On Wed, Sep 23, 2015 at 4:21 PM, Joshua Minor <[email protected]> wrote:
> Couple comments: on Jon's comments: > > - Two app-level metrics I would like to see moving forward would be a > number showing a daily/monthly visit count divided by the install base (if > it is achievable): 15% of Android users visit every day, 45% visit every > month. Along those lines, I am very interested to see the retention data > that Android is now collecting. > > Apple now provides App Retention info on iTunes Connect Analytics page, so > we can do this for both platforms. Let me know if I can help show you where > that live in our Apple account. > As an experiment, I added 1-day retention to today's report after discussion with Josh. Still need to look into the Android data. > > - Ultimately, I think 30 day views (or 28 day views) are less subject > to noise than 7-day. > > Some kind of moving average may be helpful to reduce the noise. Maybe as a > trend line laid on top of the raw data? > Yes, trendlines or generally a better handle on seasonality would be good. For some metrics like the app uniques where we don't yet have a full year of data, this might be trickier. -- Tilman Bayer Senior Analyst Wikimedia Foundation IRC (Freenode): HaeB
_______________________________________________ Mobile-l mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/mobile-l
