User Cohorts

User Cohort Analytics groups users by their unique properties, and uses the groupings to study the differences between each user groups’ main indices and their game play patterns in accordance with the time.

Classify users into cohort groups for more focused and narrowed analysis between customized groups of users. For instance, as a publisher you can assign Non-Paying users with no in-app purchase experience and VIP users with over $100 monthly in-app purchases as user cohort groups to assess the differences in each of these two groups’ DAU, number of sessions, average session time of the users, and other major indices.

1. Types of User Cohorts

Tapjoy provides default cohorts and custom user cohorts. Default cohorts are the basic set of cohorts that are provided after users SDK integration with Tapjoy. Custom user cohorts are the ones publishers can customize freely to their needs.

Default Cohort

Basic user cohorts generally used in data analysis. Default Cohorts are provided as soon as you complete your SDK integration with Tapjoy. There are 7 default user cohorts: OS, Country, Referrer, Store, App Version, Device Type, and Paying/Non-Paying

OS Cohort

• OS cohort is provided when your Tapjoy App ID/App Key is sent from your iOS/Android apps. It does not include specific store information, however.

Country Cohort

• Your country cohort is collected from information gathered from either the SIM card or IP address of the user’s device.
• You can select multiple countries when you select country cohorts. For example, you can either view the information regarding total sales in US alone, or view the aggregated Sales information in US/Germany/China all together at once.

Referrer Cohort

• The cohort is mainly used to analyze the marketing ROI and app usage behaviors of the users from different referrers. (Ad, cross-promotion, etc.)
• “Referral” cohort can trace basic referrer information of games that are published on Google Play Store through the Tapjoy SDK only.
• Please refer to the SDK integration guide to learn how to set up Referral Cohort.

Store Cohort (only applicable to Android)

• Store Cohort is provided when you register an app as a single entity without distinguishing it by market/store information and sends the same Tapjoy App ID/App Key from the app.
• If you would like to designate appstore cohort, a modification of AndroidManifest.xml file is required. Please refer to the SDK integration guide for Android for more information.

App Version Cohort

• Tapjoy SDK automatically acquires version details from apps that are published on platforms (Facebook, etc.) or stores (App Store, Google Play, etc.). Tapjoy.setAppDataVersion is not related to this cohort.

Device Type Cohort

• Based on the screen resolution and DPI data retrieved from the Tapjoy SDK; devices that are 7 inches and above are considered a ‘Tablet’, otherwise they will be labeled as a ‘Phone’.

Paying/Non-Paying

• Based on IAP purchase tracking enabled through the Tapjoy SDK.

Default Cohort variables under the same category can be used in combination with each other. For example, if you have your country cohort selected as “USA", and you would like to select multiples of game version cohorts at the same time, you can simply select in all game versions you wanted to analyze. Select game version three game version cohorts “1.0”, “1.5”, “2.0”, then you will be able to only index users from USA using the game versions from 1.0 to 2.0. There are limitations on this however; you may only choose up to 5 default cohorts from the 7 variables at a time to compare.

Custom User Cohort

Tapjoy also provides customizable user cohort variables.

Level Group Cohort (pre-defined)

• Level Group Cohort distinguishes users into groups by their levels for indexing purposes. For example, you can create groups as such: Low Level (1~10), Medium Level (11~20), and High Level (21~30). Creation and modification of Level groups can be done by selecting App Settings under the Settings menu button, then selecting analytics.
• In order to use the level group cohort, you must integrate Level Data when performing SDK integration. Tapjoy.setUserLevel(1);

Purchase Group Cohort (pre-defined)

• Purchase Group Cohort puts users into groups by their cumulative purchase amount for indexing purposes. For example, Publishers can make comparisons on average session time between Non-Paying Users and VIP Guests.
• Creation and modification of Purchase Groups can be done by selecting App Settings under the Settings menu button, then selecting analytics.
• You must integrate Purchase Event in order to use the purchase group cohort. For more details, please refer to In-App Purchase Integration page. Tapjoy.trackPurchase("RUBY_SINGLE_PACK", "USD", 0.99, null);

User Cohort 1~5

• We provide up to 5 cohort variables which you can use in addition to the pre-defined cohorts. You can refer to the following for the data integration of User Cohort Variable. Tapjoy.setUserCohortVariable(1, "DemoUser");

Similar to default cohorts, there are limits on the number of cohorts you can use at once for analysis. For custom cohorts however, the number of cohorts you may use simultaneously is 3.

2. Designing User Cohorts

Below are some examples of Custom User Cohort Variables that are generally useful in typical app. You can select up to five of these to apply in your analysis.

• Log-In Status
• Character
• Total Play Time
• Purchase Amount
• Character/Skin/Core Item purchasing status
• Player Ranking
• Friends Count

In the chart below, cohorts that are labelled with (built in) in their name are the basic cohorts provided by Tapjoy; built in cohorts will not be counted in the user cohort variable limit. Publishers can use 7 basic cohorts (Built In) and additional 5 User Cohort variables in your analysis.

Example of a User Cohort

Cohort Type	Cohort Name	Explanation	Usage and Analysis Hypothesis
Custom User Cohort	User Level Group (Built In)	Distinct user levels into various groups for analysis. (ex, Low Level (1~10), Middle Level (11~20), High Level (21~30))	- Creation and modification of Level groups can be done in the “Game/App” menu. - In order to use the level group cohort, you must integrate Level Data when performing SDK integration. (As Level Data is sent to the Tapjoy server, Level Group cohort settings screen will automatically appear on the dashboard.)
	Character	Play pattern analysis and core indexes for each character type.	- Used for assessing BM’s applicability, and also used as the basis to identify improvements.
	Total Play Time	Core indexes and play pattern analysis by differing playtimes.	- Client has the cumulative play time data of the users calculated, if not, the client needs to access DB to acquire cumulative play time data.
	Purchase	Distinguish users based on their in app purchase amount.	- Cumulative purchases since installment
	Character/Skin Purchase status	Indices and play patter analysis of the users who purchased characters and skins (Major BM)
	Play Ranking	Analyze the effects of in-game ranking on the users’ play patterns or purchase patterns.
	Friends Count (Built In)	Analysis by Number of Friends saved or number of team-play participants.	Friends count is provided by SDK in a form of pre-defined.
Default Cohort (Built In)	Country	Trace Client Device’s IP in order to analyze player’s location (Country) at the playing moment.
	Referrer	Analyze marketing activities such as advertisement, Cross promotion, etc.
	Store (Android Only)	Analysis of what marketplace the App came from.
	App Version	Analyze the effects from the updates of each game versions.
	Device Type	Analyze play patterns of the users by each device type
	OS	Per OS analysis, in case of managing multiple OS grouped in as a single app.
	Paying/Non-paying	Analysis of IAP behavior

User Cohort Design Template

Cohort Type	Cohort Name	Explanation	Usage and Analysis Hypothesis
Custom User Cohort	User Level Group (Built in)	Use( )/Not Use( )	Segment user levels into several group for analysis. (ex, Low Level (1~10), Middle Level (11~20), High Level (21~30))
	Custom User Cohort (1)	Use( )/Not Use( )
	Custom User Cohort (2)	Use( )/Not Use( )
	Custom User Cohort (3)	Use( )/Not Use( )
	Custom User Cohort (4)	Use( )/Not Use( )
	Custom User Cohort (5)	Use( )/Not Use( )
Default Cohort (Built In)	Country	Trace Client Device’s IP in order to analyze player’s location (Country) at the playing moment.
	Referrer	Analyze marketing activities such as advertisement, Cross promotion, etc.
	Store (Android Only)	Analysis of what marketplace the App came from.
	App Version	Analyze the effects from the updates of each game versions.
	Device Type	Analyze play patterns of the users by each device type
	OS	Per OS analysis, in case of managing multiple OS grouped in as a single app.
	Paying/Non-paying	Analysis of IAP behavior

• The above chart is made to support User Cohort design process, and the communication with developers