Quick Audience V4 basic concepts: workspace, data source, user ID, tag, audience - Quick Audience

Learn the key terms used in Quick Audience V4, including workspaces, data sources, user IDs, tags, audiences, and marketing models.

Term	Description
Workspace	Also called Workspace. Quick Audience supports multiple workspaces with isolated data, typically representing different brands or departments. Users must join a workspace as a Workspace Member before performing brand- or department-level operations.
Computing engine	A Computing Source is the database that stores underlying data tables for analysis objects. It performs ID Mapping to enable user identification and omni-channel data integration. The same MaxCompute database can serve as the computing source for multiple workspaces.
Analysis source	After ID mapping in the computing source, raw data is imported to the Analysis Source for analysis and marketing. Supported databases include Hologres and AnalyticDB for MySQL 3.0.
Data Sources	Data source is a collective term for Computing Source and Analysis Source.
User ID	Also called ID. A user ID indicates a user's identity. Imported user tables must contain at least one ID field. For more information about the ID field, see Quick Audience Import Data Table Requirements. IDs can be used for marketing, push, and other operations. User IDs fall into four categories: User ID: represents a user. Examples include mobile phone numbers and email addresses used to register on third-party platforms or enterprise accounts. Device ID: identifies an electronic device. Typically collected through app tracking (IMEI, IDFA, IMSI, OAID, MAC address). Not bound to a specific user. Enterprise ID: an ID generated by an enterprise business system, such as a CRM membership ID. Third-party platform ID: identifies a user on a third-party platform. Examples include UnionID, OpenID, Taobao ID, Taobao nickname, Alipay ID, and Weibo ID. After a user ID is added to Quick Audience, ID Mapping generates a unique identifier (QAID), enabling cross-channel and cross-ID-type user identification and data retrieval.
QAID	A QAID is a unique ID that Quick Audience assigns to each user. When data tables are imported or event data is received, the system performs ID Mapping to identify and deduplicate users, then assigns a QAID to each user. In user analysis, audience filtering, and marketing delivery, the QAID serves as the unique user identifier. All ID fields and data are matched through the QAID.
Tag	Tags fall into these categories: Imported tags: The User Tags records dimensional characteristics of users. When importing data to Quick Audience, you can configure tag aliases for fields to label users. Tags reference the corresponding underlying fields. Custom tags: Define tag rules based on imported data to label users who meet specific conditions. You can customize the following tags: Preference Tags: Derives a tag value from the User Behavior Table Data Requirements and Order Details Data Requirements based on the most frequent or highest-count dimension feature. Examples: preferred shopping time window, price range, channel, theme, category, or highest single-item purchase price. Loyalty Tags: Derives a tag value from the User Behavior Table Data Requirements and Details Data Requirements based on the timing or frequency of user behavior. Examples: last purchase or access time, days since last activity, or cumulative consumption/active days. Purchasing Power Tags: Derives a tag value from the User Behavior Table Data Requirements and Details Data Requirements. Examples: cumulative purchase amount, number of purchases, number of orders, or average/maximum/minimum order amount in the last year. User Stage Tags: Filters users by hierarchy logic, similar to Audience Filtering. Users meeting different conditions receive different hierarchical labels. Example: users with total purchases ≥ 1000 who prefer food get marketing priority 1; users with purchases between 100–999 in group A get priority 2. The Tag List also includes User Attributes.
User attribute	User Attributes define a user's basic information. Values come from imported user tag tables. When a table field is mapped to a user attribute, only the mapped attribute is displayed; the original tag is hidden. Combined with ID Mapping, user attributes help build a cross-channel user information system that forms the basis for user profiles.
User profile	A user profile is a tagged model built from user attributes, tags, preferences, behavior records, purchase records, and marketing records. The User 360 page displays consolidated information for users identified through ID Mapping, with labeling functions for lead analysis and after-sales follow-up.
RFM model	The RFM Model measures user value using three metrics: Recency (R), Frequency (F), and Monetary (M). Quick Audience provides an RFM model based on Order Details Data Requirements and Order Summary Data Requirements for RFM analysis and audience filtering.
AIPL model	The AIPL Model divides user-brand interactions into intimacy stages to measure user value: Awareness (A), Interest (I), Purchase (P), and Loyalty (L). Users may transition between stages as their behavior changes over time. The Quick Audience AIPL model is built on the Quick Audience Import Data Table Requirements and supports AIPL user analysis, AIPL flow analysis, and audience filtering.
People	An Audience in Quick Audience is a collection of QAIDs representing multiple users. Unlike the full user set, an audience is a filtered subset created for a specific purpose. Creation methods include: audience filtering from imported data tables, uploading user ID lists, computing intersections/unions/differences of existing audiences, and extracting users from analysis results or marketing campaigns (SMS, email, push). Audiences support insight analysis, marketing delivery, and pushing to Data Bank, Damengpan, or Kafka.
Push	Push data to external channels for storage and use. Audience Push: Push data to Brand Data Bank, Damengpan, Alibaba Cloud Kafka, or open-source Kafka. You can also push data to social interaction, retail CRM, and media delivery modules. AIPL Model Push: Push the model to Brand Data Bank.
Automated Marketing	Automated Marketing lets you design differentiated marketing strategies using a drag-and-drop canvas, which the system then executes automatically.
Behavioral Event	Also called Event. Defines specific user actions in a channel as events. Channels report real-time event data in a specified format for automated marketing. Order events are a subtype focused on purchase-related behavior. Event data can be stored in the Analysis Source. Behavior events are saved as user behavior tables, and order events as order details tables. These tables support audience filtering, RFM/AIPL model generation, and custom tag creation.