Utilizing the Audiences feature requires several steps to create and send audiences to downstream systems, e.g. Facebook, Yahoo!, Google Analytics, Nielsen. The high-level steps of how and what data is available for defining attributes, how to utilize the attributes to create audiences, and how to select the audiences to send to the downstream systems are described in the article Building an Audience Workflow.
This article dives into the details of the first of those steps, Creating and defining a Data Layer:
- Creating a Data Layer
- Configure the output location and result table names
- Select datasets to be used for attributes
- View the dependent apps of the data layer
- View attribute and audience results
Creating a Data Layer
The Data Layers page is available from the main menu (
) and then select the submenu item Data Layers under Audience or, if already within the Audience pages, via the data layer icon (
).
As noted in Creating and defining a Data Layer, typically an organization will only need a single data layer, but additional data layers can be created for each brand/customer-facing identity that has separate data set inputs and/or needs the resulting attributes and audiences to be separated. Syntasa staff will work with you to design and create the data layer(s).
If/when a new data layer does need to be created then simply click the green plus icon (
). Doing so will bring up the below screen allowing you to configure the data layer.
Configure the output location and result table names
Within a data layer, several pages are available from the left navigation bar. The initial creation and/or editing is done in the Configuration menu via the configuration gear icon (
).
Output location
Name - Unique name for the data layer. This name will be shown/selected when creating new attribute and audience apps as well as being selected within the audience sync app(s).
Event Store - Select an existing event store that will be used to house the attribute and audience result tables. It may be desired to create a new event store for this purpose, but an existing event store can be used.
However, since there is one set of attribute and audience result tables per data layer, and all attribute and audience apps created on a data layer share the result tables, each data layer must write to a unique event store.
Result tables
Partition Scheme - Typically set to Daily, this dictates the granularity of the result tables, i.e. if attribute and audience results are kept at a daily or hourly level.
Attribute Table Name - This is the name of the attribute result tables. The value is unchangeable after the data layer is created. Only one name is given here, but there are actually three tables: results, metadata, and metrics. Details are explained in the article Attribute Result Tables.
Audience Table Name - This is the name of the audience result tables. The value is unchangeable after the data layer is created. Only one name is given here, but there are actually three tables: results, metadata, and metrics. Details are explained in the article Audience Result Tables.
Attribute Metric Window Length - This field provides the breadth of a secondary analysis period for meeting attribute conditions. This is in addition to the recency field that is set at the attribute group. Thus, when reviewing attribute metric results there are two distinct time periods displaying counts for the number of visitors matching an attribute. The impact of this field is detailed in the article Attribute Result Tables.
Audience Metric Window Length - This field provides the breadth of a secondary analysis period for meeting audience conditions. The default period for analyzing audiences is a single day for the date of processing. Thus, when reviewing audience metric results there are two distinct time periods displaying counts for the number of visitors matching an audience. The impact of this field is detailed in the article Audience Result Table.
Select datasets to be used for attributes
Selecting the datasets within the data layer allows said datasets to be available as an input option when defining attributes. These datasets can be tables or views of raw clickstream data or they can be tables or views that are results of models created to generate specific audiences.
At least one dataset is required for a data layer. The application will not allow the data layer to be saved until at least one dataset is defined. On the creation of the data layer, the initial entry for a dataset will exist but be blank. The following fields:
- Dataset - This is the name of the table or view. In the dropdown, the full list of databases is shown flush to the left with the table and view names within a database indented under the database they are found.
- Time Column - The date/time column that is to be used for determining the date of the event needs to be selected. When multiple datasets are defined, these should be in the same timezone to ensure events are properly aligned.
- Identifier Column - The column to be used to identify the unique visitor/user ID needs to be selected. When multiple datasets are defined, these should refer to a common ID across datasets to ensure unique visitors are counted as such.
Adding an additional dataset - 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 - Removing a dataset is done by clicking the red minus icon (
) next to the particular dataset. At the time of saving the data layer, the system will validate that the dataset that is trying to be removed is not in use. If the dataset is in use by an app then a message stating that the data layer cannot be updated will be seen by the user.
View the dependent apps of the data layer
Within a data layer, several pages are available from the left navigation bar. The apps that are created on top of the data layer can be seen in the Dependencies menu via the dependencies icon (
).
All attribute and audience apps created on top of the data layer, when creating one of these apps users are asked to choose the data layer, will be seen here. Clicking on an app name will navigate you to the specific app, which can also be done directly from the Audience -> Apps menu.
The sync apps will not be shown here as they are can include audiences from multiple data layers.
View attribute and audience results
Within a data layer, several pages are available from the left navigation bar. Details and preview of the contents of the attribute tables of the data layer can be seen in the Attributes menu via the attributes icon (
).
Similarly, the details and preview of the contents of the audience tables of the data layer can be seen in the Audiences menu via the audiences icon (
).
As described in the article Building an Audience Workflow and above section Result tables, the attribute and audience result tables are actually three tables; each has a set within the development section and another in the production section (once an app has been activated); producing a total of six tables for each, i.e. attribute and audience.
Description of each table and their details displayed above within the attribute and audience sections of the data layer are summarized in the articles Attribute Result Tables and Audience Result Tables.