Are those popular iPhone apps popular because people want them? Or are they popular because they are more widely publicized by Apple's App Store? A study from Rahul Telang and Rajiv Garg, both of Carnegie Mellon University, sheds some light on that question -- and others related to how smartphone apps are promoted and become successful.
In their study, professor Telang and Mr. Garg looked at statistics related to popularity, measured by how many times each app was downloaded through the Apple App Store and how much money was paid for each. Their goal: to determine how many downloads an app developer can expect based on various circumstances.
It's a complex equation because the data is somewhat shrouded by Apple and changes frequently. It's also tough to determine because many apps cost money to download, while others are free but create additional revenue streams after download.
A good example of the latter is Amazon's Kindle app for iPhone, which allows users to read their Kindle-based books on their phones instead of on their Kindle book readers. You can download the app for free -- but if you want to read a book, you have to pay for that book.
Some apps create additional payment options through the Apple App Store and others, from separate websites. To the authors of the study, that meant taking into account all the dollars paid, even after the download had already been done.
The key result is that the researchers have been able to create a mathematical model that lets them predict the relationship between download rank and revenue, using just publicly available data.
Apple reveals the ranks at least daily of the top 200 apps. Mr. Garg said those that make the top 200 list stay there for an average of 100 days but that the most popular apps stay much longer. For instance, Angry Birds has been listed for more than a year -- quite a feat. He says if you debut at No. 200, you're likely to fall off the list very quickly; but if you debut at or near No. 1, you'll stay on much longer.
Mr. Telang says the study verified that consumers are more likely to download highly rated apps, just based on their ranking; although he says there is still more work needed before he can determine some specific effects.
He also traced the pattern in which the number of downloads changes as you head down the list -- in a shape called a Pareto Curve, commonly known as the long tail. Going from No. 1 to 2 creates a substantial drop; from 199 to 200, not so much.
An app ranked No. 1 for iPhone might be downloaded 52,500 times each day; while the 200th ranked app will be downloaded only 350 times that day. Similarly, a No. 1 app for iPad is worth 9,500 downloads a day, while No. 200 will only be downloaded 80 times.
While consumers may see the results of the study as verification of what they already believed, developers can use it to help determine how much money to spend for promotion and what pricing model should be used to push their apps up the rankings.
For instance, how important is it to push an app to the No. 25 rank where it is seen on the first screen of recommended downloads vs. spending less money and effort to keep it at the number 26 rank, where it doesn't get seen unless the user pages to the second screen listing the next 25 most popular apps?
The authors plan to continue their academic research.




Comments
Welcome to Post-Gazette subscriber-only commenting community. Please review our community guidelines.