合并 combine

在此页面上:​了解如何在编排的活动中使用合并定位活动,通过并集、交叉或排除操作细分入站群体。

合并​活动是一种​ 目标选择 ​活动,可以让您有效地对入站群体进行细分。 利用该功能,可合并多个群体、排除特定区段或仅保留在多个目标之间共享的数据。

可以使用以下分段选项:

  • 并集:将多个活动的结果合并为单个统一目标。

  • 交叉:仅保留所有入站群体中共同的元素。

  • 排除:根据指定的条件,从一个群体中移除元素。

配置合并活动 combine-configuration

请按照以下常见步骤操作,开始配置​ 合并 ​活动:

  1. 添加多项活动(例如​ 生成受众 ​活动),来构成至少两个不同的执行分支。
  2. 向先前的任何分支添加一个​ 合并 ​活动。
  3. 选择分段类型:并集、交集或差集。
  4. 单击​继续。
  5. 在​ 要加入的集合 ​部分中,选中您之前希望加入的所有活动。

并集 combine-union

在​ 合并 ​活动中,您可以通过选择​ 协调类型 ​来配置​并集,从而决定管理重复记录的方式:

  • 仅键(默认):当多个入站过渡共享同一键时,将保留单个记录。 仅当入站群体是同质的时,此选项才适用。

  • 列的选择:可以让您指定用于数据协调的列。 选择​添加属性。

在下面的示例中,同时使用​ 合并 ​活动与​并集,以将两个查询的结果(忠诚度会员​和​购买者)合并到一个更大的受众中,包含两个区段中的轮廓。

交集 combine-intersection

在​ 合并 ​活动中,您可以配置​交叉。 为此,您需要执行以下额外步骤:

  1. 选择​协调类型,定义如何处理重复项。

    • 仅键(默认):当多个入站过渡共享同一键时,将保留单个记录。 仅当入站群体是同质的时,此选项才适用。

    • 列的选择:可以让您指定用于数据协调的列。 选择​添加属性。

  2. 如果要处理剩余的群体,请启用​生成补集。 补集包含所有入站活动结果的并集,但不包括交集。 活动中会添加一个额外的出站过渡。

下方示例展示了两个查询活动之间的​ 交叉 ​的使用。 这用于标识属于​ 忠诚会员 ​并在上个月内购买过产品的轮廓。

差集 combine-exclusion

在​ 合并 ​活动中,您可以配置​排除。 为此,您需要执行以下额外步骤:

  1. 在​ 要加入的集 ​部分中,选择代表主要群体的​主集。 在其他集中找到的记录将从此主集中排除。

  2. 如果需要,您可以调整入站表以对齐不同维度的目标。 要在另一个维度中排除某个目标,必须首先将其带回到与主要群体相同的目标维度。 为此,请单击​ 添加规则 ​并定义更改维度的条件。 然后,使用属性或联接进行协调。

  3. 如果要处理剩余的群体,请启用​生成补集。 补集包含所有入站活动结果的并集,但不包括交集。 一个额外的出站过渡将添加到活动中。

以下​ 排除 ​示例显示了两个查询,它们配置为筛选购买产品的轮廓。 然后,从第一组集中排除不是忠诚会员的轮廓。

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 configure the Combine activity in an Orchestrated campaign to segment an inbound population through Union, Intersection, or Exclusion operations.

Intents:

  • Add a Combine activity to join at least two execution branches in an Orchestrated campaign.
  • Merge multiple populations into a single target using Union.
  • Keep only the elements common to all inbound populations using Intersection.
  • Remove elements from one population based on specified criteria using Exclusion.
  • Choose a reconciliation type (Keys only or A selection of columns) to handle duplicates in a Union or Intersection.
  • Process the remaining population after an Intersection or Exclusion by enabling Generate completement.

Glossary:

  • Combine: a type of Targeting activity that segments an inbound population by merging multiple populations, excluding specific segments, or retaining only the data shared across several targets (product-specific)
  • Union: a Combine segmentation type that merges the results of multiple activities into a single unified target (product-specific)
  • Intersection: a Combine segmentation type that retains only the elements common across all inbound populations (product-specific)
  • Exclusion: a Combine segmentation type that removes elements from one population based on specified criteria (product-specific)
  • Reconciliation type: the setting that defines how duplicate records are handled in a Union or Intersection, either Keys only or A selection of columns (product-specific)
  • Keys only: the default reconciliation type; retains a single record when multiple inbound transitions share the same key, applicable only when the inbound populations are homogeneous (product-specific)
  • A selection of columns: a reconciliation type that allows specifying which columns are used for data reconciliation, via Add attribute (product-specific)
  • Generate completement: an option in Intersection or Exclusion that processes the remaining population; the complement contains the union of all inbound activity results excluding the intersection, and adds an additional outbound transition to the activity (product-specific)
  • Primary set: in an Exclusion, the main population selected in the Sets to join section, from which records found in the other sets are excluded (product-specific)
  • Exclusion rules: in an Exclusion, used when a target from another dimension must first be brought into the same targeting dimension as the main population; added via Add a rule, defining the dimension change conditions, with reconciliation done via an attribute or a join (product-specific)
  • Sets to join: the section where the previous activities to combine are checked (product-specific)

Guardrails:

  • Keys only reconciliation is only applicable when the inbound populations are homogeneous.

Terminology:

  • Canonical name: Combine — variants: Combine activity
  • Do not confuse: “Union” (merges results of multiple activities into one target) ≠ “Intersection” (keeps only elements common to all inbound populations) ≠ “Exclusion” (removes elements from one population based on specified criteria)
  • Do not confuse: “Keys only” (retains one record per shared key, homogeneous populations only) ≠ “A selection of columns” (reconciliation on specified columns)

FAQ:

  • Q: What does the Combine activity do? — It segments an inbound population by merging multiple populations, excluding specific segments, or retaining only the data shared across several targets.
  • Q: What are the three segmentation types in the Combine activity? — Union, Intersection, and Exclusion.
  • Q: What is the default reconciliation type for a Union or Intersection? — Keys only, which retains a single record when multiple inbound transitions share the same key.
  • Q: When is Keys only applicable? — Only when the inbound populations are homogeneous.
  • Q: What does Generate completement do? — It processes the remaining population in an Intersection or Exclusion; the complement contains the union of all inbound activity results excluding the intersection, and adds an additional outbound transition.
  • Q: What is the Primary set in an Exclusion? — The main population chosen in the Sets to join section, from which records found in the other sets are excluded.
  • Q: How do you exclude a target from a different dimension in an Exclusion? — Click Add a rule in the Exclusion rules section and define the dimension change conditions; reconciliation is then done via an attribute or a join.

Examples:

  • Union: a Combine activity merges the results of two queries, Loyalty Members and Purchasers, into a single, larger audience that includes all profiles from both segments.
  • Intersection: a Combine activity identifies profiles who are Loyalty Members and have made a purchase within the last month.
  • Exclusion: two queries are configured to filter profiles who purchased a product; profiles who do not have a loyalty membership are then excluded from the first set.

Personas:

  • marketer: Can add a Combine activity to a branch after building at least two execution branches. Can select Union, Intersection, or Exclusion as the segmentation type, then check previous activities in the Sets to join section. The page states that Keys only reconciliation is only applicable when the inbound populations are homogeneous. Where: “Configure the Combine activity”.

Onboarding:

  • Level: Intermediate
  • First minimal step: Add multiple activities, such as Build audience activities, to form at least two different execution branches.
  • Success check: The Union merges the results of two queries, Loyalty Members and Purchasers, into a single, larger audience that includes all profiles from both segments.
recommendation-more-help
journey-optimizer-help