Self publishers often ask me what kind of sales they can expect for their POD title. The answer is always "it depends", but I realize that most new publishers would like a little more data:-) Before we get to the data, let me give you the "it depends" part in brief bullet form.
1) Title sales depend on customer demand. If you publish a book nobody is interested in reading, it's a really hard sale.
2) Titles sales depend on marketing. If you publish a book that a tiny percentage of the population wants to read, but you have a way to reach a large proportion of them, you can sell a lot of books.
3) Title sales depend on availability. If you have a great book and great marketing but can't deliver through a sales channel your customers are willing to use, you're losing a lot of potential sales.
4) Title sales depend on the book meeting the customers expectations. If the book doesn't meet their expectations, it will get no word-of-mouth or reviews.
The image below shows the Ingram Distribution demand and sales for my second POD title (published in 2003), which I just replaced with a new edition. It shows just over five years of distribution sales for the book. It does not include my direct sales, ebook sales, overseas sales or school sales (unless they ordered through Ingram). It is also atypical for a large trade book, but the expected result for a niche title.
The first thing you'll notice is no initial demand! I did no advertising or promotions for this book and only launched the web version a few months before publishing the POD book. Consequently, the first full month it was available (July 2003), I sold all of 13 copies through Ingram distribution (sold more than that by direct mail order). If I was an acquisitions editor at a large trade publisher, I would have lost my job. The first five months the title was available it averaged well under a sale a day through Ingram, until the Christmas rush jumped it up to 86 copies in a month.
In 2004, the sales rate more than doubled as the website gained traffic and the book gained word-of-mouth and useful Also Bought and Better Together relations on Amazon. Sales through Ingram would have been over 1,000 for the year except there was a distribution snafu for Lightning Source books in September and October of that year, as you can guess from the sudden drop in what would normally be the beginning of the high season.
In 2005, the third year of the title's life, the sales rate through Ingram was more than triple the rate of the first year, and this despite the fact that 2005 was the record year for ebook sales of the title. I believe March 2005 was also the first time the Barnes&Noble warehouse laid in a stock of books from Ingram, but I didn't keep careful enough records to work it backwards.
In 2006, sales through distribution dropped a little, and I don't remember if it was due to the book losing many Also Bought and Better Together relations on Amazon due to a software problem on their platform, or new competition, of which there was always plenty. Amazon dropped Lightning Source ebooks halfway through 2006, effectively putting me out of the ebook business for the next two years, but the paper book received no sales bump from this, which leads me to believe that ebook and paperback book buyers on Amazon have limited overlap. I think this agrees with Amazon recent Kindle statements suggesting that Kindle owners buy more paper books.
So the title peaked in 2005, and drops off into remainder sales and obscurity, right? Not with POD. Even today, copies rarely come up for sale used, and never last long. The sales bounced back and 2007 may have been its best stocking year at Barnes&Noble, with the book appearing on the shelves at about one in four superstores that I checked. Sales for 2007 through Ingram set a record for the title, running at quadruple the sales rate of the first year.
In 2008, returns started coming in for the book, eventually reaching nearly 10% of sales. I believe the main source was from store shelves where the 2003 copyright date was becoming a turn-off for buyers and bookstore purchasing software alike. A couple of college instructors who use the book for a course text were also asking when a new edition would be available, so I finally published a new edition last month. As soon as the start of the school year has safely passed, I'll take the original edition out-of-print, so you can take the data above as the complete life cycle of the title through distribution. The 2008 year was on track to do around the same as 2006, and only time will tell if releasing new edition was premature and a long term mistake.
The title in question was a $14.95 book of diagnostic flowcharts, on which Foner Books netted approximately half the cover price on Ingram sales. I didn't add up the sales numbers above, but it looks to be a little over 6,000 copies, or about $45,000 in net through the Ingram channel. Over a quarter of those sales were generated directly through the Amazon Associates links on the Foner Books website. How many of the sales were generated indirectly through those links is unknown, but my guess is you could at least double the number, attributing over half of the Ingram sales to customers who ordered through a bookstore, or bought through Amazon after the Associates link lost track. Over the years, the website sent over 18,000 book shoppers to Amazon to look at the book. And since I never approached Barnes&Noble to stock the book, I believe that the two or three years they did stock it were thanks to walk in demand generated by the website.
A few concluding notes about the title to help you make comparisons. First, computer hardware books are a niche subject, and not a popular one. Over the five year life cycle give above, they became even less popular, so my title was really gaining market share in a shrinking market. I would estimate that between 2003 and 2008, my POD title was the #2 computer hardware book sold by Amazon, yet it didn't even average in the top 10,000 books for the last couple years. A few examples of other niche genres that sell as many or more books than computer hardware would be: root cellaring (far more:-), concrete garden ornaments (around the same), how to read books (far more), beer (around the same). In two of the examples given, the lead book averages in the top 1,000 on Amazon and was published more than fifteen years ago!
Back in 2004 I did a one year case study including a detailed month-by-month sales account of the first POD title I published, including direct mail order sales and UK distribution. That case study is also included in my title about POD publishing.