Event Design and Transmission Guide¶

What Is Event Taxonomy?¶

Event Taxonomy is a classification system that organizes collected events and attributes into a consistent structure with naming rules.

If the many actions that occur in a game, such as login, purchase, and leveling up, are collected without any rules, different teams may send the same data under different names, or it may become difficult to understand what each piece of data means during analysis. Event Taxonomy prevents this confusion and helps make collected data immediately usable for analysis.

Why Is It Needed?¶

Structure¶

Analytics event taxonomy consists of two layers: Event and Attribute.

Event
├── Platform Attribute   ← Automatically collected with SDK integration only
└── Event Attribute      ← Sent directly with the event

What Is an Event?¶

An event is a unit that records a specific action performed by a user in the game as data. It contains "when, who, and what happened" and becomes the most basic unit of analysis.

• Example: A user purchased an item -> item_purchase event occurs
• Example: A user cleared a specific stage -> stage_clear event occurs
• Example: A user logged into the app -> login event occurs

What Is an Attribute?¶

An attribute is detailed information recorded together when an event occurs. With only the event, you can know only "what happened", but attributes let you understand "under what conditions and with what values" it happened.

• Example: Attributes of the item_purchase event -> item ID, payment amount, currency unit
• Example: Attributes of the stage_clear event -> stage number, clear time, used character

Attributes are divided into two types depending on who collects them.

• Platform Attribute: Attributes automatically collected by Hive through SDK integration only, without additional configuration. Includes country, OS, app version, server ID, etc.
• Event Attribute: Attributes sent directly with the event. You define and send the items needed for analysis.

Sending Custom Events¶

Custom events defined directly in the game can be used in Analytics for analysis if they are sent according to the collection schema below.

Actual data transmission uses Hive SDK Client Log Transmission. The source code required for transmission can be automatically generated by language in the [Source Generation] tab of Event.

Before sending events, if you define the event name and attributes in Event, it is convenient for sharing and management within the team. However, even if you send data without prior definition, Analytics automatically recognizes the received data and reflects it in the event list within up to 1 hour.

Analytics Collection Schema¶

This is the default schema used when sending custom events. All events must be sent in the format below.

Event Name Naming Rules¶

It is recommended that event names used in eventName follow the rules below.

• Recommended characters: English letters (a-z, A-Z, case-sensitive), numbers (0-9), underscore (_), hyphen (-)
• Separator: Separate words with underscore (_) rather than spaces
• Form: A noun_verb structure that clearly describes the event is recommended
• Examples: item_purchase, level_up, tutorial_complete, stage_start

Example Structure of hiveAttribute / eventAttribute¶

hiveAttribute and eventAttribute are each sent as JSON objects.

hiveAttribute - Platform attributes provided by the Hive provisioning server. Market, server ID, and similar values belong here. Currently, developers must set and send the values directly. This will be changed to an automatic collection method in the future.

{
  "market": "GO",
  "serverId": "Asia"
}

eventAttribute - Attributes sent directly with the event. These contain detailed information needed to analyze the event.

{
  "level": "100",
  "contentsType": "dungeon"
}

Attribute Naming Rules¶

It is recommended that the key names of eventAttribute and hiveAttribute follow the rules below.

• Recommended characters: English letters (a-z, A-Z, case-sensitive), numbers (0-9), underscore (_), hyphen (-)
• Separator: Separate words with underscore (_) rather than spaces
• Examples: item_id, contents-type, stage_level

Attribute Key Conversion Rules¶

When attributes are sent, the system automatically converts some key names according to the rules below. Be careful, because the actual key names queried during analysis may differ.

-Hive Suffix (hiveAttribute)¶

-Hive is automatically appended to the end of all keys in hiveAttribute.

-Event Suffix (eventAttribute)¶

-Event is automatically appended only when a key in eventAttribute overlaps with one of the following 8 common fields:

• timezone, dateTime, identifierProvider, userId, deviceId, appId, category, eventName

Recommended Events¶

Below are examples of events that are highly useful in Analytics. If you send each event according to eventName and eventAttribute, you can analyze them immediately using all Analytics features such as chart, funnel, and retention.

Level Up¶

eventAttribute (common):

Currency¶

eventAttribute (common):

Content¶

eventAttribute (common):

In-App Store¶

eventAttribute (common):

Mate¶

eventAttribute (common):

Social Activities¶

Related Menu¶

• Event - Register and manage events and attributes
• Metric - Save frequently used event aggregation settings as metrics