Justin Williams has excellent insights on some of his approaches to App Store marketing. Specifically, on why he doesn’t read reviews, do app sales or randomly distribute promo codes.

I agree completely with him on the reviews and codes.  While I’ve gotten a lot of nice reviews, they do tend to skew toward people who just want to complain and their input isn’t usually terribly productive.  I’ve also tried some promo code giveaways, and can’t say I can figure out and meaningful benefit from them.

The app sales issue is a little more interesting. I’m not sure I have any hard conclusions regarding the value of sales, but I have experimented with sales on Terminology and thought some real data would help drive us from opinion to analysis on the topic.

Below are two charts showing the results of a recent one week sale I ran on Terminology for iPad.  The regular price of the iPad version is $2.99. For the week starting Friday, March 11th, I ran a $0.99 sale to welcome the arrival of the iPad 2.

The first shows the effect on the ranking of Terminology in the iPad Paid Reference Charts for the two week period surrounding the sale.

The second shows the effect on unit sales and revenue during the same period.

The basic trend here is consistent with other sales I’ve run in the past. The first day of the sale draws a significant spike in unit sales, typically ~3x.  Then there is a quick drop and it levels off for the remaining days of the sale. Typically, I’ve seen it level off where unit sales remain above non-sale levels, but revenue slightly decreased.

I attribute the quick spike primarily to the many websites and social media robots dedicated to announcing “App Store Price Drops.”  There is a market for people who really like to feel like they are getting a deal.  Frankly, at App Store prices they are already getting incredible deals, but that doesn’t defeat the psychology of the bargain hunter.  I think that these are sales that would not otherwise happen, so some benefit is gained from the sale pricing.

After that first 1-2 days, I think it’s questionable whether you want the increased unit sales.  Revenue is slightly decreased, and without the option to charge for upgrades or contact your buyers in the App Store, there’s no gain in simply having a larger user base.  If I do run any sales in the future, I don’t intend to have them run longer than two days.

The exception to that rule is for a new app launch.  I think introductory pricing is a significant selling point, and it’s far more important to make an appearance in the charts and draw App Store feature attention early on.  Also, I think it’s less likely to generate the sort of uneven user toll feedback that Justin mentions.  Early adopters should get a a benefit, and are the users most likely to help spread the word about your app.