For a business built primarily on words and ideas, publishing is surprisingly obsessed with numbers. Well, perhaps not obsessed, exactly—it's more of a love/hate relationship. First printings, size of advances, total sales: it seems we can't decide whether to tell the truth about them, or believe what other publishers tell us, or decide they are meaningless. But there is one (sort of) numerical thing we value as a true indicator of success: the bestseller lists as compiled by PW, the New York Times, Book Sense and a couple of other pivotal places. If a book makes "the list," it's deemed a success; if it gets to the top spot, it's an even bigger one.

But becoming a bestseller is not all that it used to be—at least it's not as meaningful in a long-term, stay-in-the-public consciousness sort of way. Not according to Lulu.com, a print-on-demand Web site that conducted a study of 50 years of bestsellers.

Among its findings:

The average number of novels to reach #1 on the New York Times bestseller list in the 1960s was 2.8 per year, compared to an average of 18.2 novels in each of the first five years of our new century.

The novels that have had the longest continuous reigns in the top spot are Allen Drury's Advise and Consent (57 consecutive weeks), The Source by James Michener (43 weeks) and Erich Segal's weeper, Love Story(41 weeks).

Shockingly, the longest unbroken spell atop the heap for The Da Vinci Code—widely believed to be the bestselling novel of the century, if not all time—was only 13 weeks. (It has been on everybody's list for more than three years, of course, and has moved in and out of the top spot.)

What does this "churning" of #1 bestsellers indicate about the way we're doing business? Does the relatively rapid seat-swapping indicate that people who read are reading more often and more widely?

It sure would be pretty to think so, but I'm afraid I think this admittedly small study (it covers only one list, and it's not PW's) suggests something a bit more worrisome: namely, that because of publishers' reliance on "big books" to make back the inflated advances they get, the effort to get on to the list—anywhere on the list—is equally outsized. And while such efforts may work, at least for a week or three, creating a short-term bestseller should not be the point of publishing; besides, it rarely makes a house money in the long run. Hollywood counts on a boffo opening and impressive grosses to spur interest and make news. Publishers can't.

Last year at PW, 20 new books hit #1 on our hardcover fiction list, 17 of them in their first week. But eight of those were gone in a week, victim of the one-day, all-out laydown. Any publisher worth his BookScan password knows that healthy backlist is where the money is—that good books shouldn't be forgotten. They should apply this thinking to frontlist as well.

So, publishers: dispense with the Hollywoodthink. If you must indulge in media-envy, look to Broadway, where the cognoscenti know that while there's glamour in opening big—the smell of the greasepaint, the dull roar of the crowd!—the bankable glory comes from producing the longest-running shows.

Agree? Disagree? Tell us at www.publishersweekly.com/saranelson