We launched Circulets on May 9 on the Apple App Store and the Google Play Store. It was expected to show up on the Amazon App Store and the Samsung App Store as well. We were working with these two stores for the first time and did not clearly understand their processes, resulting in some small snafus that made the game appear later on them.
In this post, I am only going to talk about numbers about our game’s performance on the store (basically downloads). We have no available reference for such numbers, so we cannot benchmark this market performance, but the objective is to share our lessons learned and get your help (hopefully, you’ll leave a comment) in making sense of this data.
All data in this post comes from the following sources:
- Stores: Apple App Store, Google Play Store, Amazan App Store, and Samsung App Store
- Analytics: App Annie analytics for the stores and game; Flurry for in-game data
A little chaos…
There is a little chaos in the domain for numbers from the mobile market. Here’s why:
- Apple App Store gives me data for the previous day at 5pm (IST) of today. So there is approximately a 17 hour delay from real-time data.
- Google Play Store gives me data for the previous day at 8pm (IST) of today. So there is an approximate 20 hour delay from real-time data.
- App Annie reports data after a delay of 48 hours, ie when the stores report yesterday’s data, App Annie reports day-before-yesterday’s data.
- Flurry on the other hand updates every 6-8 hours, but only reports data from online users. And is mostly at war with all the data out there.
I am going to assume that the most reliable and useful data is coming from the Stores (irrespective of any data that is what I will eventually be paid for). Even though this is at times at a constant conflict with data from everywhere else or what could be logically correct, the data from stores is going to be my gold standard.
Any data from Flurry and App Annie will only be taken as indicative by me, I will not base any conclusions on them.
So to begin with, Sales/Downloads. Over the first week (May 8 – May 16), the game was downloaded a total of 17,460 times out of which 17,345 were on the App Store, another 114 on the Play Store and just 1 on the Amazon store.
The spike that you see came from our sale on World Family Day, when we made the app free for a day on the App Store.
Here is how we fare in our top countries (there were so many, that I stopped after the total download crossed 100)
But according to Flurry (which is technically not only latest data, but also includes pirated copies whose users are online) reports:
which brings us to an interesting comparison. For the sake of simplicty, lets attribute the extra numbers to piracy for now.
Now here is another number. I have personally scourged the internet for pirated copies of the game and found hundreds of websites providing APKs for download, averaging about 200 per site. Similarly, I have chanced upon another couple websites offering IPAs for download, the total not being more than 500 downloads when last checked.
It is an impossible task to know exactly how many copies were pirated (we were hoping Flurry would help), but looking at the widespread distribution of APKs and the lack of relevant numbers in Flurry, I am inclined to assume a safe number of at least 20,000 pirated copies of the game.
None of this makes any sense for now.
Now, we have about 13,242 users according to Flurry. These users have played a total of 21,713 sessions (averages to about 1.64 sessions per user). Here is how frequently the game has been played this week.
Sessions refer to a complete app cycle – launch to termination, it does not imply the number of rounds played in the game. Here is how number of rounds fare:
The numbers, esp the revenue numbers, are not looking any hopeful or bright at all. The only concession that we allow ourselves right now is that this is the first week, and the game will be on store for a longer time. An immediate concern is lack of visibility, since we are not even mentioned on any of the major gaming media.
Its only been a week and we are not able to highlight any trend yet from our user behavior. We also need to optimize our data funnels to make more meaningful sense of data. For now, all our internal game data is a little nonsensical, with a bit of overlap, something that we will immediately fix with an update.
We will follow this up with a numbers post after finishing a month on store. Hopefully, by June 15th. As of now, we are completely clueless about what to make with this data.