Having unsuccessfully tried such a pricing model myself, I have serious doubt about it's viability. Majority will not pay anything, very few will pay reasonable amount, and some will pay very little but will be very high maintenance.
Payer: God dammit, I paid for your service and I want one hour of your support time. Company: How much did this payer paid? 50 cents! We need to hire someone who will take 30 cents to support this payer for an hour.
This is one of those pricing strategy that sounds good and people will claim to support but in reality wouldn't.
Leanpub and Gumroad are two services that follow this strategy. I don't know if they are successful ventures, but I notice they have maintained this pricing strategy since their inception and never changed or stopped it.
Leanpub says "...the pattern we have seen with multiple books, about a third of people will pay the suggested price, a third will pay the minimum price and a third will pay somewhere in between. Finally, the odd person will pay more than the suggested price" [1]. Some anecdotal experiences of individual authors [2],[3],[4] support it.
[1]: https://leanpub.com/help/author_faq [2]: http://blather.michaelwlucas.com/archives/1626 [3]: http://blog.ppenev.com/2013/07/16/what-i-have-learned-from-l... [4]: http://blog.gumroad.com/post/75707736685/is-pay-what-you-wan...
When coupled with our sliders which clearly show the (very high, 90% minus 50 cents) royalties that are earned by the author(s), and with our free update distribution, we get a lot of benefits for our authors, including:
1. When there is a low or a free minimum price, readers who otherwise could not afford a book can get it legitimately. This can help build a community around the book. For example, https://leanpub.com/rprogramming has over 80,000 readers (free + paid), and while it has a free minimum price, it is also our top book by lifetime revenue.
2. Quite often, there is so much extra income earned by readers paying above the minimum price that it entirely offsets refunds. (We have a 45-day 100% refund policy.)
3. Because of the royalty transparency, variable pricing, 45-day refunds and the fact that readers are supporting authors by buying in-progress books, there's a very, very low amount of nastiness. I really like seeing author and reader email; quite often "I love Leanpub" is part of the sentiment, regardless of whatever issue is occurring. (We also sell completed books, but much of what Leanpub is based on is our writing workflow and our focus on selling in-progress books...)
Your page for Roger Peng's R Programming book also contains a link to Lulu for those who prefer print.
Does following that link bypass your system or are you collecting some kind of fee for that? I'm assuming Dr Peng gets a reasonable chunk of coffee money from Lulu directly as usual.
Cheers
That said, having used Lulu myself in the past (back in 2006), Lulu pays good royalty rates on print books and ebooks, and we recommend that all our authors consider using Lulu or Amazon CreateSpace to produce a print book once their book is done. If you search "Leanpub" on Lulu, you'll see a handful of Leanpub books there.
To help our authors produce print books, we have two features:
1. Print-ready PDF export. Click a button, get a PDF with proper page numbering (alternating sides), chapters always starting on the right page, no cover image (since Lulu and CreateSpace have wraparound cover upload features), etc. I think that most of our authors who have print books, including Dr. Peng, went this route since it's a lot easier than the second choice.
2. InDesign export. Click a different button, get InCopy (ICML) files. These can be given to a designer who is good at InDesign, and he or she can make a beautiful custom-designed book. Our InDesign export is pretty basic compared to the print-ready PDF export, but our hope is that it's a better starting point for a designer than a Word document with a bunch of formatting that they need to throw out.
For us, making the print book production process as easy as possible just adds to the value created by using Leanpub. We want to create more value than we capture, and this helps with that...
Cheers
A quick 'back of the envelope' estimate of revenue and cost of BandCamp service (enough public info available) seem to suggest this pricing model might have low profitability.
Not sure if it's applicable outside that market, but it might.
I don't know if unlocking everything by default would work that well.
Also, funny thing, looking at payment logs you can clearly tell the average is often purposefully inflated by some people (I guess the dev studios).
I feel a more apt pricing strategy would be to provide free/low-cost yearly access to students and a free basic course + paid advanced courses to non-students.
Now that I think about it, I would like to see that as an option.
They along with their institutional partners are missing out revenue from people like me who'd like to pay some amount, but not the amount they fix. I'm happy to pay some amount without a certificate. They should consider giving a pay-what-you-like option for all their courses.