I'm quite well disposed towards the technical book publisher O'Reilly. I've been a customer of theirs for almost two decades now due to the high quality of their books.
They've always been quite "geek-friendly," too, so it was not particularly surprising (though of course very pleasant) to find that when they started selling e-books, they chose a strict no-DRM policy. That took some of the sting out of the prices; somehow it feels as if, for a $45 paper book, paying $35 for the e-book is too much, though the manufacture and shipping of the paper book probably do cost less than $10.
I received a marketing e-mail from O'Reilly recently with a new, apparently on-going promotion: their e-book Deal of the Day RSS feed. Every weekday they discount one of their e-books to $10. For the more expensive $30 and $40 e-books, that works out to a substantial discount.
Whether they'll be able to make money from this, I don't know. But in my case, at any rate, it's certainly generating sales they would probably not otherwise make.
$10 or less is about my "worth taking a chance" price point. At that price, if the book looks moderately interesting and appears to be more or less relevant to my needs, I'll probably just buy it on the spot, even if I don't intend to read it immediately. For books costing much more than that I need to have a reasonably serious need for the book, and I also need to look through it (and perhaps check out some reviews) to investigate its quality. I may do this right away or, if I don't have an urgent need for the book, I'll likely put it on my "things to look at later" list, especially since in the fast-moving field of computer books a new edition or better alternative might appear at any time.
As I've mentioned previously, Skip's comment here is important. Even if the book is one I would later buy after doing the research, that's only going to happen if it comes to my attention again at the appropriate time. And even when I try to make sure that this happens, it often doesn't. I have a long list of interesting-looking books spread across various to-do lists, calendar entries and other files that fell into that "not ready to buy yet" category and remain unbought, some of them years later.
So what has O'Reilly's program sold me so far? Beautiful Testing (RRP $40) fell into the "check it out first" category and would have gone on to my to-look-at-later list if it hadn't been $10. I still don't have any idea when I'll have time to read it, but it's there whenever I care to get to it, and even if I never read it, I'm not out a huge amount of money. Apprenticeship Patterns (RRP $24) is a book I don't need now, but might need in the future. It too would have gone my to-look-at-later list, and might never have gotten off it, but it's worth $10 now just because I might need it later, or just get even a few good ideas out of it for myself. The Art of Concurrency is not worth the RRP of $36 to me (I'm already quite familiar with the subject matter), but might have $10 worth of useful new information. The New How (RRP $20) is a book which I probably would not have bought at all for more than $10 (and it just barely scraped in at that price). The topic matter is interesting to me, but not something I'm likely to use any time in the next few years, and I have no idea if the book is any good at all.
So O'Reilly's come out with $40 in revenue from me this year so far, as opposed to some possibility of $40 and a lesser possibility of $24 at some indeterminate points in the future. That's certainly left them ahead in this case.
But I'm only one person. The real questions here are, does this marketing strategy have the same effect on other people, and how much will O'Reilly lose from people who delay a full-price purchase they'd otherwise make on the chance that the book may come up on the daily discount list soon?