So you’re thinking about self-publishing an I.T. book?

Here's the tl;dr: * Peter made $48k from Flexible Rails. * $34k of that was from sales of the PDF.

I'm biased, but it seems to me that giving away the PDF (where you make close to 100% profit) to sell a paper book (where you make much less -- The 40% profit from the article is by far the highest I've ever seen) is crazy.

If you have an audience and a good book, you should be able to do well on an e-book. Keir Thomas in the article has both, but he gave away 500 000 copies of his ebook and only made $9k on the paper book. Yikes! He could've sold the book for 2$ a pop and got 1% of the downloads and still made the more.

Keir is right that you have to give away something for free. So, give away a good chunk of the book (say, 1/3 or so) as a PDF and sell the full thing.

Or, give away the whole book for free as HTML and sell the PDF. That's what Eric Ries http://leanpub.com/startuplessonslearned and Nivi and Naval from Venture Hacks are doing (http://leanpub.com/venturehacks) and they're doing quite well.

Really, I don't meant to be too hard on Keir. It was a good post, with interesting data. I think he has shown that giving away the e-book and selling the print book is not necessarily the way to go.

I'd love to see a lot more experiments and write-ups about this. I truly believe that authors can and will soon be able to make good money self publishing technical e-books.

I keep saying this: People don't like to pay for eBooks (yet). When I wrote the Rails Upgrade Handbook, I initially put out market feelers at $20, but most people said that was too expensive. I settled pretty arbitrarily at $12. It sold well, but as sales began to slope off, I ran a $6 sale for some time (stated as a celebration of Rails 3 stable I think and then just continued on just because), and wouldn't you know, sales soared for a while. I plan on implementing a pricing structure similar to what he's done here once Rails 3.1 hits next month; we'll see what that experiment yields.

My eBook on writing eBooks (http://authoringebooks.com) hasn't sold well by most mainstream standards, and I can almost guarantee the $49 price tag is partially to blame. Unfortunately, it'd be really problematic to have a print edition of a book on writing eBooks, so I'm just going to have to figure out a better way to market it. :)

Maybe one day soon people will be willing to pay more for eBooks, but at this point I can't blame them. You don't get a tangible good that you can give specific value to. With current rights scehemes, if you delete the file on accident or something, it's possible that you're hosed for good. The medium in general still isn't a solved problem, but maybe one day soon it will be...

I would sell the PDF for the same price as print (or slightly cheaper) and try 2 other ways for generating publicity/traffic.

1. Serialize the book as blog post i.e. post each self-contained part as spaced blog posts, with a link to buy the whole PDF at the end of each post.

2. Publish the whole book as html, but at any given time only part of the book would be available. I would (automagically) change which part e.g. every week.

Both those tactics are meant to drive search traffic, give away some of the value of the book but not all of it, so that people have a reason to buy the whole thing, either as PDF or as paper version.

Here's my impression:

1. Lulu is the easiest to work with. They accept the widest variety of files, don't require any special formatting, allow you to choose your price and discounts, and most importantly, update changes almost instantly. The only down side to using Lulu is that they don't have the same mindspace as Amazon or Barnes & Noble.

2. Amazon is the next-best. They'll accept .doc files and convert them for you, which means it's extremely easy to just write and upload. Their conversion is also very accurate; I didn't have to fuss with the file once I read the guidelines. On the down side, they take forever to update, so you need to check your information and read your entire book in the little preview window to make sure everything is perfect before you submit, otherwise you'll be waiting another 24-36 hours before you can submit a change that won't go in for ANOTHER 24-36 hours. They also don't allow you to download a copy of your own ebook to verify locally or on a Kindle before it goes live.

3. Barnes & Noble is the worst of the three, in my experience. They operate the same way as Amazon, but with some caveats. The first is that their conversion is nowhere near as good as Amazon's; I ended up copying all of the text out of my book and manually formatting it using CSS and HTML tags to get a book that looked the way it was supposed to, with images, line breaks, indentation, and text alignment. The second is that they, like Amazon, take a long time to update your listing. On the up side, however, they do let you download a copy of your ebook so you can look at it locally or on your Nook.

That having been said, how have sales worked out for me?

Over 60% of my sales so far have gone to Amazon. I attribute that to both the name (Amazon is a trustworthy brand) and also to the popularity of the Kindle. The remaining 40% have gone to the PDF version on Lulu. Both versions are DRM-free and the PDF version is slightly more expensive than the Amazon version ($1.53 compared to $0.99).

If you look at the site I'm using to advertise (www.branchrock.com), the Amazon button comes first, followed by the B&N button, then the Lulu button, which is simply labeled "PDF" to bypass any PC/Mac confusion. Perhaps the positioning of the Amazon button has also contributed to the number of sales from there.

I don't, unfortunately, have any statistics on how many people are reading on a Kindle vs computer.

I hope this helps other people who are looking into publishing ebooks themselves.

It'd be interesting to see what the outcome would be if he repeated the experiment with a title they were dying to publish. That would help determine if they're worth the huge percentage they take because they're good at promotion, or if all they're doing is detecting titles that will sell well on their own.

On a related note, I myself have developed a rule of thumb policy to don't buy technical books anymore because there is almost always an acceptable free way to get the info I need, and therefore this frees up more cash for getting things I can't get for free otherwise, or at least, to get higher quality or more convenient things. Like food, shelter, computer equipment, etc. I am still very willing and do frequently buy fiction books, in part because I know how hard it is to write fiction, how poor the pay often is, and because my interest is not a commodity/generic interest but rather a specific one for a particular story or author that is attractive. (I gladly pay for Stephenson, Scalzi and Simmons, for example.) Also, I don't like to handle dirty/secondhand books, and I've read a pretty big percentage of old public domain literature that I'm interested in -- if I haven't read something written before say 1850 there's a pretty good chance I never will because I've passed on it several times already, whereas there are still tons of books written even in the last 20 years that sound attractive from a distance but I still haven't gotten around to reading yet. When I want to Get Something Done I can often do tactical lookups in man pages, help docs, online pages, Googling around, asking somebody, etc. and very rarely have no other recourse than a retail book.

I'm also coming from the perspective of someone who did previously frequently buy technical books and build up a pretty large library of computer/software-related ones, and found that only a small percentage of them were fundamental enough to still be useful and must-keepable today (think Stevens TCP/IP Illustrated, many Addison-Wesley UNIX/Internet tomes). Whereas a lot of them were too tied to transient/temporary/version-specific technology (most of my O'Reilly buys, for example -- even though I love O'Reilly) such as Java 1.2-era books, old versions of Perl, early HTML books, books on Solaris (heck just saying a book on Solaris is antique and useless makes me feel old, I remember when I ran into it at lots of companies before the coming of the Linux wave), etc. Book buying is also a highly discretionary line item in my budget. Whenever there's any sense of a cash crunch or big upcoming need to spend money on more important things, book buying is one of the first things to disappear. Gas prices up 75%? A few less technical books bought and you're back in balance for quite a while.

All this said, I admire writers because I know how hard it is. And I think there will always be a non-zero-dollar market demand for quality written technical books, even if only in purely electronic form. I just think writers will need to increasingly be aware of what their competitors are, because there's going to be an increasing number of good free alternatives to their work, as well. And an increasing percentage of people are going to be giving away portions of their work for free on the web as a sort of resume++ and lead generator.

Anyone have any experience with this? Has a blog post or ebook created opportunities for you?

Why is S3 so important for file distribution? Couldn't you do about the same with some cheap shared hosting?