These use cases show the flexibility and power of data views in Customer Journey Analytics.
For example, when creating a data view, you could create an Orders metric from a pageTitle schema field that is a string. Here are the steps:
On the Components tab, drag the pageTitle into the Metrics section under Included Components.
Now highlight the metric you just dragged in and rename it under Component Settings on the right:
Open the Include/Exclude Values dialog on the right and specify the following:
The “confirmation” phrase indicates that this is an order. After reviewing all the page titles where those criteria are met, a “1” will be counted for each instance. The result is a new metric (not a calculated metric.) A metric that has included/excluded values can be used everywhere any other metric can be used. It works with Attribution IQ, filters, and everywhere else you can use standard metrics.
You can further specify an attribution model for this metric, such as Last Touch, with a Lookback window of Session.
You can also create another Orders metric from the same field and specify a different attribution model for it, such as First Touch, and a different Lookback window, such as 30 days.
Another example would be to use the Visitor ID, a dimension, as a metric to determine how many Visitor IDs your company has.
Previously, integers would automatically be treated as metrics in Customer Journey Analytics. Now, numerics (including custom events from Adobe Analytics) can be treated as dimensions. Here is an example:
Drag the call_length_min integer into the Dimensions section under Included Components:
You can now add Value Bucketing to present this dimension in a bucketed fashion in reporting. (Without bucketing, each instance of this dimension would appear as a line item in Workspace reporting.)
You can use a numeric dimension to get “metrics” into your Flow visualization.
This capability is specifically applicable to array-based fields. The include/exclude functionality lets you do filtering at the sub-event level, whereas filters (segments) built in the filter builder only give you filtering at the event level. So you can do sub-event filtering by using include/exclude in Data Views, and then reference that new metric/dimension in a filter at the event level.
For example, use the include/exclude functionality in Data Views to focus only on products that generated sales of more than 50 Dollars. So if you have an order that includes a 50 Dollar product purchase and a 25 Dollar product purchase, we would remove only the 25 Dollar product purchase, not the entire order.
These new settings allow you to view only high-value revenue and filter out anything below $50.
Your company may have spent time training your users to expect “Unspecified” in reports. The default in Data Views is “No Value”. You can now rename “No Value” to “Unspecified” in the Data Views UI.
Another example would be a dimension for a membership program registration. In this case, you could rename “No Value” to “No Membership Program Registration.”
Using the Duplicate feature at the top right, create a number of Revenue metrics with different attribution settings like First Touch, Last Touch, and Algorithmic.
Don’t forget to rename each metric to reflect the differences, such as “Algorithmic Revenue”:
You can determine whether a session is indeed the first-ever session for a user or a return session, based on the reporting window that you defined for this data view and a 13-month lookback window. This reporting lets you determine, for example:
What percentage of your orders are coming from new or return sessions?
For a given marketing channel, or a specific campaign, are you targeting first-time users or return users? How does this choice influence conversion rates?
One dimension and two metrics facilitates this reporting:
Session type - This dimension has two values: 1) New and 2) Returning. The New line item includes all of the behavior (i.e. metrics against this dimension) from a session that has been determined to be a person’s defined first session. Everything else is included in the Returning line item (assuming everything belongs to a session). Where metrics are not part of any session, they fall into the ‘Not applicable’ bucket for this dimension.
First-time Sessions. The First-time Sessions metric is defined as a person’s defined first session within the reporting window.
Return Sessions The Return Sessions metric is the number of sessions that were not a person’s first-time session.–>
To access these component:
95%-99% of the time, new sessions are reported accurately. The only exceptions are:
When a first session occurred before the 13-month lookback window. This session will be ignored.
When a session spans both the lookback window and the reporting window. Let’s say you run a report from June 1 to June 15, 2022. The lookback window would encompass May 1, 2021 to May 31, 2022. If a session were to start on May 30, 2022 and end on June 1, 2022, because the session is included in the lookback window, all sessions in the reporting window get counted as return sessions.
Schemas in Adobe Experience Platform contain Date and Date-Time fields. Customer Journey Analytics data views now support these fields. When you drag these fields into a data view as a dimension, you can specify their format. This format setting determines how the fields are displayed in reporting. For example:
For the Date format, if you select Day with the format Month, Day, Year, an example output in reporting might look like: August 23, 2022.
For the Date-Time format, if you select Minute of Day with the format Hour:Minute, your output might look like: 20:20.
We currently support dates after Jan 1, 1900 (with the single exception of Jan 1, 1970) and date-time values after Jan 1, 2000 00:00:00.
Date: A travel company is collecting the departure date for trips as a field in their data. They would like to have a report which compares the Day of Week for all departure dates collected to understand which is most popular. They would like to do the same for Month of Year.
Date-Time: A retail company is collecting the time for each of their in-store point-of-sale (POS) purchases. Over a given month, they would like to understand the busiest shopping periods by Hour of Day.