How Many Copies of a Book are Sold, On Average?
Steve Laube recently posted an article on average book sales. The figures he posted are interesting and yet, most authors never sell that many books. Across 10 authors and 51 books, he came up with a figure of 12,455 copies. Someone else came up with a figure of 5,000 copies. Most of the one million books published in the United States each year will never sell more than 100 copies. Why the differences in numbers?
One thing we have to ask when trying to find the average is what we really mean by the average. As you are probably aware, there is more than one way to calculate an average. The most frequently used method is to calculate the mean. This is done by taking the total books sold and dividing by the number of titles. This is what Steve Laube did. But the mean has little meaning when a few books have very large numbers and the rest are smaller. For example, if we have three books with sales of 5,300, 7,000, and 1,000,000, the mean will be 337,433. We wouldn’t tell an author who has written a book that he will probably sell 300,000 books, based on those numbers. Steve Laube tried to get around the problem by removing the very high numbers and the very low numbers. The problem with that is that it is arbitrary to decide where the cutoff point is. Also, as so many industry professionals tend to do, by focusing only on the major publishers, Steve pretends that more than half a million books published in the United States don’t exist.
The median is a better method of calculating the average in a situation like this because it more accurately represents the typical case. That is usually what people are asking for when they ask about the average book sales. They realize there are some books that sell much larger numbers and some books that much smaller numbers. It eliminates the problem of the extremes because it finds a midpoint at which half the books sell less than the median and half sell more. Given the three books mentioned above, the median is 7,000. But here again, we have the question of whether we should include the half-million books that are selling less than 100 copies. If we do, our average is likely to come out at 100 copies per book, nowhere close to Steve Laube’s 10,000 or the other fellow’s 5,000.
Industry professionals typically leave out the lower half-million because the vast majority of those books are self-published and poorly written. Industry professionals are going on the assumption that by the time a book makes it through the filtering process of obtaining an agent, publisher selection, and improved editing and marketing the book has a much better chance of success. They aren’t concerned about books they wouldn’t publish. For them, that makes sense, but the rest of us have to decide which books we should include in the average.
If you are considering self-publishing, there’s no doubt about it, you must include the half-million self-published books in your figures. While you may be convinced your book is better, you have no way of knowing, since you haven’t read all most of those books. But you can help yourself out by focusing only on those books that fall within your genre. Comparing a children’s book to a computer book is like comparing apples to oranges. But if you are seeking a publisher, don’t assume you can use Steve Laube’s numbers. Until you have a contract in hand, you will have to include the sale of all those people who also don’t have a contract. I don’t know what that number is, but I do know there are a lot of books that are sitting on the shelf and will never be published.
To say that this is complicated is an understatement. At best, we can compare our book to the books most similar to ours, though that is hard enough. But we can’t find a simple average number of books sold.
One thing we have to ask when trying to find the average is what we really mean by the average. As you are probably aware, there is more than one way to calculate an average. The most frequently used method is to calculate the mean. This is done by taking the total books sold and dividing by the number of titles. This is what Steve Laube did. But the mean has little meaning when a few books have very large numbers and the rest are smaller. For example, if we have three books with sales of 5,300, 7,000, and 1,000,000, the mean will be 337,433. We wouldn’t tell an author who has written a book that he will probably sell 300,000 books, based on those numbers. Steve Laube tried to get around the problem by removing the very high numbers and the very low numbers. The problem with that is that it is arbitrary to decide where the cutoff point is. Also, as so many industry professionals tend to do, by focusing only on the major publishers, Steve pretends that more than half a million books published in the United States don’t exist.
The median is a better method of calculating the average in a situation like this because it more accurately represents the typical case. That is usually what people are asking for when they ask about the average book sales. They realize there are some books that sell much larger numbers and some books that much smaller numbers. It eliminates the problem of the extremes because it finds a midpoint at which half the books sell less than the median and half sell more. Given the three books mentioned above, the median is 7,000. But here again, we have the question of whether we should include the half-million books that are selling less than 100 copies. If we do, our average is likely to come out at 100 copies per book, nowhere close to Steve Laube’s 10,000 or the other fellow’s 5,000.
Industry professionals typically leave out the lower half-million because the vast majority of those books are self-published and poorly written. Industry professionals are going on the assumption that by the time a book makes it through the filtering process of obtaining an agent, publisher selection, and improved editing and marketing the book has a much better chance of success. They aren’t concerned about books they wouldn’t publish. For them, that makes sense, but the rest of us have to decide which books we should include in the average.
If you are considering self-publishing, there’s no doubt about it, you must include the half-million self-published books in your figures. While you may be convinced your book is better, you have no way of knowing, since you haven’t read all most of those books. But you can help yourself out by focusing only on those books that fall within your genre. Comparing a children’s book to a computer book is like comparing apples to oranges. But if you are seeking a publisher, don’t assume you can use Steve Laube’s numbers. Until you have a contract in hand, you will have to include the sale of all those people who also don’t have a contract. I don’t know what that number is, but I do know there are a lot of books that are sitting on the shelf and will never be published.
To say that this is complicated is an understatement. At best, we can compare our book to the books most similar to ours, though that is hard enough. But we can’t find a simple average number of books sold.
Comments
Example:
"Most of the one million books published in the United States each year will never sell more than 100 copies."
If it takes me a year to write a book I know I am very unlikely to make a year's salary from self-publishing (and without factoring in the promo / time it takes to get your book noticed) But I guess like most creative 'hobbies' you just have to do it and the enjoyment comes from that - any income is a bonus .... least that's what I keep telling myself !!!
But also...
Amazon recently announced their eBook (Kindle) sales have for the first time eclipsed physical book sales... I guess with Apple and Google the only way is up !!