I make videos and ebooks, and sell them on my blog. People like Patrick McKenzie and Amy Hoy constantly reiterate the advice that you should raise your prices and charge more than you think. So I took this advice. I put my first video on sale for $39, even though I was initially thinking of $19. Then I got an email from a happy customer who bought at $39 and wanted to tell me that my video would be underpriced at twice the price, so I doubled the price, and then increased it a little further, to $97. The video continued to sell very well, so I was happy.
With subsequent products, I made it a rule to charge more than I thought was reasonable, and also, to increase the margin of unreasonableness with each new product. (I also made it a rule to always set my prices at prime numbers, with only one exception, but there's really no good reason for that, and it's probably not important in this discussion.) Some of my products did well, some of them didn't, but when products didn't do well, the reason always seemed to be content or marketing, not price. My pricing strategy did not seem to have any downsides at any point. In fact it worked so well that I now consider charging unreasonable amounts of money to be a major theoretical pillar of my business. But I also noticed that Amy Hoy's husband Thomas Fuchs frequently tweeted discount codes for his ebooks (in particular the Retina one, which is quite good), and I also got a few tweets and complaints that I was charging too much, so rather than say "well yeah, that's the whole idea," I decided to run a sale.
One caveat before the story proceeds: I have a 100% no-questions-asked refund policy. If you have the slightest dissatisfaction with your purchase, you can have your money back. I probably wouldn't implement this kind of pricing strategy -- "charge more than you could possibly imagine people being willing to pay" -- if I didn't also have this approach to refunds. Because with this approach to refunds, charging as much as you possibly can is pretty much just an honest exercise in not knowing what the correct price is, choosing a number semi-randomly, and finding out. If I had a whole bunch of rules and hoops for people to jump through in order to get their refunds, it would be less pleasant, less ethical, and also a much less reliable source of feedback. There would be too many variables involved with that type of feedback for me to call it useful information about price and only price.
Anyway, as you can probably guess by the fact that I'm running a new sale right now, my sale went well. I did several other sales and they all went well. In fact I did one or two which were based around live-tweeting quick runs to the grocery store. The sales began when I left the house and ended when I got home and unpacked my groceries. The first one I did as an experiment, because I couldn't decide if I should work on my business or buy some groceries, so I figured I'd just do both at the same time, and then I did the second one because the first one went well.
The reason my sales work well might be because I set high prices when I release my products. For instance, here's some feedback about my Ember book:
I actually agree with both these tweets, but the Ember book sells at its normal "too expensive" price, so after this sale, it will of course go back up to that price.
The thing is, if you're in a hurry to learn the essentials of Ember, because you're really excited about it, or maybe you've got a big new consulting deal which will net you thousands and thousands of dollars if you can produce results with Ember right away, then my Ember book is easily worth the $47 I first charged for it, on day zero. You'll make that money back before you've finished your first cup of coffee on the first day of the new project. But if you're a more cautious person, or you're just sort of curious about Ember, in an academic sort of way, with no looming deadlines, you might not want to pay that much, and you certainly wouldn't have the same immediate, compelling incentive that makes it an indisputably great business decision for other people.
But you might still be down to buy it if it's on sale.
This is basically the easiest form of customer segmentation ever. Your customers who are willing to pay top dollar will pay top dollar, and your customers who would only buy at a lower price will buy when you offer the lower price. Best of all, if you retain records of how many units you sell, and at what price you sell them, you can probably come up with some math which will tell you what your optimal price should actually be. It's good to be able to draw upon data when you want to answer a question like that.
Another caveat, however: discounts don't actually have a huge influence on the sheer number of products sold, for me. They drive short-term sales spikes, and they enable me to do price differentiation, but the majority of my products sell at their higher original prices. Probably one explanation is the fact that my discounts have narrow windows, like a week, a weekend, or even just the time it takes me to get to the grocery store and back. It's not a pure, scientific experiment, after all; there's no control group.
But there's another, simpler explanation: the "set high prices" advice from Hoy, McKenzie, and others is good advice. I follow it, and I recommend you do the same. My products are almost always about programming, so if you use the information in them well, you'll make enough new money to cover the purchase price very rapidly. This is what the brilliant information marketing pioneer Dan Kennedy -- who invented nearly all this stuff back when it was all mail-order and physical books -- calls "buying free money." Say your new Ember contract is worth $47,000 to your consulting business. Would you pay $47 to get $47,000? Of course.
PS: my sale's here.