Bestsellers

Data about bestsellers is based on the number of books sold, and not on the number printed, the number shipped or the dollar value of the sales. Hence, it is a measure of books' popularity, and not a measure of their quality or of what people could have read or should have read. It is also a relative measure, being valid only for the period under study or for the point in time when the question about sales was asked. Yesterday's bestseller may become today's pulp.

The practice of regularly publishing data about bestsellers in the United States dates back to 1895, when the literary magazine The Bookman began publishing monthly lists of bestsellers in various cities. Today, the best sources of information about them are:

•   The New York Times  , America's newspaper of record, which has been listing them since 1942
•   Publishers Weekly  , America's authoritative trade magazine about bookselling and book publishing, which has been listing them since 1912

Both conduct nationwide surveys and they both attempt to balance input from chain stores and independent booksellers. As a result, they tend to avoid the skewing that may afflict surveys having a more limited scope. For example: The Washington Post focuses on sales in metropolitan District of Columbia, and its list tends to be dominated by books about politics; the Los Angeles Times focuses on sales in metropolitan Los Angeles, and its list tends to be dominated by books about business and celebrities. Nationwide surveys show that, in general, Americans are more interested in being religiously inspired, losing weight, becoming rich and, since the September 11 attacks, in fighting terrorism and reading right-wing tirades against anyone not right-wing.

Both publish weekly bestseller lists. Rankings in these lists are based on sales in bookstores, and nowhere else. Data on which the rankings are based is gathered a few weeks prior to the list's publication. Hence, there is a gap of a few weeks between the time when the sales occurred and the time when they are reported on the lists.

Publishers Weekly also publishes annual bestseller lists in mid- to late-March of the year following the year they report on. For example: the list of 2001's bestsellers was published in the March 18, 2002 issue; the list of 2002's was published in the March 24, 2003 issue. Rankings in these lists are based on sales in bookstores and other distribution channels as well, such as newsstands, supermarkets, drug stores and gas stations. Because they are based on a wider sample of outlets, these lists provide a broader view of the marketplace for books and they include many books that never appear on a weekly list.

Currently, both organizations produce separate lists of the top fifteen bestsellers in various categories, such as fiction and nonfiction, and hardcover and paperback. Fiction lists tend to identify America's favorite authors; nonfiction lists tend to identify the subjects that interest Americans most at a given point in time. Because of the their role in identifying subjects of current interest, the lists of nonfiction bestsellers often ignore bestselling standard nonfiction books. For example: new versions of the Bible and some cookbooks and dictionaries may sell hundreds of thousands or more copies per year, year after year, but they are generally identified as bestsellers only in the first year or two of their publication. If such standards were listed every year in which they sold in numbers large enough to be included on the lists of nonfiction bestsellers, the lists would appear to change little over time.

Although important nonfiction titles such as the 9/11 Commission Report are sometimes initially published in paperback, indexers are generally most interested in the category of hardcover nonfiction bestsellers, which represents the largest market for their services. These can be measured in terms of the number of distinct titles listed during a given period or in terms of the number of available positions on the lists published during the period. The data in the tables below begins with the year 1975 because information about the presence or absence of an index in a book is not readily available for books published prior to that year.

The tables indicate that the long-term gradual decline in back-of-the-book indexing, which coincided with the period of greatest consolidation in American publishing industry, may be ending. Between 2000 and 2003, the downward trend had continued. During those years, the percentage of bestselling titles with indexes was 37% (2% less than the period 1990 to 1999) and the percentage of positions with indexes was 33% (6% less than the period 1990 to 1999). In 2004 and 2005, however, 60% of the annual bestsellers had indexes. Those years tie 1976, 1978 and 1991 for second-highest percentage ever. (The highest percentage is 73% in 1980.)

If only new titles are considered — i.e., only titles not carried over from a previous list — the picture is even somewhat more encouraging:

For details about decennial periods, link to the pages for the periods

Indexes are slightly more common in hardcover nonfiction bestsellers appearing on weekly lists, where about half the books listed have an index. These lists account for a much smaller number of outlets than do the annual lists.

Weekly Bestseller Lists, 2000-2006 New York Times (NYT), Publishers Weekly (PW)
	Overall		NYT		PW		Both
2000:  2001:  2002:  2003:  2004:  2005:    Average 2000-2005:    2006 so far:  Change so far:	46% 50% 46% 49% 48% 46% ----- 48% ----- 46% –2%		47% 54% 46% 50% 48% 41% ----- 48% ----- 47% –1%		40% 50% 44% 50% 49% 45% ----- 46% ----- 49% +3%		39% 56% 45% 53% 47% 36% ----- 46% ----- 51% +5%

For details about weekly lists, link to the pages for the years

In some fields, the percentage of books having indexes greatly exceeds that for all nonfiction books. An example is foreign affairs and international relations.

For details about foreign affairs and international relations bestsellers, link to the pages for

To top of page