The Most Popular iTunes Apps Aren't Always The Cheapest

While the most popular aren’t always the cheapest, on average, the Top 10 Paid apps tend to be cheaper than less popular ones (those ranked 45 to 55 or 91 to 100):


The situation varies across categories and in this post I’ll briefly examine a few of the larger ones. In both the Books and Games categories, the mean price of the Top 10 most popular paid apps tend to be lower than less popular ones. In other large categories, such as Navigation†† and Travel, the situation isn’t as clear: the mean price of the Top 10 most popular paid apps aren’t always lower.


(Click here for a larger version of the chart above.)

Since the mean tends to be susceptible to outliers (a few high-priced apps), I decided to graph the price distributions for the top paid apps in the categories displayed above (click here for the graph). I looked at statistical densities††† on three dates: 3/8 (24 weeks ago), 5/31 (12 weeks ago), and last week. In the Book category, the top 10 paid apps now seem to be dominated by lower-priced (99 cent) titles. In the Game category, the top 10 game apps were comparatively lower-priced 24 weeks ago but things have changed slightly: the top 10 game apps are no longer substantially cheaper than less popular ones (rank 45 to 55, or rank 91 to 100).

(†) I refer to a paid app as being in the Top N, if it was listed among the N most popular apps, sometime during the given week.

(††) For display purposes (i.e. to avoid distorted looking graphs), I omitted a couple of popular (top 10) but unusually high-priced Navigation apps (MobileNavigator and TomTom).

(†††) Based on small samples, the approximate densities drawn are far from robust, but they provide another tool for comparing categories. Boxplots over time would be another method.

tags: , ,
  • Fascinating analysis.

    Taken further, this analysis can inform pricing efforts. For example: it would be interesting to see how much total revenue is being generated by an app. Also, can app developers modify the price to maximize popularity or maximize revenue?

    Is the raw data for this generally available?

  • bowerbird

    money. is that all you guys care about?


  • The conclusion is, in the more “serious” app categories, people are more willing to spend money on a good product.

    Higher priced games are also popular if they’re a good app.

    Nothing can beat the sustainability of a good app with a good brand name.

  • Interesting analysis with a not totally unexpected finding: if i would be the owner of a in-demand app, i would necessarily give it away for a relative low price, excellent stuff has it’s price.

  • Your chart seems to imply that Top Paid or Most Popular rankings do not factor in price. (Which chart are you using? You use both terms. And these are two different ranking types used on the App Store.)

    I’ve recently seen it alleged elsewhere too that the App Store charts are “unfair” to higher priced apps because they do not reflect price.
    “These Top 100 ranks are based on unit sales rather than revenue and are therefore skewed towards lower priced applications.”

    To the contrary, in my studies of the “Top Paid” apps lists and “Most Popular” rankings on the App Store, it looks like these lists DO indeed factor in price. (i.e. They do not just reflect unique sales. There seems to be a weighting given for app price.)

    Also I’ve read that “Top Paid Apps” reflects a 24-hour rolling window. Clearly the “most popular” rank and “Top Paid” charts do not match (in addition to the obvious fact that “most popular” includes free apps). So perhaps the “most popular” rank reflects a longer window, thus giving successful apps some “momentum” on their lists.

    Are these your impressions?

    I’m sure the iPhone analytics companies as well as large vendors who have had multiple apps on the same chart know the answers to these questions because they have access to multiple individual app sales numbers as well as the App Store numbers. It would not be difficult to extrapolate the formulas from that data.

    It would be nice if small, indi developers (like myself) had access to those formulas in order to tune their pricing in this ultra-competitive market.