Pattern vs. Field Concatenation

Patterns and Field Concatenations are both powerful tools that help standardize and structure campaign metadata. While they can solve similar problems, each comes with unique capabilities and trade-offs that make them more or less suited to specific use cases. Scroll down for a quick comparison.

Patterns

Patterns create a key to which all metadata in downstream systems (ex. Adobe Analytics or Snowflake) will be associated. They can string together fields across field sets.

Elements include:

• Constant
• Variable (a field that changes and is not static; this can be a text field, a list, a date, an autofill, or a concatenation); you can autofill and concatenate any field in the template, no matter the field set
• Date
• Auto# (create a unique key with an incrementing number)
• Random (create a unique key with a random string of numbers and/or letters)

By default, Patterns lock once the submission has been committed. Once submitted, it cannot be edited by a user.

If the Pattern Elements include a Variable that the user wants to change, for example: If the Pattern is EM-{campaign}-{platform}-Auto# and returns EM-AlwaysOn-CheetahMail-123 but the user realizes they used the wrong Campaign Name, there are two options:

1. Create a new submission 
2. Have the admin convert the Pattern to an Unlocked Pattern and edit the dataset

Otherwise, the Pattern will not update to match the edited value.

NOTE - if you leave a field out it will result in a double delimiter.  

Field Concatenations

Field Concatenations string together any fields within the field set.

• Concatenations can be used in other Concatenations (for example, Final Campaign Name=Brand-LOB_Campaign Name_Flight Date).
• Field Concatenations can be modified by editing the submission! Fields are not locked and will update the concatenation as the included fields are updated.
• Because Field Concatenations can be easily modified, we do not recommend using Concatenated Fields for tracking code/campaign code/key/UTM ID purposes
• NOTE - if you leave an optional field out of a concatenation it will not result in a double delimiter as a pattern would.  It would just leave the missing field and extra delimiter out.

Quick Comparison

Patterns

Ideal for managing naming logic across multiple field sets or generating unique, secure keys.

Key Benefits:

• Supports Random elements to create a unique key with a random string of numbers and/or letters
• Can aggregate across field sets, enabling consistent logic across business units or data types 
• Can include constants, dates, auto-incrementing numbers, and variables from any field set
• Useful for standardizing IDs or naming logic across multiple datasets
• Pattern output locks on submission unless explicitly configured as "unlocked"

Best For:

• Content or campaign ID generation
• Secure and consistent naming across regions or teams
• Scenarios requiring reusable patterns across datasets

Field Concatenations

Straightforward, editable, and ideal for single field set structures.

Key Benefits:

• Easy to configure and modify—output updates with field changes
• Great for dynamic values that may change post-submission
• Supports nesting (e.g., combining concatenations within other fields)
• Automatically skips delimiters for empty fields

Limitations:

• Cannot include Random elements
• Restricted to fields within a single field set - cannot be used across multiple field sets which limits usefulness to control naming across disparate business units
• Only supports auto-numbers by default

Best For:

• UTM paramters
• Creative or campaign naming
• Simpler, template-specific metadata rules

How to Choose

• Use Field Concatenations for simplicity and when working within a single field set
• Use Patterns when you need randomized elements for more secure keys, cross-field-set logic across multiple datasets, or repeatable standards across templates - now with the option to unlock

Choosing the right method depends on where and how the data will be used, as well as how much flexibility and uniqueness you need in the output.