My game publishing experience on Google Play (6) -- First Monthly Revenue Report

    Finally we obtained the access to our ad-revenue back-end. We are notified that from now on, in the beginning of each month, we'll receive the revenue report of the preceding month. The earnings would be actually paid in the next month.

    So here's the table of our earnings:

Day#	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
US$	0.1	4.5	4.4	7.1	16	4.9	1.3	2.9	1	0.5	0.6	0.7	1	0.8	1.3

    Furthermore, I guess it's a good timing to investigate the relationship between downloads and earnings:

Figure 1: Earning (US$) vs Downloads (times)

Definitions:
Daily Installs- Number of devices that had downloaded our app today.

Current Installs- Number of devices that currently had our app installed.

    Obviously, neither daily installs nor current installs is a good enough estimator for the revenue. Thus I made up a model assuming only α: [1,0] players who installed on day n would contribute to the revenue on day n+1.

    Thus we could model the estimator to revenue with the Exponential Weighted Moving Average of the daily download. The error sum is minimized when α = .45 (Sum =7.3), obviously better than the other two estimators above (Sum=16.8 for Daily and Sum =22 for Current)

    What does the analysis above indicates? It reveals that nearly half of our user lost their interest to our game after a day! We really should take some efforts on the retention rate.

This is a series sharing my publishing and cooperating experience on Google Play game.
 Last Post: App's ranking in Google Play
Next Post: Another peak!