Mobile apps need to measure their success with their users, says Ben Hsieh, CEO of mobile consultancy firm 3Play Media.
But that data has to be consistent across platforms and devices, and there’s a lot of uncertainty around what that means.
That’s why Hsieb and his team are using an algorithm called the median formula to help determine how many people are getting the best results for their apps.
This chart shows the median app performance for the top 50 U.S. mobile apps (according to a survey of 3Play clients).
The vertical axis shows the average app performance, and the horizontal axis shows how many times the app has been installed.
Read more: How to fix iOS app metricsThe median formula was developed by a team at 3Play, a consulting firm that works with many of the most popular mobile app developers.
It is based on a metric called “play time,” which measures the number of plays a user has in an app over a period of time.
Play time is the time between a user playing an app and the time it actually plays.
Hsiea says this data can be hard to interpret, because it doesn’t take into account how long the app takes to load, or whether it has a high download or low upload rate.
“You can’t say, ‘OK, the users are going to have a lot more play time than they actually do,’ ” Hsiehe says.
“If you are not tracking this time, you are probably not getting the metrics you need.”
The median formula is the first step in figuring out what users want.
But if the data is consistent across devices and devices with similar behavior, Hsiec and his company can estimate which users are most likely to experience better performance.
Hsieh says the median is more reliable than the standard performance metric, and it’s an easy way to know which apps are getting more plays.
“It’s the only metric that we use that gives us confidence in the app performance,” he says.
Hsui, 3Play’s CEO, says that in his experience, “it’s very clear when an app is performing well.”
The company’s algorithm uses a number of factors, including app performance in the first month, app download size, the number and type of players in an average user’s family, the user’s current location, and other variables.
The median calculation gives the company a measure of which apps have a chance of performing well in the future.
“In the case of iOS, it’s pretty clear what the median should be based on the data that we have,” Hsiei says.
3Play then compares that median to what apps users have downloaded, and in the case that the median does not match what the user is downloading, it breaks down that user’s play time into chunks of a few seconds each, to give users an idea of how much of the app’s time has been downloaded.
Hsingh says it’s a good way to compare apps that have similar performance across devices.
3 Play, which was founded in 2010, now has over 2 million users across the globe.
“We believe in the median,” Hsien says.
3Play also provides advice to the mobile app industry on how to improve their metrics.
The company is using the median to help it create a new metric called Play Time Metric.
3play is developing a platform that will allow app developers to submit their apps for approval.
3 plays will publish a new “App of the Week” each week and ask users to rate each app on a scale from one (the least positive) to 10 (the most positive).
The app developers will be able to choose from one of three categories: “high, low, or neutral.”
3 Play is also creating a platform where users can ask 3Play to share its findings and recommendations on improving mobile app metrics.
“They can also contact us directly to ask about how to do things better,” Hsingi says, noting that the company is working on a new app review tool for iOS.
As 3Play sees more apps becoming the norm, Hsingy says, the company will use the median as an indication of which platforms are performing better.
“That is an important thing for us to look at, and we’ll keep on doing that,” he adds.