You can use a Flow visualization with the Dataset ID dimension.

1. Log in to Customer Journey Analytics and create a blank Workspace project.
2. Select the Visualizations tab on the left, and drag a Flow visualization to the canvas on the right.
3. Select the Components tab on the left, and drag the dimension Dataset ID to the center location labeled Dimension or Item.
4. This flow report is interactive. To expand the flows to subsequent or previous pages, select any of the values. Use the right-click menu to expand or collapse columns. Different dimensions can also be used within the same flow report.

If you would like to rename dataset ID dimension items, you can use a lookup dataset.

The lookback window for rekeying depends on your desired frequency of data replay. For example, if you set up stitching to replay data once every week, the lookback window for rekeying is seven days. If you set up stitching to replay data every day, the lookback window for rekeying is one day.

In some situations, it is possible that multiple people log in from the same device. Examples include a shared device at home, shared PCs in a library, or a kiosk in a retail outlet.

The transient ID overrides the persistent ID, so shared devices are considered separate people (even if they originate from the same device).

In some situations, an individual user can associate with many persistent IDs. An example is an individual frequently clearing browser’s cookies or using the browser’s private/incognito mode.

For field-based stitching, the number of persistent IDs is irrelevant in favor of the transient ID. A single user can belong to any number of devices without impacting Customer Journey Analytics’s ability to stitch across devices.

For graph-based stitching, a single person can have many persistent ID in the identity graph. Graph-based stitching uses the persistent ID based on the specified namespace. In case there are more persistent ID for the same namespace, the lexicographic first persistent ID is used.

Live stitching is available approximately one week after Adobe enables stitching. Backfill availability depends on the amount of existing data. Small datasets (less than 1 million events per day) typically take a couple days, while large data sets (1 billion events per day) can take a week or more.

Cross-device analytics is a feature specific to traditional Adobe Analytics that allows you to understand how people operate across devices. It offers two workflows to link device data together: field-based stitching and the device graph.

Cross-channel analysis is a use case specific to Customer Journey Analytics that allows you to understand how people operate across both devices and channels. It stitches a dataset’s person ID, allowing that dataset to be seamlessly combined with other datasets. This feature operates in design similarly to cross-device analytics field-based stitching, but the implementation is different due to differing data architecture between traditional Analytics and Customer Journey Analytics. See Stitching and the cross-channel analysis use case for more information.

Adobe handles privacy requests in accordance with local and international laws. Adobe offers the Adobe Experience Platform Privacy Service to submit data access and deletion requests. The requests apply to both the original and rekeyed datasets.

To illustrate, imagine the following data for identities, events before stitching and after stitching.

Current process for privacy request

When a privacy request is received for customer with CustID Bob, the rows with strikethrough entries are deleted. Other events are restitched using the identity map. For example, the first stitched id in the stitched dataset is updated to Alex.

Identity Map	Id	timestamp	persistent ID	persistent namespace	transient id	transient namespace
	1	ts1	123	ecid	Bob	CustId
	2	ts2	123	ecid	Alex	CustId

Events dataset	Id	timestamp	persistent ID	persistent namespace	transient id	transient namespace
	1	ts0	123	ecid
	2	ts1	123	ecid	Bob	CustId
	3	ts2	123	ecid	Alex	CustId

Stitched dataset	Id	timestamp	persistent ID	persistent namespace	transient id	transient namespace	Stitched ID	Stitched namespace
	1	ts0	123	ecid			Alex	CustId
	2	ts1	123	ecid	Bob	CustId	Bob	CustId
	3	ts2	123	ecid	Alex	CustId	Alex	CustId

New process for privacy request

When a privacy request is received for customer with CustID Bob, the rows with strikethrough entries are deleted. Other events are restitched using the persistent id. For example, the first stitched id in the stitched dataset is updated to 123.

Identity Map	Id	timestamp	persistent ID	persistent namespace	transient id	transient namespace
	1	ts1	123	ecid	Bob	CustId
	2	ts2	123	ecid	Alex	CustId

Events dataset	Id	timestamp	persistent ID	persistent namespace	transient id	transient namespace
	1	ts0	123	ecid
	2	ts1	123	ecid	Bob	CustId
	3	ts2	123	ecid	Alex	CustId

Stitched dataset	Id	timestamp	persistent ID	persistent namespace	transient id	transient namespace	Stitched ID	Stitched namespace
	1	ts0	123	ecid			123	ecid
	2	ts1	123	ecid	Bob	CustId	Bob	CustId
	3	ts2	123	ecid	Alex	CustId	Alex	CustId

If the Persistent ID field is blank on an event in a dataset being stitched , the Stitched ID for that event in determined in one of two ways:

• If the Transient ID field is not blank, Customer Journey Analytics uses the value in Transient ID as the Stitched ID.
• If the Transient ID field is blank, Customer Journey Analytics also leaves the Stitched ID blank. In this case, Persistent ID, Transient ID, and Stitched ID are all blank on the event. These types of events are dropped from any Customer Journey Analytics connection using the dataset being stitched where Stitched ID was chosen as the Person ID.

Be cautious of ‘person collapse’, which occurs when stitching is applied to data that uses placeholder values for Transient IDs. In the example table below, undefined person IDs originating from a dataset sourced from a CRM system are populated with the value ‘Undefined’ resulting in incorrect representation of persons.

Certain metrics in Customer Journey Analytics are similar to metrics in traditional Analytics, but others are different, depending on what you are comparing. The table below compares several common metrics:

Other metrics can be similar in Customer Journey Analytics and Adobe Analytics. For example, the total count for Adobe Analytics custom events 1-100 is comparable between traditional Adobe Analytics and Customer Journey Analytics (whether stitched or unstitched). Differences in capabilities) such as event de-duplication between Customer Journey Analytics vs. Adobe Analytics can cause discrepancy between the two products.

No, Customer Journey Analytics cannot currently use Identity Map fields for stitching.

Data does not have to be reingested into Experience Platform, however it will need to be reconfigured in Customer Journey Analytics. Please follow these steps:

1. Setup the new graph-based stitched dataset.
2. Configure the new dataset as part of a new connection in Customer Journey Analytics.
3. Switch your existing Data View to use the new connection (and as such the new graph-based stiched dataset
4. Remove the old connection that was using the field-based stitched dataset.

Not if you follow the steps outlined above. Otherwise, please ask Adobe Consulting for additional support.