(E)Book Pricing, Part 2

The folks over at Wrox are running a pricing experiment. They’re discounting their ebooks by 50% to see how much extra volume they get, and whether they can convert people who steal ebooks into paying customers. While this is an interesting exercise, and will hopefully give them some good data, there are a couple of caveats.

  1. 2 points may not make for a very good demand curve, especially if one point has twice the price of the other.
  2. Many people may be influenced by the promotion of “50% off” as much as the change in price. They may become acclimated to the new price and require further promotions to drive changes in buying behavior.
  3. The variable cost is next to nothing, and they are trying to develop a category (ebooks) that could bring them higher margins even at lower prices, so even with points 1 & 2, this may be a helpful longterm promotion. However, this would NOT work in markets with significant variable costs, including physical books.

Does anyone else have examples of favorite pricing experiments that they have run or seen?


  1. Jon


    My position is: if you’re not doing pricing experiments, you’re not doing pricing! Pricing research is OK, but its imprecise & imperfect. At least pricing experiments tell you what customers actually do…not what they say they will do. Experiments often work out cheaper than over-priced conjoint and discrete choice research too.

    I’m constantly devising pricing experiments for my clients, which for reasons of confidentiality, I’m not able to disclose.

    But one of my favourite pricing experiments, which is in the public domain, was one run by the airline Hapag-Lloyd on 2nd May 2004, when they commenced flights between Newcastle (UK) & Hanover (DE). Sixty passengers booked a 3 day stayaway on the inaugural flight and we told they didn’t have to pay until the return leg…when they could pay what they thought the flight was worth.



  2. Jim Minatel

    Reuben: Thanks for noticing our price experiment and yes, as you’ve pointed out there are some flaws in the experiment method but we are collecting some very useful data. What I would point out though is that the variable cost is not as low as you might think, especially compared to print costs. The authors still get a royalty percentage on this, which is usually as high or higher a variable cost than the physical print costs. And in the case of current PDF ebooks, we’ve also paying a per transaction DRM cost that isn’t as low as you might guess. So especially for driving good return for authors through their royalty, there probably is a floor on the pricing.

  3. Reuben Swartz

    Jon and Jim– thank you for your thoughtful comments.

    Jim, in particular I want to say that I am not criticizing your experiment, just warning folks that this is not the typical way to do price experiments.

    (Nor is the recent fire sale on housing by Hanoverian that an alert reader brought to our attention. They slashed home prices by up to 30% and sold almost as many units in 3 days as they had in the prior 3 months. If you think the people who bought an iPhone before the price drop are upset…)

    Best of luck with the efforts– I’m a big fan of ebooks, especially for reference-type volumes that you want to be able to search easily.

    You may have heard that Harlequin is moving much of their romance catalog to ebooks, as well, with lower pricing than print versions. This probably won’t cause a lot of direct competition with your efforts. 😉

    Please let us know the results.

Comments are closed.