Do you guys give much credence to the retention figures that are given in iTunes Connect? My recently released iOS game seems to have pretty lousy retention %ages in iTunes Connect (D1 25%, D2 10%, D3 4%, etc.). However, I recently released a new version about a month out from initial launch and based on the number of updates that got downloaded, it looks like I have a 30%+ 1-month retention. Do you use update download numbers as a metric for gauging retention, or is it misleading to use such stats?
I haven't spent much time looking at the figures (but I'm going to try and allocate an hour a week to examine them), but I think the 1 day retention is very useful. If people aren't using the app on the first day it's installed then I think that's a problem. It might be that a promotion has been targeted at the wrong users, or it could be more serious in that the right group got it but didn't like it!
What about update download numbers? My update downloads at one month out from launch are high, which suggests that the iTunes retention stats are not very accurate at all.
Update download numbers are tricky. An update just means the user has downloaded the update, but it might not even be on purpose - they may have just clicked "update all". It's good that they haven't deleted the app, I suppose. I'd put much more weight on retention and session stats than update download numbers because of this. I've seen iPhones with hundreds of apps, many unused, the user just hasn't got around to deleting everything they don't want. Many users have a purge once they start to run out of space. Also, what about users that have a library of apps in iTunes that get updated, but actually never get installed to a device? Users will download everything they can, and keep it "just in case" they need/want it another day. I have many apps that I've never used, but haven't deleted.
The App Store Analytics seem broadly accurate, though as compared to the numbers we get from our core Amazon analytics solution they are slightly overstated, possibly because the retention data only tracks users who have opted in. They seem to be at least internally consistent though, so if you are using it to track your retention over time it is as good as any other provider. 30% D1 retention or 30% D28 retention? If the latter, that seems wrong unless you have very low user numbers and a few of them loaded it up to see the update. If the former, then that's a plausible jump from 25% based on an update. No. You shouldn't really do that, especially when you have the retention reports right there. Anecdotally, I update apps I don't use and haven't even ever loaded all the time, and other users do that too.
You are lucky getting actual downloads. My new word game "Headwords Up" has like zero downloads. You would think parents would want to teach their kids and even themselves bits if English in a fun way, but no. I should have spent my hard earned job salary hiring programmers and artist to do games like Flappy Bird, Color Switch etc. On that note I will be glad when software is so advanced I dont have to hire expensive programmers and can do it myself.
"You are lucky getting actual downloads." We've all been there, it's not fun to check the download stats and see a big fat zero. I don't want to derail this thread, but you need to allocate time to market your app. Give out promo codes, send them to relevant sites, forums and people that can help to spread the message. Marketing is hard (really hard), but with over a million apps on the app store it needs to be done. I'll admit to being pretty rubbish at it, but I'm learning! The icon for "Headwords Up" is a bit "busy", and I don't think the screenshots really show-off the app, so maybe think about improving them first? Download/purchase decisions are made in a split-second, and if the icon isn't enticing then they won't even click through to the app's page.
