Frequently asked questions

Customer Journey Analytics (CJA) is our next-generation analytics product. Following are answers to frequently asked questions about CJA. For more information, review Customer Journey Analytics feature support.

1. Prerequisites

 Do I need Private Device Graph or Device Coop for Customer Journey Analytics?

No, Private Device Graph or Device Coop are not required for Customer Journey Analytics. In fact, they are not yet supported.

 Do I need Experience Cloud ID (ECID) for Customer Journey Analytics?

No, Customer Journey Analytics supports any ID in a dataset, whether that’s ECID or any other ID you choose.

 What if I need to ETL (Extract, Transform, Load) my data prior to Customer Journey Analytics?

Customer Journey Analytics includes Data Prep capabilities to help transform your data before putting it into Adobe Experience Platform data lake. If you need ETL after the data has already been ingested, Adobe Experience Platform Query Service provides some limited options, although there may be extra fees involved.

2. Stitching data (Cross-Channel Analytics)

 Can Customer Journey Analytics "stitch" across devices or across datasets?

Yes. Customer Journey Analytics has a stitching solution called Cross-Channel Analytics (CCA). It lets you re-key a dataset’s person ID, which enables a seamless combination of multiple datasets.

 Is stitching from anonymous behavior to authenticated behavior supported?

Yes. Cross-Channel Analytics looks at user data from both authenticated and unauthenticated sessions to generate a stitched ID.

 How does 'replay' work in CCA?

CCA “replays” data based on unique identifiers it has learned. Replay causes new devices to the connection to become stitched. Learn more

 How does stitching historical data (backfill) work in CCA?

When first turned on, Adobe provides a backfill of stitched data that goes back as far as the beginning of the previous month (up to 60 days.) In order to do this backfill, the transient ID must exist in the unstitched data that far back in time. Learn more

 What is the expected behavior for non-stitched profile dataset records?

Example scenario: You join 2 datasets in a CJA connection by using CRMid as the Person ID. One is a Web Event dataset with CRMid in all records. The other dataset is a CRM profile dataset. 40% of the CRM dataset has CRMid present in the Web event dataset. The other 60% is not present in the Web event dataset - do these records appear in reporting in Analysis Workspace?

Answer: Profile rows that have no events tied to them are stored in CJA. However, you cannot view them in Analysis Workspace until an event tied to that ID appears.

3. Getting data into Customer Journey Analytics

 Can I combine data from different Adobe Experience Platform sandboxes in one Customer Journey Analytics connection?

No, you cannot access data across sandboxes. You can combine only datasets that are located within the same sandbox. Learn more

 How do I connect online data to offline data in Customer Journey Analytics?

As long as the person ID matches between datasets, Customer Journey Analytics can connect filters, attribution, flow, fallout, etc. across datasets.

 How do I bring my offline data into Customer Journey Analytics?

Your entitlement to Customer Journey Analytics allows you to ingest data into Experience Platform. You can then create connections to that data and data views in Customer Journey Analytics, for reporting in Analysis Workspace. The Experience Platform’s data on-boarding team can help provide recommendations or consulting for you, if needed.

 How do I get Adobe Analytics data into Customer Journey Analytics?

Adobe Analytics data can be connected to Experience Platform through the Adobe Analytics Source Connector. Most Adobe Analytics fields are brought over in XDM format, but other fields are not yet available.

 How long does it take to assemble dataset elements into a data view?

A few hours to get started, and a few days to backfill the last 13 months of data.

 Is it necessary to bring PII data to establish connections between the data?

No, you can use any ID, including a hash of a customer ID, which is not PII.

 What are the limits for ingesting past or future dates/timestamps into CJA event datasets?
  • Regarding past dates/timestamps: Event data up to 10 years old.
  • Regarding future dates/timestamps: Event data (predictive) up to 1 month in the future.

4. Latency considerations

NOTE

There is no fixed data size in CJA and thus Adobe cannot commit to a standard ingestion time. We are actively working to reduce these latencies through new updates and ingestion optimization.

 What is the expected latency for Customer Journey Analytics on Adobe Experience Platform?
  • Live data or events: Processed and ingested within 90 minutes, once data is available in AEP. (Batch size > 50 million rows: longer than 90 mins.)
  • Small backfills - For example, a lookup dataset of 10 million rows: within 7 days
  • Large backfills - For example, 500 billion rows: 30 days

5. Set rolling window for Connection data retention

The Enable rolling data window setting lets you define CJA data retention as a rolling window in months (3 months, 6 months, etc.). It is set at a connection level, not at a dataset level. Data retention is based on event dataset timestamps and applies to event datasets only. No data retention setting exists for profile or lookup datasets since there are no applicable timestamps.

The main benefit is that you store or report only on data that is applicable and useful and delete older data that is no longer useful. It helps you stay under your contract limits and reduces the risk of overage cost.

6. Implications of deleting data components

When it comes to data deletion, we are concerned with 6 types of components: sandbox, schema, dataset, connection, data view, and Workspace project. Here are some possible scenarios around deleting any of these components:

If you… This happens…
Delete a sandbox in Adobe Experience Platform Deleting a sandbox will stop data flow to any Customer Journey Analytics connections to datasets in that sandbox. Currently, Connections in CJA tied to the deleted sandbox will not automatically be deleted.
Delete a schema in Adobe Experience Platform, but not the dataset/s associated with this schema Adobe Experience Platform does not allow for the deletion of schemas that have one or more datasets associated with them. However, an Admin with the appropriate set of rights can delete the datasets first and then delete the schema.
Delete a dataset in Adobe Experience Platform data lake Deleting a dataset in AEP data lake will stop data flow from that dataset to any CJA Connections that include that dataset. Any data from that dataset is automatically deleted from associated CJA Connections.
Delete a dataset in Customer Journey Analytics Contact your Adobe Account Team to set in motion the process for deleting a dataset within a connection that has been saved.
Delete a batch from a dataset (in Adobe Experience Platform) If a batch is deleted from an Adobe Experience Platform dataset, the same batch will be removed from any CJA Connections that contain that specific batch. CJA is notified of batch deletions in Adobe Experience Platform.
Delete a batch while it is being ingested into Customer Journey Analytics If there is only one batch in the dataset, no data or partial data from that batch will appear in Customer Journey Analytics. The ingestion will be rolled back. If, for example, there are 5 batches in the dataset and 3 of them have already been ingested when the dataset was deleted, data from those 3 batches will appear in Customer Journey Analytics.
Delete a connection in Customer Journey Analytics An error message will indicate that:
  • Any data views created for the deleted connection will no longer work.
  • Similarly, any Workspace projects that depend on data views in the deleted connection will cease working.
Delete a data view in Customer Journey Analytics An error message will indicate that any Workspace projects that depend on this deleted data view will cease working.

7. Considerations when merging report suites in CJA

If you plan to ingest Adobe Analytics data through the Adobe Analytics source connector, consider these ramifications when merging 2 or more Adobe Analytics report suites.

Issue Consideration
Variables Variables such as eVars may not line up across report suites. For example, eVar1 in report suite 1 may point to Page. In report suite 2, eVar1 may point to Internal Campaign, leading to mixed and inaccurate reporting.
Sessions and People counts They get deduplicated across report suites. As a result, counts may not match.
Metric deduplication Deduplicates instances of a metric (for example, Orders) if multiple rows have the same transaction ID (for example, Purchase ID). This prevents over-counting of key metrics. As a result, metrics like Orders may not add up across report suites.
Currency Currency conversion is not yet supported in CJA. If the report suites you are trying to merge use different base currencies, problems may arise.
Persistence Persistence extends across report suites, which impacts filters, attribution, and so on. Numbers may not add up properly.
Classifications Classifications do not automatically get deduplicated when merging report suites. When combining multiple classifications files into a single lookup dataset, you could encounter problems.

8. Traditional Adobe Analytics components

 Can I share/publish filters (segments) from Customer Journey Analytics to Experience Platform Unified Profile, or other Experience Cloud applications?

Not yet, but we are actively working to deliver this capability.

 What happened to my old eVar setting?

eVars, props, and events in the traditional Adobe Analytics sense no longer exist in Customer Journey Analytics. You have unlimited schema elements (dimensions, metrics, list fields). So all of the attribution settings you used to apply during the data collection process are now applied at query time.

 Where are all my session and variable persistence settings now?

Customer Journey Analytics applies all of these settings at report time, and these settings now live in Data Views. Changes to these settings are now retroactive, and you can have multiple versions by using multiple Data Views!

 What happens to our existing segments/calculated metrics?

Customer Journey Analytics no longer uses eVars, props, or events and instead uses any AEP schema. This means none of the existing segments or calc metrics are compatible with Customer Journey Analytics.

 How does Customer Journey Analytics handle Uniques Exceeded limitations?

Customer Journey Analytics has no unique value limitations, so no need to worry about them!

 If I am an existing Data Workbench customer, can I move to Customer Journey Analytics right now?

It depends on your use case - please work with your Adobe Account team. Your current use cases may already be a good fit for Customer Journey Analytics!

9. Estimate connection size

Refer to Estimate and manage usage.

10. Regarding usage overages

Usage limits are regularly monitored and enforced by Adobe. “Rows of data” means the daily average rows of data available for analysis within Customer Journey Analytics.

For example, let’s say your contract entitles you to one million rows of data. Suppose that on day 1 of using Customer Journey Analytics, you upload two million rows of data. On day 2, you delete 1 million rows and keep your usage at that committed maximum (i.e., one million row of data) for the remainder of your License Term. Depending on your contractual terms, you may still incur prorated over-usage fees for day 1, since you exceeded your “rows of data” license entitlement.

11. Diagnose data discrepancies

In some cases, you may notice that the total number of events ingested by your connection is different than the number of rows in the dataset in Adobe Experience Platform. In this example, the dataset “B2B Impression” has 7650 rows, but the dataset contains 3830 rows in Adobe Experience Platform. There are several reasons why discrepancies can happen, and the following steps can be taken to diagnose:

  1. Break down this dimension by Platform Dataset ID and you will notice two datasets with same size but different Platform Dataset IDs. Each dataset has 3825 records. That means Customer Journey Analytics ignored 5 records due to missing person IDs or missing timestamps:

    breakdown

  2. In addition, if we check in Adobe Experience Platform, there is no dataset with Id “5f21c12b732044194bffc1d0”, hence someone deleted this particular dataset from Adobe Experience Platform when the initial connection was being created. Later it got added to Customer Journey Analytics again, but a different Platform Dataset ID was generated by Adobe Experience Platform.

Read more about the implications of dataset and connection deletion in Customer Journey Analytics and Adobe Experience Platform.

Business.Adobe.com resources

Docs links
Flow Analysis Lookalike Modeling Data Ingestion Ad Hoc Analysis Predictive Analytics Data Analytics

On this page