For a long time, product creation was a cause of anguish for me. Questions kept swirling in my mind. Am I doing the right thing? Will people buy? How will I actually sell it? It was mentally crippling.
Of course, I had no clue about selling. I didn't know how to bring traffic to my web site, I didn't know what makes people buy, and I didn't know how to make a product people want. Having spent a decade in software development, I thought sales and marketing were somewhat dirty words.
Go forward a couple of years, and sales of my self-published and self-promoted book C++11 Rocks have recently passed $10,000. And that's despite the fact it's written for a ridiculously narrow audience: experienced C++ developers who use Visual Studio and want to master C++11 (the latest version of the language).
This book was a result of my participation in 30x500, an online course for bootstrappers. It was extremely useful. It transformed how I approach product development and marketing in a way no other course, book or article had before then.
So here are some things I learned about marketing and selling a book.
1) Marketing isn't hard and mysterious.
My marketing basically consists of putting up useful C++ posts and cheatsheets on my blog. This is basically the cornerstone of the 30x500 marketing approach, and is neither voodoo nor rocket science.
The interesting thing is that I only published 13 posts in the course of the last 12 months or so. $10K in sales from this seems pretty good to me!
I link to my posts on Reddit, a few LinkedIn groups, and my Twitter account - that's it. My readers then share the posts if they find them interesting.
2) Mailing list size or Twitter follower count doesn't really matter.
In the last 1.5 years, I've grown my mailing list from a miniscule 250 subscribers to an insignificant 460. I've barely passed 100 Twitter followers. Does it matter? Apparently, not so much.
3) It's very useful to get noticed in the industry.
20-30% of the sales came from links to cpprocks.com on isocpp.org, a site dedicated to promoting C++.
For example, a few of my posts and my book page have been linked on isocpp.org - a site dedicated to promoting C++. These links created a significant proportion of my sales.
Getting noticed is also a result of creating useful content, but without any other effort on my part it took a while. I think it's valuable to put effort into establishing contact with people in the industry earlier.
4) Company/site licences sell, although not as much as I'd like.
20% of the revenue came from $499 company licences. This is good, but it's only about 2% of the number of purchases. It would be nice to increase this percentage, but I don't know how.
5) It's OK to ask people for testimonials.
I send out two followup emails to each person who buys the book: one with several useful links, and another one explicitly asking for a testimonial.
The response rate isn't high at 3-4% but I think it's better than I would have otherwise. And I don't really need hundreds of testimonials.
Note: A while ago I got advice from Joel Spolsky not to incentivise people to give testimonials in any way, and I still think it's great advice.
6) Some people will complain about the price. Ignore them.
I think everyone selling online gets these complaints. The best thing to do is to ignore them.
I had one email suggesting my book should cost $5 instead of $47. If I sold at that price, my hourly rate would be a couple bucks an hour. However, some people just have a sense of entitlement, or simply like to get things on the cheap. It's best to ignore them, and focus on the people who are happy to pay for the value they get.
7) Traditional publishers offer a crap deal.
I talked to a traditional publisher and could have gone with them. I was considering that as a way of outsourcing my marketing. However, if I'd gone down that path, it's likely I would have earned less.
The typical royalty percentage is low (10%) unless you sign up with Pragmatic Programmers. My royalties are close to 95%.
From what I read, most technical books don't get to $10,000 in royalties. The average royalties for books published in the US (all of them, not just technical books) are about $7,000.
To sum up, I count this book as a success (and I'm not done with it yet!), and I think that my sales are a testament to the 30x500 system.
As a side note, I'm actually selling 2 editions of the book, but since I released the beta of the 2nd edition, sales of the 1st edition have been almost non-existent. I could have written one edition, but two editions were a result of ruthlessly scaling the initial scope of the book down to a MVP for the first edition.