构建区段定义 build-segments

流式客户细分

当新数据流入系统时，受众的用户档案列表会实时保持最新。

流式分段是一个持续的数据选择过程，会根据用户活动更新受众。 构建区段定义并保存生成的受众后，该区段定义将应用于传入 Journey Optimizer 的数据。 这意味着当个人资料数据发生变化时，将会在受众中添加或删除个人，从而确保您的目标受众始终相关。 请参阅Adobe Experience Platform文档以了解详情。

note important
IMPORTANT
评估方法和事件数据功能是独立的概念。 在以批处理方式处理特定事件类型时，可以将区段配置为流式传输。 自​2024年11月1日起，任何客户沙盒或组织中的Journey Optimizer都不支持跟踪和反馈数据集中的真实流式分段发送和打开事件。 这些事件仍可以在批处理区段中使用，但如果包含在流式处理区段中，则会以批处理方式评估这些事件，并且无法通过流式处理资格提供确定性的当天排除。 当区段定义符合Adobe Experience Platform的流式分段资格标准时，点击和其他跟踪事件可用于流式分段。 跟踪收藏集本身不会停止。 发送事件产生的相关排除、退回和延迟事件也会受到影响。 对于禁止使用案例，请使用本机频率上限或其他业务规则（如果可用）。 对于在同一历程中较早发送的支持消息的响应，请使用反应事件，而不是将逻辑重建为受众。

批次分段

每24小时评估一次受众的用户档案列表。

批量分段通过区段定义一次处理所有用户档案数据，创建可保存和导出以供使用的受众快照。 与流式分段不同，批量分段不会连续实时更新受众列表。 在批处理之后输入的新数据直到下一次批处理才会反映在受众中。 尝试强制立即更新不会覆盖每日周期。 要立即进行增量更新，请考虑使用流式或按需分段选项。

有关更多详细信息，请参阅Adobe Experience Platform分段服务文档

边缘分段

Edge分段功能可在Adobe Experience Platform的边缘🔗上即时评估中的区段，启用同页和下一页个性化用例。 当前只有选定的查询类型可使用边缘分段进行评估。 有关更多详细信息，请参阅Adobe Experience Platform分段服务文档

AI Knowledge Reference

This section contains structured knowledge intended to support interpretation, retrieval, and question answering related to this topic.

For complete understanding, this information should be combined with the documentation on this page. Neither source is intended to stand alone; the page describes the feature, while this section provides additional context that helps disambiguate terminology, intent, applicability, and constraints.

• TL;DR: This page explains how to build a segment definition from attributes and events to create an audience, choose its streaming, batch, or edge evaluation method, and evaluate audiences on demand for targeting in Journey Optimizer.

Intents:

• Create an audience with the Build rule method from the Audiences menu
• Use attributes and events (and components from existing audiences) as building blocks of a segment definition
• Choose the streaming, batch, or edge evaluation method for the audience
• Run on-demand (flexible) evaluation of selected audiences before targeting them
• Understand backfilling timing after an audience is first defined

Glossary:

• Segment definition: A new audience definition created with Adobe Experience Platform Segmentation Service using the Build rule method (product-specific)
• Build rule: The audience creation method selected from Create audience that opens the segment definition screen (product-specific)
• Attributes: One of the two basic building blocks of segment definitions, for example the city where profiles live or the birth year (product-specific)
• Events: One of the two basic building blocks of segment definitions, for example a purchase made within the last 7 days (product-specific)
• Streaming segmentation: Evaluation method where the audience profile list is kept up to date in real time as new data flows in (product-specific)
• Batch segmentation: Evaluation method where the audience profile list is evaluated every 24 hours as a snapshot (product-specific)
• Edge segmentation: Evaluation method that evaluates segments instantaneously on the edge for same-page and next-page personalization (product-specific)
• Flexible audience evaluation: The Audience Portal ability to run a segmentation job on demand for selected audiences (product-specific)

Guardrails:

• The frequencyMap attribute is not supported for use in segment definitions and cannot be used as part of audience segmentation criteria; for frequency-based targeting, consider frequency capping rules under business rules.
• As of November 1st, 2024, streaming segmentation no longer supports the use of send and open events from Journey Optimizer tracking and feedback datasets; this applies to all customer sandboxes and organizations.
• Send and open events can still be used in batch segments, but if included in a streaming segment they are evaluated in a batch manner; clicks and other tracking events remain available for streaming segmentation.
• Edge segmentation can currently evaluate only select query types.
• Batch segmentation is evaluated every 24 hours; attempts to force an immediate update do not override the daily cycle.
• After an audience is first defined, backfilling the audience from prior data can take up to 24 hours.
• Flexible audience evaluation lets you evaluate up to 20 audiences at a time, and ineligible audiences are automatically excluded.

Terminology:

• Canonical name: Segment definition — Acronym: n/a — variants: segment
• Do not confuse: “Streaming segmentation” (real-time updates) ≠ “Batch segmentation” (evaluated every 24 hours) ≠ “Edge segmentation” (instantaneous on the edge)

FAQ:

• Q: How do I start creating a segment definition? — From the Audiences menu, click Create audience and select Build rule.
• Q: Can I use the frequencyMap attribute in a segment definition? — No; it is not supported and cannot be used as segmentation criteria. Consider frequency capping rules under business rules instead.
• Q: Can I still use send and open events in streaming segmentation? — No; as of November 1st, 2024, streaming segmentation no longer supports send and open events from Journey Optimizer tracking and feedback datasets, though they remain usable in batch segments.
• Q: How long does backfilling take? — Backfilling the audience from prior data can take up to 24 hours after the audience is first defined.
• Q: How many audiences can I evaluate on demand at once? — You can evaluate up to 20 audiences at a time with flexible audience evaluation, and ineligible audiences are automatically excluded.