Update: December 12 – see the bottom of the article for a different, possibly more accurate graph..
Are 99¢ apps harming the iPhone app ecosystem?: The Pragmatic Argument
Two days ago Craig Hockenberry wrote an open letter to Steve Jobs complaining about the prevalence of 99 cent apps on the App Store and the “rush” to that price point. He suggests the rush to the bottom is damaging the chances of more significant apps being developed for the platform. We’ve covered the “gold rush” on the iPhone App Store before, and dissected the App Store data to discover that the cheapest apps aren’t the ones making the most money (sounds obvious, but many of the complaints assume it isn’t so).
99 cent (or free!) apps do no more harm to pricier iPhone apps than open source or shareware apps do to commercial PC or Mac software. But the point still stands.. are 99 cent apps really that popular on the App Store? In terms of quantity, there are a lot, but are they significantly more popular than more expensive apps?
Popularity vs Price Band: The Data Argument
We ran the data from the iPhone App Store through the Mobile Orchard supercomputer and produced the graph above. It shows every iPhone app (free or otherwise) plotted in one of three bands (free, 99¢ apps, and more expensive apps). The y-axis within each band is random and just gives the data more room to breathe. The x-axis is an app’s “popularity” in terms of downloads – shown logarithmically (the App Store provides popularity information within a range of 0 to 1 and > 90% of the apps have a popularity of under 0.01, so we’ve plotted the data logarithmically otherwise you wouldn’t be able to see anything).
If you look at the different popularity bands, you’ll see the 99 cent iPhone apps occasionally have a very slender edge. For example, there are two 99 cent apps in the 0.5-1.0 band – with none from the higher price band. Look at the 0.03-0.05 band, however, and just 32 99¢ apps are represented against 25 from the higher band. This isn’t a whitewash for the 99¢ apps by any means. Visually, you can see that even though 99 cent apps appear to have the edge at higher levels of popularity, it’s just not significant and if you were to multiply the price by the sales obtained at each level of popularity, the higher priced apps would beat the 99 cent apps in overall revenue (we looked at this before in the Games category).
Variations between categories
Above, we looked at the iPhone App Store in general, but the difference between the popularity of 99 cent apps versus higher-cost apps becomes more pronounced if you hone in by application category.
Let’s run the data for several categories through the Mobile Orchard supercomputer:
In the Business and Productivity categories a clear preference for higher priced apps is shown. In the Games category it’s less obvious, but visually the 99 cent band is no stronger than the higher priced band. The Entertainment graph, however, shows that 99 cent apps are certainly a lot more popular there than the higher priced apps.
The Social Networking category shows a strong preference to higher priced apps. This category seems to have quite a few solidly-priced instant messenger apps and in comparison to the 99 cent apps, they’re doing very well. As always, however, the free apps completely trounce the pay-for apps.
Where next?
Seeing the differences between application popularity and price visually is very appealing, but we’re now working on processing this data numerically to produce “willingness to buy” scores for each of the application store categories. Which categories have people spending the most and what sorts of applications are people most likely to pay for? This information will undoubtedly be of use to you if you’re an iPhone developer or speculating on what to develop for the platform.
Update
Graph with all price points (> $50 mixed into top line) – click on the graph to get full size
There’s no analysis on this just now, but to help address those who want to see how it looks if price is more accurately traced on the Y-axis, we have this very quickly produced graph (same process as the others but with the y-axis accurately portrayed and no scale shown).
It is worth noting, however, that the main part of this article address whether 99 cent applications are more popular than all other higher priced applications – which is why the bands are defined as they are.
