So, is there a solution that can make these challenges go away? YES there is and the solution to these challenges is Customer Journey Analytics. Customer Journey Analytics or CJA is an enterprise wide analytics product that is built on Adobe Experience Platform. CJA allows us to join different data sources (online & offline) to give a complete view of our customers in real-time across channels. Please note that CJA is considered an add-on to Adobe Analytics, also available for Non-Platform (AEP) customers and works natively with Adobe Experience Platform.
In this article, I'll compare Adobe Analytics with CJA based on a set of standard capabilities which are common between the two solutions and highlight some of the differences. The writeup is long but I've combined all the content in a single matrix at the end so feel free to scroll down to view it in one tabular view.
Adobe Analytics
In this section, I've listed the various capabilities of Adobe Analytics and added a high level writeup explaining each of these separately. I've done the same for Customer Journey Analytics.
1. Data Capture
o Primarily takes place based on the AppMeasurement library (client-side-web), Mobile SDK (mobile app), Data insertion API and Bulk Data Insertion API (server-side).
2. Data Usage
o Data is stored in Report Suites usually setup to receive data globally or individually based on the requirement.
o Virtual Report Suites (VRS) can be created to “split” data based on web/mobile, region or Business group and can be setup based on custom session timeouts, expiration and time zones.
3. Reporting and Analysis
o Data is visualized in Analysis Workspace or the legacy UI.
o Workspace panel includes Freeform, Cohort, Fallout etc. options available to visualize data.
o Calculated metrics can be created, and marketing channels can be used for further analysis.
o Robust data export capabilities (PDF, CSV etc. formats) as well as access to raw data feeds.
o Ability to setup alerts in case of anomalies.
4. Identity
o Primarily based on cookies for client-side web tagging.
o Based on ECID for mobile app (tied to each installed instance of the app).
o Customer IDs converted to ECID for server-side implementations in general.
o Device graph data can be accessed via the People metric or leveraged via Cross-Device Analytics.
5. Segmentation
o Segmentation built into Analysis Workspace
o Visitor, Visit and Hit segment containers available.
o Sequential segmentation and exclusion capabilities available to users.
6. Data Limitations
o Limited to 200 eVars/props and 1000 events.
o UI limited to 500K unique rows of data per month (Low Traffic).
7. Data Classifications
o Classifications subject to the same restrictions as the UI in terms of only classifying the top 500K rows.
8. Historical Data Ingestion
o Historical data sent in but out of order hits can affect the sequence of events and attribution of eVars and marketing channels.
9. User Permissions
o User permissions are granted via the Admin Console at a more granular level for report suites etc. at a product profile level.
10. Data Latency
o Data can take up to 2 hours to be fully available in Adobe Analytics.
|
Customer Journey Analytics
In this section, I've put CJA through the same set of capabilities as I did for Adobe Analytics. Please note that there are some features that CJA lacks compared to Adobe Analytics which the product team is working on to add support for. Those are explained in more detailed here.
1. Data Capture
o Data needs to be conformed to Adobe Experience Platform’s XDM schema to bring in any type of data.
2. Data Usage
3. Reporting and Analysis
4. Identity
5. Segmentation
6. Data Limitations
7. Data Classifications
8. Historical Data Ingestion
9. User Permissions
10. Data Latency
1. Data Capture
o Data needs to be conformed to Adobe Experience Platform’s XDM schema to bring in any type of data.
o Web SDK is used for real-time data streaming and streaming API will be available for sending data server-side.
2. Data Usage
o Data is stored in datasets created within Adobe Experience Platform and added to CJA as Connections.
o Data Views are similar to VRS which also allow us to define data based on the type of datasets being analyzed as well as setting custom session timeouts, expiration and defining separate time zones.
3. Reporting and Analysis
o Data in CJA is visualized in Analysis Workspace.
o Workspace panel includes Freeform, Cohort, Fallout etc. options available to visualize data.
o Calculated metrics can be created for further analysis, but marketing channel support is not available yet, but support is planned.
o No current ability to export data in CJA (Workspace) but support is planned. However, Query Service and Data Access API provides the ability to export data.
o No current ability to setup alerts but support is planned.
4. Identity
o Tied directly to the Namespace defined within Adobe Experience Platform.
o ID can be based on anything be it cookies, CRM id, Loyalty ID or Phone number.
o Custom namespaces can be defined.
o Data in the device graph is NOT available yet but support is planned.
5. Segmentation
o Filters built into Analysis Workspace.
o Person, Session and Event segment containers available.
o Leverages the same standard segmentation UI/features as Adobe Analytics.
6. Data Limitations
o Unlimited metrics and dimensions and data in eVars/props is available in XDM format within CJA.
o Unlimited number of rows and unique values.
7. Data Classifications
o Lookup Datasets created in Platform are not subject to any volume restrictions in terms of volume but there is a 1 GB limit which isn't "enforced".
8. Historical Data Ingestion
o Any missing historical data can be uploaded into Adobe Experience Platform and then leveraged in CJA including support for out of order hits for a person.
9. User Permissions
o Only product admins (not all users) can now perform granular tasks such as deleting, updating and sharing Workspace dashboards with other users.
10. Data Latency
o Data isn’t available in near real-time can be take up to 2 hours, but real-time support is being looked into.
|
Here's the matrix which consolidates the capabilities compared above in a tabular format. Please note that I took a stab at also calling out which solution is (currently) better for a particular capability by adding a checkmark. If there's no checkmark, then it means that the two solutions are on par with each other or support is planned to add that feature to CJA by the product team.
Hope this article provided you with some more information and context to figure out some similarities and differences between Adobe Analytics and Customer Journey Analytics. The key points to consider would be to see if you analyze large amount of dimensional data (exceeding 500K unique rows per month), often analyze customer data across multiple channels, need to add missing historical "hit level" data after the fact or connect offline data with online with the aim to get a single view of the customer, then you should seriously consider CJA.
Are you in the process of considering this tool or have any further questions? Feel free to post them here.
Are you in the process of considering this tool or have any further questions? Feel free to post them here.
3 comments:
Thanks for this. Nice article.
Hi Rohan. This is great info. Thanks for providing. My question is as we are heading to AEP (which is the future) will Analytics workspace and current visitorAPI libraries services be discontinued let's say in the next 2 or 3 years and everyone will have to transition to Customer Journey Analytics by then? Also would be great if there is any info or guidelines on the best practices to transition from current libraries to AEP.
Hi Sharan, I haven't heard anything yet about a set date when customers will have to migrate to CJA but the product team will know more. I have the same response for the client side libraries but my advise would be for you to take a closer look at the AEP Web SDK given that it's a streaming platform along with a server-side component which should serve you better for a future without cookies.
The main thing you can do is set up a POC of the Web SDK in your local environment and prepare for it as diligently and even more like clients did when they migrated from DTM to Launch. Hope it helps!
Post a Comment