Pinch Media collects stats for iPhone apps, e.g. Clayton uses them to track usage statistics for the Oracle People Search app.
Last week, they released an analysis of apps that has some pretty interesting metrics. The one that caught my eye was that for free apps, only about 20% of users return to the app the day after they installed it. TechCrunch calls this “shelf-life”, which seems as good a term as any.
The shelf-life declines precipitously every day thereafter, creating what looks a lot like a Pareto distribution, a.k.a. a long tail, more on that in a jiff.
The Pinch Media slide deck is embedded below for reference, and here’s that particular slide.
The analysis is based on metrics Pinch Media collected from the apps developed by their users. Their analytics package is free and is being used to track a couple hundred apps, translating into more than 30 million downloads. Seems like a good sample size.
I love graphs and data. So, I immediately mapped this to my own experience with iPhone apps. As I’ve said before, the apps I keep and find most useful accomplish units of work for me, e.g. Flashlight, Oracle People Search, TwitterFon, FlickIt, Call a Cab, OpenTable.
I don’t find apps that do more than a single task very useful, e.g. Facebook, Brightkite. I’m also not that fond of browsing the ‘tubes on the iPhone because it’s a bit slow (remember, I’m still using my OG iPhone on Edge) and the browser is small. Yeah, it’s better than anything else I’ve used, but it’s still small. The speed thing bugs me the most.
My usage probably fits that curve nicely. Apps fill the units of work that I occasionally have, e.g. I have to look up an Oracle person’s phone number, I have to get a cab, I need to find the lock in the dark, I need to tweet about the kit car parked in front of my house, etc.
I use my iPhone all the time though, so what do I do with it? I make calls and send email. Sometimes, I listen to music.
Analyzing my overall usage iPhone of apps, including the ones that Apple installs, maps nicely to a Pareto distribution, i.e. use on the vertical axis, number of apps on the horizontal.
I use a very few apps a lot. This is the head.
I use a a lot of apps infrequently, trending toward not using them at all. This is the long tail.
There’s no real point here, just a data nerd’s confession of joy at mapping stuff on a graph. How does your usage of apps fall on a graph?
You’d think app developers would want to fall into the head area of the graph, but my guess is those apps will always be subject to Apple’s whims, i.e. if it’s a popular function, Apple may just add it into the next firmware.
I think the money lies in units of work. As the slide deck suggests, usage metrics support charging upfront as a superior model to free, supported by ads. So, the key is finding the right itch to scratch.
What would be interesting is to see overall app (including Apple’s) usage. I’m curious to know if the Phone app is the top app; I would think so, but you never know. Maybe Email is. Or iPod.
Anyway, the slide deck is interesting. Clayton’s app is nearing 1,000 unique users, so I may blog a bit about that later this week. Seems like a good milestone number. He seemed a bit disappointed by that number. Stay tuned for more on that.
Sound off in the comments about iPhone app and usage. Do you think they’re worthless, like Topper does?
