Improve your catalog through optimizations

Optimizations allow you to enhance and transform your product catalog data so it is properly structured, enriched and adapted to the requirements of different marketing platforms.

By applying optimizations, you can improve the quality of your product information as well as enhance the visual appeal, making your catalog more relevant, consistent, and attractive to potential customers, while also ensuring compliance with platform specifications.

Each optimization is composed of two key elements:

• Filters: define which products the optimization will be applied to.
• Product Data Rules: define what changes will be applied to those products.
  
  

Filters

1. Filter conditions (operators)
2. Subfilters: AND vs OR

Product Data Rules

1. Replace
2. Compose
3. URL Params
4. Layout

Order of the optimizations

Filters

Filters are the first step when creating an optimization. They allow you to define the specific group of products that will be affected by the Product Data Rules.

• If no filter is applied, the optimization will affect the entire catalog
• If a filter is defined, the optimization will only apply to the selected subset of products

This provides full flexibility to apply both global and granular optimizations depending on your strategy.

Filter conditions (operators)

Filters are based on conditions that use different operators to define which products should be included or excluded from an optimization. By combining fields (such as title, price, brand, availability or custom attributes) with these operators, you can build very precise product selections.

Below is an overview of the available operators and how they work:

String-based operators (text and identifiers)

These operators work with any string value, meaning they can be applied to both text fields (e.g. title, brand) and alphanumeric fields (e.g. item_group_id, id).

• CONTAINS: selects products where the field includes a specific value.
  Example: title CONTAINS “running” → products with “running” in the title

• NOT CONTAINS: excludes products containing a specific value.
  Example: title NOT CONTAINS “used”

• IS EQUAL TO: matches an exact value.
  Example: brand IS EQUAL TO “Nike”

• IS NOT EQUAL TO: excludes an exact match.
  Example: brand IS NOT EQUAL TO “Adidas”

• STARTS WITH: matches values that begin with a specific string.
  Example: item_group_id STARTS WITH “PB”

• NOT STARTS WITH: excludes values that begin with a specific string

• ENDS WITH: matches values that end with a specific string
  Example: item_group_id ENDS WITH “V2026”

• NOT ENDS WITH: excludes values that end with a specific string

List-based operators

These operators allow you to filter using multiple values at once.

• IN: Selects products that match any value in a list (comma-separated)
  Example: ID IN “556789, 556790, 556791”

• NOT IN: Excludes products that match any value in the list

Length-based operators

Useful for validating or optimizing text fields.

• IS LONGER THAN: selects values with more characters than a given number.
  Example: title IS LONGER THAN 6

• IS SHORTER THAN: selects values with fewer characters than a given number.

Numeric operators

Used for numerical fields such as price, stock, or custom metrics.

• IS GREATER THAN
  Example: price IS GREATER THAN 50

• IS GREATER THAN OR EQUAL TO

• IS LESS THAN

• IS LESS THAN OR EQUAL TO

These operators are especially useful for price segmentation, stock thresholds, or performance-based optimizations.

Time-based operators

These operators are based on the field new_since, which is automatically generated and tracks when a product was added to the catalog.

• NEWER THAN X DAYS: Recently added products

  Example: ''new_since'' newer than 7 days

• OLDER THAN X DAYS: Older products in the catalog

  Example: ''new_since'' older than 30 days

These operators enable you to build highly targeted and strategic product selections, allowing you to apply optimizations only where they are most relevant and impactful.

Subfilters: AND vs OR

You can refine your selection further by adding subfilters, combining multiple conditions using AND or OR logic.

• AND: All conditions must be met

  Example: brand = “Nike” AND price > 50

  Only Nike products with price above 50

• OR: At least one condition must be met

  Example: brand = “Nike” OR brand = “Adidas”

  Products from either Nike or Adidas

Using subfilters allows you to build more advanced and precise targeting strategies within your catalog.

 

Product Data Rules

Replace

The Replace rule is one of the most useful Product Data Rules when it comes to refining and optimizing your catalog data. It allows you to modify specific parts of a field by replacing exact characters or words with new ones, helping you standardize and improve your product information for better performance across advertising platforms.

In essence, this rule scans the selected field and replaces a defined string of characters with another, exactly as specified.

How it works

When configuring a Replace rule, you need to define:

- The field you want to modify

- The text to find

- The text to replace it with

When to use Replace

This rule is particularly useful when:

• Standardizing product titles or descriptions

• Fixing inconsistencies in naming conventions

• Enhancing product data with additional keywords

• Aligning catalog data with feed requirements

Examples

If we need to remove unwanted or invalid characters from the attributes:

If we want to update values from the attributes, like changing the title from: “latest generation mobile” to “latest generation mobile phone”, we would configure the rule as follows:

This allows you to enrich product titles, improve keyword coverage and align your catalog with platform requirements or campaign strategies.

 

Compose

The Compose rule is a powerful Product Data Rule that allows you to fully control how your catalog fields are created, structured and optimized. It allows to create new fields, overwrite existing ones using static or dynamic data and delete fields from the catalog.

1. Create new fields with static or dynamic information

To add new information to your catalog, the first step is to write the name of the new field and then define whether the value will be static or dynamic.

• Static value: use this option when you want to assign the same value across a product set or the entire catalog.

Examples:

If we want to add the ''brand'' field to the catalog, we would do as follows:

If we want to group products than are on sale under a custom_label:

If we want to tag products to not be read by the platform:

 

• Dynamic value: use this option when you want to transfer information from one existing field to another. You need to define the new field name and use double curly brackets to reference the source field.

Examples:

If we want to send the information from the field performance_data to platforms, but it is not an accepted field, you can map it into a valid one like custom_label:

If you need to transform field data for layout purposes without altering the original catalog values, it is recommended to create a new field dynamically. If you later need to adjust or refine these values, you can combine this setup with the Replace Product Data Rule:

 

2. Concatenate fields dynamically

You can combine multiple fields dynamically by using double curly brackets . This allows you to enrich your catalog by merging attributes into a single field.

Examples:

If you want to optimize the title field by adding more relevant information such as brand, size and color, you can do it like this:

You can also combine static and dynamic data, for instance:

 

3. Remove information

You can delete the content of a field by leaving the value empty.

Example:

Although fields that are not required or recommended by the platform will not be read by it, there are times when it is necessary to remove the information from an attribute. This can be done easily by adding the field you want to delete and leaving its value blank.

URL Params

The URL Params rule allows you to append static parameters to your product URLs. This is especially useful for correct tracking campaigns, attribution and analytics. 

With this rule, you can include the static parameters indicated in the URL, ensuring that all product links carry the necessary tracking information for your paid media campaigns.

Example:

If we want to add utm_source=Facebook and utm_content=christmas_campaign we would simply add them separated by the symbol &

Layout

Once you have created your design using Creatives, you can easily apply it through the Layout Product Data Rule. This rule allows you to optimize your visual assets by updating the relevant fields in your catalog with the new design.

Depending on the type of layout you create, different fields will be optimized:

• Image layouts: If your design uses an image placeholder, the rule will overwrite the image_link and additional_image_link fields with the new images.

• Video layouts: If your design uses a video placeholder, the video[0].url field will be updated with the new design after applying the rule.

Example

Suppose you want to apply a design to improve the image_link in 1:1 format and also add the same design in 9:16 and 4:5 formats. You would follow these steps:

1. Select your design template: In the Design Template dropdown, choose the layout you want to apply. A preview is available to help you easily identify the correct design.

2. Select the number of pages to apply: In this case, select all three formats (1:1, 9:16 and 4:5).

After applying the Layout Product Data Rule, the selected fields will be automatically updated with your new design, ensuring your catalog images or videos are fully optimized for different ad placements.

Learn more about how to apply a layout.

Order of optimizations for the layout

The order in which optimizations are applied is critical because the tool executes them in a cascading sequence. This means that later optimizations can overwrite the results of earlier ones if they affect the same fields or product groups.

Best practices for optimization order

1. Start with general optimizations:
   Place optimizations that are generic or apply to the entire catalog or large sets of products at the beginning. These could include tasks like standardizing titles, adding static values or filling missing attributes.

2. End with specific optimizations:
   Optimizations that target specific products or fields should be placed later in the sequence. If two optimizations affect the same product group or field, the last one executed takes precedence, as it will overwrite the previous one.

3. Layout-specific considerations:

   • Any new fields created to be used in a layout must be optimized before the Layout rule is applied.

   • If the Layout optimization runs first, these new fields will not be incorporated into the final design, and the layout will not reflect the intended changes.