Creating a Data layer [CCDP]

As mentioned in the Data Layer Overview, most organizations typically require only a single data layer. However, additional data layers can be created to support specific needs, such as distinct brands or customer-facing identities with separate dataset inputs or cases where resulting attributes and audiences need to remain independent. Syntasa staff will assist you in designing and creating the necessary data layer(s).

This article covers following topics:

• Creating a new Data Layer

• Data Layer fields

• Updating the Data Layer Details

NOTE: This article has been repurposed to include categorical, scheduling and activation fields required particularly for CCDP. The original article for creating a data layer can be found here.

Creating a New Data Layer

To create a new data layer, follow these steps:

1. Open the Audience module by clicking the hamburger menu and selecting Audience.

2. From the left-side menu, select Data Layer (the first item) to navigate to the Data Layer list screen.

3. Click the Create New (+) icon to open the Create Data Layer screen.

4. Complete all the required fields and click the Save button to finalize the creation.

   

Data Layer Fields

Let's discuss each field in detail:

Name

• Represents the unique name of the data layer.

• This name is used across all Audience screens, including Attributes, Sync, Operations, Datasets etc. Selecting this name on the screens ensures that only data related to the chosen data layer is displayed.

Event Store

• It allows you to select an existing event store to house the attribute and audience result tables.

• You can create a new event store for better organization, though an existing one can also be used.

• A single set of attribute and audience result tables is maintained per data layer, shared by all attribute and audience apps within that data layer. Therefore, each data layer must be written to a unique event store to ensure data separation.

• An event store already used by another data layer cannot be reused.

Attribute Table Name

• Specifies the name of the attribute result tables.

• The name is set to syn_attribute by default.

• This value cannot be changed once the data layer is created.

• Represents three tables:

  • Results Table: Stores the detailed attribute results.

  • Metadata Table: Contains metadata related to the attribute results.

  • Metrics Table: Includes metrics derived from the attribute results.

• For more information, refer to the Attribute Result Tables article.

Audience Table Name

• Specifies the name of the audience result tables.

• The name is set to syn_audience by default.

• This value cannot be changed once the data layer is created.

• Represents three tables:

  • Results Table: Stores the detailed audience results.

  • Metadata Table: Contains metadata related to the audience results.

  • Metrics Table: Includes metrics derived from the audience results.

• For more information, refer to the Audience Result Tables article.

Attribute Metric Window Length

• Defines the secondary analysis period for evaluating attribute conditions, extending beyond the recency field set at the attribute group level.

• The value is set to 7 by default.

• When reviewing attribute metric results, you’ll see two distinct periods:

  • The Recency Period: Displays the number of visitors matching the attribute based on the recency setting and processed dates.

  • The Metric Window Period: Shows the visitor count based on the additional analysis period defined here.

• For more details on the impact of this field, refer to the Attribute Result Tables article.

Audience Metric Window Length

• Defines the secondary analysis period for evaluating audience conditions, extending beyond the default setting.

• The value is set to 7 by default.

• The default period for analyzing audiences is a single day, corresponding to the processing date.

• When reviewing audience metric results, you’ll see two distinct periods:

  • The Default Period: Displays the number of visitors matching the audience for the processing date.

  • The Metric Window Period: Shows the visitor count based on the additional analysis period defined here.

• For more information on the impact of this field, refer to the Audience Result Tables article.

Datasets to be used for Attributes

• Selected datasets serve as input options when defining attributes.

• These datasets can include:

  • Tables or views of raw clickstream data.

  • Tables or views that represent results from models created to generate specific audiences.

• At least one dataset must be defined for a data layer.

• On the creation of the data layer, the initial entry for a dataset will exist but be blank.

• Each dataset entry includes the following 3 fields:

  • Dataset: Represents the name of the table or view to be used.

    • The dropdown displays a complete list of production databases, with:

      • Database Names are aligned to the left.

      • Tables and Views are indented under the respective database.

  • Time Column: Specifies the date/time column used to determine the event date.

    • For multiple datasets, ensure that all time columns are in the same timezone to maintain proper event alignment.

  • Identifier Column: Specifies the column representing the unique visitor/user ID.

    • When using multiple datasets, the identifier column should reference a common ID across datasets to ensure that unique visitors are accurately identified.

  • Category: Maps the dataset to a specific logical category within the Campaign Rule Builder.

    • Predictive Traits (AI): Select this for datasets containing AI-generated model scores, such as churn probability or lifetime value (LTV).

    • Profile Attributes: Use this for datasets containing static user properties like location, gender, or loyalty tier.

    • Behavioral Events: Select this for datasets that track user actions and events over time.

    • Audience Segments: Use this for datasets that represent membership in pre-existing lists or segments.

  • Type: Defines the processing and evaluation frequency for the dataset logic.

    • Batch: Data is evaluated at standard intervals (e.g., daily or every 15 minutes).

    • Streaming: Data is evaluated in real-time as events occur.

  • Campaign Type: A multi-select field that determines which delivery channels can access and use this dataset during campaign configuration.

    • Options include Web, Email, and Paid Media.

• Adding an additional dataset is done by clicking the green plus icon () under the existing datasets. When adding additional datasets, keep in mind the selections of Time Column and Identifier Column as noted above.

• Removing a dataset is done by clicking the red minus icon () next to the particular dataset.Note: During the saving process, the system validates whether a dataset being removed is currently in use. If the dataset is used by an attribute group, the system will prevent the data layer from being updated. A message will be displayed to inform the user that the data layer cannot be updated due to the dataset being in use.

Updating the Data Layer

A data layer can be modified at any time after its creation. The following fields become locked after the data layer is created and cannot be updated:

• Event Store

• Attribute Table Name

• Audience Table Name

• Partition Scheme