Mobile game operational Data analysis metrics

The content of this article comes from TalkingData's mobile game data Analysis Whitepaper, just a simple transcription.

I. User acquisition (ACQUISTION)
Number of new users (Daily new USERS,DNU): Number of users who register and log in to the game daily.
Solve the problem:
* New user share of channel contribution;
* Macro trend, whether the need for delivery;
* Whether there is a channel cheating behavior.
Note:
* The number of Zhou Xi users is the sum of the number of new users in the 7 days of the week;
* The number of new users calculated as above;
* According to the needs, can be subdivided into natural growth users (non-promotional period) and promotional users (promotion period).

Number of Session users per day (Daily one session Users,dosu): The user of a conversation, that is, only one session in the newly logged user, and the session duration is below the specified threshold value.
Solve the problem:
* Whether the promotion channel has the brush quantity cheating behavior;
* Channel promotion quality is qualified;
* User Import whether there is a barrier point, such as: Network status, load time and so on.
Note:
* The number of sessions per week is the sum of the total number of session users in the 7 days of the week;
* Monthly session User count is calculated as above;
* One of the game guide design analysis points;
*dosu helps to assess the quality of new users, and further analysis requires defining the number of monthly session users for active users.

User Acquisition Cost (Customer acquisition COST,CAC) = Promotion Cost/valid new login user
Solve the problem:
* What is the cost of obtaining a valid new subscriber?
* How to choose the right channel to optimize the delivery;
* How much is the channel promotion cost?
Note:
*CAC calculations are broken down according to the channel.

Second, the user active (Activation)
Daily active users (Daily active Users,dau): Number of users who log in to the game daily
Solve the problem:
* What is the core user size of the game?
* Game Product cycle change trend measurement;
* Game products old user churn and active situation;
* Channel active user life cycle;
* How sticky the game product is (combined with Mau).
Note:
Dau the size of the core user needs to be cautious about the changes in the Dau of new users and returning users, in order to understand the size and quality of users based on detailed dau segmentation.

Weekly Active users (Weekly active Users,wau): The number of users who have logged in to the game in the last week (including 7 days of the day), usually calculated according to the natural week.
Solve the problem:
* Game Cycle user size is how much;
* Game product periodicity (weekly) Change trend measurement.
Note:
Wau to analyze the user scale according to the week as a cycle, it is advantageous to find the problem and master the fluctuation of the game user scale in the dimensions of different active users.

Monthly active Users (Monthly active Users,mau): The number of users who have logged in to the game in the last one months (including 30 days of the day), usually in natural months.
Solve the problem:
* What is the overall user size of the game?
* Game product user scale stability;
* Promotion effect assessment;
* How sticky the game product is (combined with dau).
Note:
*mau level of user size changes relatively small, can show the stability of the user scale, but the promotion of a certain period and version update on the impact of Mau may be more obvious;
* In addition to the game life cycle at different times, Mau changes and stability is also different.

Number of daily participation (Daily Engagement Count,dec): The use of mobile games is recorded as one participation, the number of day participation is the total number of users participating in the game daily.
Solve the problem:
* Measurement of User stickiness (daily average number of participants);
* What channel, what user participation frequency is high;
* How often the user participates in the product.
Note:
* General recommendation 30 seconds to repeat the record for a full use, not alone metering;
* Weekly participation is the total number of participants in the game for one week;
* The number of monthly participation is ibid.
* Average daily participation: the average number of per-user participation in the game on that day.
Calculation formula: Number of daily participation/daily participation of users;
* Through the analysis of the distribution of different participation times, can help to analyze the impact of version updates, promote channel stimulation.

Daily use Duration (Daily avg.online time,daot/at): Average Daily Online duration of active users. That is: daily total online hours/days active users. General Lean Calculation Company: At=acu*24/dau
Solve the problem:
* User's game participation degree;
* Product Quality Control index:
* Channel quality;
* Combined with single-use duration analysis of retention and loss issues;
* The user's ability to play continuously.
Note:
* Average single use time: within a certain period of time, the average number of times each game used by the user = The total number of times users use time/participation;
* Help to analyze cheating behavior, version stickiness and effects;
* Depending on your needs, you can observe the average duration of the user's weekly, biweekly, and monthly usage, and learn about the stickiness of the game.

User Activity level (DAU/MAU)
Solve the problem:
* User's game participation degree;
* The popularity of the game is growing, declining, stable;
* How many days the user is active.
Note:
Dau/mau theory is not less than 0.2,0.2*30=6 days, that is, the number of users logged in no less than 6 days.

Iii. Retention & Loss (Retention & Churn)

User retention (Users Retention): In the statistical time interval, the login usage of the new user at subsequent different periods.
Next day retention rate (Day 1 Retention Ratio): The number of users logged in on the next day (excluding the day of first sign-on) accounts for new users.
3rd Retention rate (Day 3 Retention Ratio): The number of users logged in on the third day (excluding the day of first sign-on) accounts for new users.
7th Retention rate (Day 7 Retention Ratio): The number of users logged on on the seventh day (excluding the day of first sign-on) accounts for new users.
Monthly retention rate (Day Retention Ratio): The number of users logged on on the 30th day (excluding the day of first sign-on) accounts for new users.
Retention rates require long-term tracking and can be set 30th, 60 days, or 90 days, as required.
Solve the problem:
* User's adaptability to the game;
* Assess the quality of channel users;
* Delivery channel effect assessment;
* How sticky the user is to the game;
* New users when the loss of time will intensify.
Note:
* The retention rate in some sense represents the new users of the game satisfaction;
* Pay attention to the retention rate at the same time need to focus on user churn node;
* The statistics and calculations of retention rates can also be analysed in terms of natural and natural months, such as the retention of new users in the following weeks in the last week;
* The next day retention rate represents game satisfaction, mainly reflecting the beginner's adaptability to game guidance and gameplay in the early stages of the game.

user churn (users churn): In the statistical time interval, the user leaves the game at different times.
daily churn rate (Day 1 churn Ratio): Statistics Day login game, but then 7th shows users logged in to the game as the percentage of active users of the statistics day , this definition extends the observation length as required, see Remarks;
weekly churn rate (Day 7 churn Ratio): Last week, you logged in to the game, But the percentage of users who did not log in to the game this week accounted for the weekly active users of the week;
monthly churn rate (Day churn Ratio): Logged into the game last month , but the percentage of users who logged in to the game this month accounted for active users in the month.
Fix the problem:
* What is the life cycle of the active user;
* which channel has a high turnover rate;
* The way to pull revenue, version update for users of the impact of the loss of how much;
* When the rate of attrition is higher.
Note:
* loss rate + retention rate is not equal to 100%, the retention rate is subject to the criteria defined above;
* Daily churn rate can be defined according to the requirements, such as statistics on the day of the game, but then 14th or 30th the number of users not logged in the game;
* Wastage rate in the game into a stable period is worth attention, the stability period of active and income are more ideal, if the loss rate fluctuations, it is necessary to arouse vigilance. Need to pay close attention to which part of the user left the game, wastage rate as a vane, with early warning effect.

Four, game revenue (Revenue)
There are currently three forms of mobile game revenue generation:
* paid Download
* in-app ads
* in-app pay
here to reconsider the third situation for the indicator definition, the following description does not separately describe the recharge and consumption, only to pay collectively.
monthly payment rate (Monthly Payment ratio,mpr): The percentage of active users for paid users within the statistical time interval. Generally in months. Calculation company: Mpr=apa/mau where APA is the number of paid users for the month (see below)
Fix the problem:
* The payment guide for the game product is reasonable;
* User pay preference and willingness (combined with first-time pay features, props, ratings, Overall analysis);
* Pay conversions to achieve the desired results.
Note:
*mpr includes users who have paid for the payment in the statistical timeframe by historical paying users and paid users who are newly converted within the statistical timeframe, and
*mpr does not necessarily mean an increase or decrease in the number of paid users of the game;
* different game types, The corresponding MPR performance is also different.

active paid users (active Payment Account,apa): Number of users who have paid successfully during the statistical time interval. Generally in months. If you calculate on a monthly basis, you have the following relationship: Apa=mau*mpr where MAU is the number of active users per month, MPR is the monthly payment rate.
Fix the problem:
* What is the size of a paid user for a game product;
*apa? The percentage of whale users, dolphin users, and small fish users;
* Overall stability of paid users.
Note:
*apa includes users who have paid for the payment in the statistical timeframe by historical paying users and those who have been converted to pay within the statistical timeframe, and
*apa can be subdivided to recharge active users and consume active users as needed.
Average per user revenue (Average Revenue per UERS,ARPU): The average revenue generated by active users for the game during the statistical time interval. Generally in months.
arpu= earnings/players
Month arpu= earnings/mau
Calculation: Total game revenue divided by the number of overall active users of the game, generally calculated by month, that is, arpu= monthly total revenue/monthly active users (MAU)
Solve the problem:
* How the quality of the user is obtained from different channels;
* how the game proceeds;
* The relationship between active users and the per capita contribution;
* How to increase the level of income of game players.
Note:
* strictly defined ARPU is different from the domestic understanding of ARPU, domestic arpu= total income/paid users;
*arpu for the initial product positioning at different scales of revenue estimates.

Average revenue per paid user (Average Revenue per Paying User,arppu): The average abdominal mass of paid users to the game during the statistical time interval. Generally in months.
Arppu= Revenue/Paid users
Monthly arppu= Income/apa
Solve the problem:
* What is the average pay level for game-paying users?
* What is the overall pay trend for paying users?
* Analysis of whale users.
Note:
*arppu vulnerable to whale users, small fish users, the analysis should be cautious;
*arppu and APA, MPR and so on can be used for the retention of paid users, the loss of specific paid groups to conduct in-depth analysis to ensure payment quality and scale.

Lifecycle Value (Life time VALUE,LTV)
Lifecycle (Life time): The average number of times a user has averaged from the first participation to the last participating game.
Life cycle Value: The total amount of revenue the user creates for the game during the life cycle. Can be seen as a long-term cumulative ARPU value.
Calculation method: The average LTV per user is calculated as follows:
Ltv=arpu*lt (average life cycle by month)
Where LT is life time, that is, the lifetime, according to the monthly statistics, that is, the average number of months the player retained in the game.
For example, a game of arpu=2 Yuan, lt=5, then ltv=2*5=10 Yuan.
Solve the problem:
* How long the user will stay in the game;
* What is the value of the user's contribution to the game;
* User base and channel profit contribution (LTV>CAC).
Note:
*ARPU follows strictly defined terms, that is, total revenue/number of active users;
*LTV is calculated for active users, and there are no paid and non-paid users.
The following indicators are only a representative part of the mobile game indicators, in the actual analysis process, according to the analysis of dimensions, can be carried out in-depth indicators, such as the income analysis can be added to the return of the user contribution, continuous pay user contributions, paid to retain users, paid user churn rate, two pay analysis, user Pay Cycle conversion and so on.
In addition, some of the common indicators are not described in detail, here are only a few notes:
PCU (peak Concurrent Users): Maximum simultaneous number of players online
ACU (Average Concurrent Users): Average number of simultaneous online players
new users Converstion rate: conversion rates (can be divided by channel) Clicks->install->register->login

K-factor:k factor
K-factor= Rate of infection * conversion
Conversion rate: The rate at which the infection is converted to a new user.
Infection rate: The number of invitations sent per user is generally averaged.
If the k>1, the game user group through self-propagation growth faster;
If the k<1, the game user group to reach a certain size will stop through self-propagating growth.

Mobile game operational Data analysis metrics