Each section of the list has ten titles—30 titles on fiction weeks, 20 titles on nonfiction weeks. As far as I can tell, these lists are not curated by the Washington Post but sourced from the American Booksellers Association, which publishes a list regularly that is in turn picked up by many newspapers and other news outlets—many of which do not feature the overzealous paywall that the Washington Post has. In other words, get this generic list somewhere else.
USA Today
I’m going to level with you, USA Today is my personal favorite of the publications in this article for bestsellers. Like the NYT, they get their data from reporting vendors—but USA Today’s reporting vendors are all booksellers. They don’t have wholesaler data in the mix. Next, they disclose a selection of those reporting vendors right on their site so you know exactly where they’re getting data from—they don’t claim that they’ve disclosed every single vendor but they give what I feel is a representative sample. Their methodology is the most transparent.
Next, they don’t split books up by format but instead report on bestselling titles. Most books come out in either hardcover or paperback plus an e-book to start, sometimes with a paperback or mass market paperback following later. If a title has, for instance, hardcovers, mass market paperbacks, and e-books all selling at the same time, USA Today counts all of those sales together toward that title’s rank. This makes more sense to me than making, for instance, the hardcover edition of something and its later paperback edition compete on two separate lists.
The list is more intuitive to use: USA Today displays all 150 titles together and you can filter by fiction or nonfiction and by genre. They offer a PDF download of their list, which makes it more accessible.
And none of this is behind a paywall. It’s freely available.
Publishers Weekly
PW (I pronounce this pee-dub) is the industry rag so I don’t understand why they have the most opaque methodology of any of these outlets. Where do the sales figures come from? When does the reporting week begin and end? So many questions.
They have sixteen lists plus a “top 10 overall” list. Some of the lists specify that they’re reporting on frontlist titles, while others don’t specify so I can assume they’re reporting on frontlist and backlist together. How is PW defining frontlist and backlist, given that those terms don’t have universally understood definitions? So many questions.
PW features an article each week called “This Week’s Bestsellers” that gives a brief, low-effort overview of a few select titles from that week’s list, mostly directing readers to recent PW reviews or interviews for further reading. It does not contain a useful analysis of the week’s list or highlight any trends.
Look, there’s a reason I have been filtering PW Daily directly to trash like fifteen years.
Amazon
Finally, there is Amazon. Amazon does not tell you about US publishing sales, Amazon tells you about Amazon publishing sales. The Amazon bestseller list is the list of books that are selling best on Amazon. All the other publications’ lists have most of the same books, but the Amazon bestseller list is completely different with just a few overlapping titles.
There are lots of books that are only available on Amazon—books published only through KDP, for example. Also, there are books that sell well on Amazon but don’t sell well in brick-and-mortar stores.
Consider, for example, the book in the #10 slot as of right this second, a COVID-denial screed by a person in the medical industry best known for anti-vaxxism and hawking dietary supplements. I notice that in a sample of independent bookstores I checked online, many were not carrying this book (I wonder why). Further, to buy this book in a store, you would run this risk of other human beings seeing you purchase this trash. I contend it’s possible that this book is selling enough copies on Amazon to make it onto this list, but not enough copies across reporting vendors nationwide to make it onto any other list.
Also, since the Amazon list updates hourly—while the other outlets tally once per week—Amazon will report sales trends much more quickly than the other lists.
I suspect that Amazon also represents the purchasing power of many American readers who do not have easy access to a brick-and-mortar bookstore but do have delivery service through USPS—rural areas in particular. This could also be a factor in why the makeup of Amazon’s list is so different from that of its peers.
What it means to be a bestseller is murky and vague, and often misleading. An author who has one book sell enough, quickly enough, at the right outlets, to make spot #15 on one of the NYT’s lists for one week, and then drops off and never appears on the lists again, can call themself a “NYT Bestselling Author” for life. Meanwhile, a book with moderate but steady sales over a longer period of time may never make the list but could outsell the first book by an order of magnitude.
At the end of the day, a book is “a bestseller” if it appears on any list of bestselling books, at any spot on the list, one time. There’s a big difference between being “New York Times #1 bestseller” and being “#1 in Paranormal Demons & Devils Romance” on Amazon, but both of those authors could reasonably call their book “a #1 bestseller.”
My best advice, your best bet is to write the best book you can—then cross your fingers and hope for a great marketing manager.
If there is anything you would like me to write about, please drop a line and let me know. I expect to be down and out for a few days a couple weeks out, sleeping off my second COVID shot, so I’m trying to get a ahead and I want to populate the calendar with the topics you’re most interested in. Have you always wanted to know what a dangling modifier is but been too embarrassed to ask? Want to know who makes the best dot grid journal to write in with your Pilot G2? I have the all the answers. You only need to ask.
If you have questions that you'd like to see answered in Shelf Life, ideas for topics that you'd like to explore, or feedback on the newsletter, please feel free to contact me. I would love to hear from you.
For more information about who I am, what I do, and, most important, what my dog looks like, please visit my website.
After you have read a few posts, if you find that you're enjoying Shelf Life, please recommend it to your word-oriented friends.