We’re doing some maintenance on the parts of appFigures that handle syncing and uploading reports as well as the archive. Everything should be back up by
9pm 2am EST. While this is going on you may see empty region reports and other weirdly displayed data, but have no fear your data is as safe as can be.
When it’s all behind us you can expect a genius importing engine that can eat even the most disgustingly mangled reports.
A few months ago we (somewhat) sponsored and (mostly) hung out at an incredible iPhone development conference, 360iDev.
Part of being a sponsor meant we could throw a Tuesday-night nerd party. So naturally we chose to have a sumo competition for an iPad. Recently a video of this abomination to good taste surfaced and we knew we needed to show it to you.
This was the final showdown. A lot was at stake. Men were made and broken in inflatable suits. To the winner: everlasting glory and a smooth, sexy electronic joy-machine. To the loser: a beer and a promise that we wouldn’t tell his wife.
Take a look:
Sometimes you just don’t want a monthly recurring charge. Until now that was the only way we could bill you: one month at a time every month. Well now you’ve got another option. We just released the ability to prepay your account with your credit card or with PayPal.
As you may or may not know, appFigures is sponsoring 360iDev in April. It’s a great iPhone development conference in San Jose, and as sponsors we get to throw a really wild party on April 13th. The problem is that for a really wild party you need the right beer.
We’ve been kicking ideas around the office and can’t come to any sort of agreement, suggestions run from Arrogant Bastard Ale and Corona to Steel Reserve and paint thinner with flour. We figure if we can’t come to any sort of consensus, we should let our users and the conference attendees decide, and what better way to do this than through the internet-magic of twitter?
So here’s the deal: between today, March 23rd and April 1st tweet @appFigures with your beer suggestion and tag it with #ibeeripoll
We’ll take the top 5 most tweeted and retweeted beers and bring them along to the party, so get excited and may the best beer win.
Here are some suggestions (click to tweet your vote):
We couldn’t help but notice our community being all a-twitter about huge, unexpected rank changes in the last two days. We thought we’d break out the old calculator and examine it a little bit.
As you know, appFigures tracks top-200 ranks in App Stores around the world. We took this data and looked at each app’s changes by comparing the hourly change of each app and taking the largest hourly move per day. We added up all the apps that had a big change > 50 (up or down) and compared those numbers to the same days of last week. The goal was to get a picture of high-magnitude changes in rank, the sort that would seem unexpected. The numbers spoke and told us that something was not right.
Tuesday was slower than usual…
Wednesday, and Thursday were crazy!
Let’s get numeric:
During a normal day, 5.77% of apps experience an hourly rank change of more than 50 places (up or down). On Wednesday, that number shot up to over 70%, indicating that it was a particularly volatile day in terms of high-magnitude hourly moves. Thursday experienced similar churn, with nearly 50% of apps experiencing these high-magnitude changes.
Here’s an overview of the last 30 days on the App Store. Look at that NBA-level jump…
Daily % of apps that moved 50 or more places in one hour.
And here’s a direct comparison to last week:
Apps that moved 50 or more positions in one hour.
Then we dove a little deeper into the nature of the changes, taking not only hourly rank movements of 50+, but breaking them down into four groups: 0-50, 51-100, 101-150, and 151+ places. On normal days, 94% of apps see no hourly rank change greater than 50 places. A bit less than 2% see a shift of 51-100 places, and the same for 101-150 and 150+. On Wednesday and Thursday, only about 40% of apps stayed within 50 places, with around 20% seeing a change of 51-100, and likewise for 101-150 and 150+ places. Take a look at the breakdown:
This is what a normal day on the App Store looks like:
Maximum change in rank for 1/25-2/1
This is what Wednesday and Thursday looked like:
Maximum change in rank for 2/3-2/4.
We were puzzled as to why the big movers were distributed relatively uniformly (in that the percent moving 51-100 places ~ 101-150 places ~ 150+ places), rather than showing a more normal distribution, but that’s for another day.
Another peculiarity we found on 2/3 and 2/4 was the average highest hourly rank change per app. That number was consistent around -8 places up until 2/2. On Wednesday and Thursday it jumped dramatically to around +25 places each day. So, though the app store has shown big swings on those two days, an app’s biggest hourly swing is on average 25 places up.
Well that’s it for now, we’ll have some more analysis coming soon.