This document outlines key recommendations while designing your Adobe Campaign data model.
Adobe Campaign system is very flexible and can be extended beyond the initial implementation. However, while possibilities are infinite, it is critical to make wise decisions and build strong foundations to start designing your data model.
For a better understanding of Campaign built-in tables and how they relate to each other, refer to this section .
💡 Read out this section to get started with Campaign schemas.
💡 Learn how to configure extension schemas in order to extend the conceptual data model of the Adobe Campaign database in this page.
Adobe Campaign is a powerful cross-channel campaign management system that can help you align your online and offline strategies to create personalized customer experiences.
While most email service providers are communicating to customers via a list-centric approach, Adobe Campaign relies on a relational database in order to leverage a broader view of the customers and their attributes.
To access the description of each table, go to Admin > Configuration > Data schemas, select a resource from the list and click the Documentation tab.
What data should be sent to Adobe Campaign? It is critical to determine the data required for your marketing activities.
Adobe Campaign is neither a data warehouse nor a reporting tool. Therefore, do not try to import all possible customers and their associated information into Adobe Campaign, or import data which will only be used to build reports.
To make the decision whether an attribute would be needed or not in Adobe Campaign, ask yourself if it would fall under one of these categories:
If not falling into any of these, you are most likely not going to need this attribute in Adobe Campaign.
To ensure good architecture and performance of your system, follow the best practices below to set up data in Adobe Campaign.
A field is required to be stored in a table if it has a targeting or personalization purpose. In other words, if a field is not used to send a personalized email or used as a criterion in a query, it will unnecessarily take up disk space.
In addition to the autouuid and autopk defined by default in most tables, you should consider adding some logical or business keys (account number, client number, and so on). It can be used later for imports/reconciliation or data packages. For more on this, see Identifiers.
Efficient keys are essential for performance. With Snowflake, you can insert nnumeric or string-based data types as keys for tables.
Adobe Campaign resources have three identifiers, and it is possible to add an additional identifier.
The following table describe these identifiers and their purpose.
|Name (or internal name)||
Adobe Campaign primary key is an auto-generated UUID for all built-in tables. A UUID can also be used for custom tables. Learn more
Even if the number of IDs is infinite, you should take care of the size of your database to ensure optimal performances. To prevent any issue, make sure to adjust your instance purge settings. For more on this, see this section.
Primary keys are required for every table created in Adobe Campaign.
Most organizations are importing records from external systems. While the physical key of the Recipient table is the “id” attribute, it is possible to determine a custom key in addition.
This custom key is the actual record primary key in the external system feeding Adobe Campaign.
When creating a custom table, you have two options:
An autouuid should not be used as a reference in workflows.
Beware of the “own” integrity on large tables. Deleting records that have large tables in “own” integrity can potentially stop the instance. The table is locked, and the deletions are made one by one. So it’s best to use “neutral” integrity on child tables that have large volumes.
Declaring a link as an external join is not good for performance. The zero-id record emulates the external join functionality. It is not necessary to declare external joins if the link uses the autouuid.
While it is possible to join any table in a workflow, Adobe recommends defining common links between resources directly in the data structure definition.
Link should be defined in alignment with the actual data in your tables. A wrong definition could impact data retrieved via links, for example unexpectedly duplicating records.
Name your link consistently with the table name: the link name should help understand what the distant table is.
Do not name a link with “id” as a suffix. For example, name it “transaction” rather than “transactionId”.
By default, Adobe Campaign will create a link using the primary key of the external table. For more clarity, it is preferable to explicitly define the join in the link definition.
When you design a link, make sure that the target record is unique when a 1-1 relationship has been declared. Otherwise the join may return multiple records when only one is expected. This results in errors during delivery preparation when “the query returns more rows than expected”. Set the link name to the same name as the target schema.
Define a link with a cardinality (1-N) in the schema on the (1) side. For example, the relation Recipient (1) – (N) Transaction should be defined in the transaction schema.
Note that a reverse cardinality of a link is (N) by default. It is possible to define a link (1-1) by adding the attribute revCardinality=‘single’ to the link definition.
If the reverse link should not be visible to the user, you can hide it with the link definition revLink=‘NONE’. A good use case for this is to define a link from recipient to the last transaction completed for example. You only need to see the link from recipient to the last transaction and no reverse link is required to be visible from the transaction table.
Links performing an external join (1-0…1) should be used with care as it will impact the system performance.
Adobe Campaign is neither a data warehouse nor a reporting tool. Therefore, to ensure good performance of the Adobe Campaign solution, database growth should stay under control. To achieve this, following some of the best practices below may help.
Regarding retention, built-in log tables in Campaign have pre-set retention periods on them, generally limiting their data storage to six months or less.
The following are the default retention values for built-in tables. Be aware that the retention configuration is set by Adobe technical administrators during implementation and values may vary for each implementation, based on customer requirements.
Custom tables are not purged with the standard cleanup process. While this might not be required on day one, do not forget to build a purge process for your custom tables as this could lead to performance challenges.
There are a few solutions to minimize the need of records in Adobe Campaign:
You can declare the “deleteStatus” attribute in a schema. It is more efficient to mark the record as deleted, then postpone the deletion in the cleanup task.
💬 As a Managed Cloud Services user, reach out to the Adobe consultants or technical administrators to learn more about retention or if you need to set retention for custom tables.
In order to ensure better performance at any time, follow the best practices below.
Adobe Campaign relies on third-party database engines. Depending on the provider, optimizing performance for larger tables may require a specific design.
Below are a few common best practices that should be followed when designing your data model using large tables and complex joins.
The table size is a combination of the number of records and the number of columns per record. Both can impact the performance of queries.
The number of rows impacts performance as well. The Adobe Campaign database is not designed to store historical data that are not actively used for targeting or personalization purpose - this is an operational database.
To prevent any performance issue related to the high number of rows, only keep the necessary records in the database. Any other record should be exported to a third-party data warehouse and removed from the Adobe Campaign operational database.
Here are a few best practices regarding the size of tables: