4.1.2 Connect 51ºÚÁϲ»´òìÈ Experience Platform Datasets in Customer Journey Analytics
Objectives
- Understand the Data Connection UI
- Bring 51ºÚÁϲ»´òìÈ Experience Platform data into CJA
- Understand Person ID and data stitching
- Learn the concept of data streaming in Customer Journey Analytics
4.1.2.1 Connection
Go to to access Customer Journey Analytics.
On the Customer Journey Analytics-homepage, go to Connections.
Here you can see all the different connections made between CJA and Platform. These connections have the same goal as report suites in 51ºÚÁϲ»´òìÈ Analytics. However, the collection of the data is totally different. All data is coming from 51ºÚÁϲ»´òìÈ Experience Platform datasets.
Let’s create your first connection. Click Create new connection.
You’ll then see the Create Connection UI.
You can now give a name to your connection.
Please use this naming convention: --aepUserLdap-- – Omnichannel Data Connection
.
You also need to select the correct sandbox to use. In the sandbox menu, select your sandbox, which should be --aepSandboxName--
. In this example, the sandbox is Tech Insiders. You also need to set the Average number of daily events to less than 1 million.
After selecting your sandbox, you can start adding datasets. Click Add datasets.
4.1.2.2 Select 51ºÚÁϲ»´òìÈ Experience Platform datasets
Search for the dataset Demo System - Event Dataset for Website (Global v1.1)
. Enable the box for this dataset to add it to this connection.
Stay in the same screen, and now search and check the checkbox for Demo System - Event Dataset for Call Center (Global v1.1)
.
You’ll then have this. Click Next.
4.1.2.3 Person ID and Data Stitching
Person ID
The goal now is to join these datasets. For every dataset you selected, you’ll see a field called Person ID. Each dataset has its own Person ID field.
As you can see, most of them have the Person ID selected automatically. This is because a Primary Identity is selected in every schema in 51ºÚÁϲ»´òìÈ Experience Platform. As an example, here’s the schema for Demo System - Event Schema for Website (Global v1.1)
, where you can see that the Primary Identity is set to ecid
.
However, you can still influence which identifier will be used to stitch datasets together for your connection. You can use any identifier that is configured in the schema linked to your dataset. Click on the dropdown to explore the IDs available on each dataset.
As mentioned, you can set different Person IDs for each dataset. This allows you to bring different datasets from multiple origins together in CJA. Imagine bringing in NPS or survey data which would be very interesting and helpful to understand the context and why something has happened.
The name of the Person ID field isn’t important, as long as the value in the Person ID fields correspond. Let say we have email
in one dataset and emailAddress
in another dataset defined as Person ID. If delaigle@adobe.com
is the same value for the Person ID-field on both datasets, CJA will be able to stitch the data.
Review the CJA FAQs here to understand the nuances with identity stitching: FAQ.
Stitching the data using the Person ID
Now that you understand the concept of stitching datasets using the Person ID, let’s choose email
as your Person ID for each dataset.
Go to each dataset to update the Person ID. Now fill the field Person ID choosing the email
on the dropdown list.
Once you have stitched the two datasets, you’re ready to continue.
You also need to ensure that for both datasets, these options are enabled:
- Import all new data
- Backfill all existing data
(Don’t forget to enable both these options for the second dataset)
You also need to select a Data source type for each dataset.
These are the settings for the dataset Demo System - Event Dataset for Website (Global v1.1).
These are the settings for the dataset Demo System - Event Dataset for Website (Global v1.1).
Click Add datasets.
Click Save and go to the next exercise.
After having created your Connection it may take a few hours before your data is available in CJA.
Next Step: 4.1.3 Create a Data View