How to Exclude Specific Product Fields Like SKU or GTIN from Automatic Translation

To keep product identifiers like SKUs, GTINs, or model numbers identical across languages, you need to prevent the translation system from altering them. This is done by setting up exclusion rules that tell the translation agent which fields or patterns to skip.

Prerequisites for Setting Up Field Exclusions

Before configuring exclusions, ensure you have:

• Access to your SeaText Translation Agent settings
• Administrator privileges in your ecommerce platform integration
• A clear list of product fields that must remain untranslated (e.g., SKU, GTIN, MPN, ASIN)
• Knowledge of your product data structure (field names as they appear in your feed or CMS)

Step-by-Step Process to Exclude Specific Fields

1. Log in to your SeaText dashboard and navigate to the Translation Agent settings.
2. Under Translation Rules, select Field Exclusions or Do Not Translate These Fields.
3. Enter the exact field names you want to protect, such as sku, gtin, model_number, or mpn. Use the field names as they exist in your product data source.
4. For pattern-based exclusions (e.g., all fields ending in _code or values matching a barcode format), click Add Regex Rule and enter a valid regular expression, such as:
• ^\d{8,14}$ to match GTIN-8, GTIN-12, GTIN-13, or GTIN-14
• ^SKU-[A-Z0-9]+$ for custom SKU formats
5. Save the rule and apply it to your active translation jobs.
6. Run a test translation on a small product batch to verify that the excluded fields remain unchanged.
7. Review the translated output in all target languages to confirm identifiers are preserved.
8. If any field is still being translated, double-check the field name spelling and regex syntax, then reapply the rule.

Technical Implementation Details for Developers

Understanding how regex patterns interact with different data feeds is crucial for reliable implementation. The behavior differs between JSON feeds and XML sitemaps due to structural variations.

In JSON feeds, field names are explicit keys within objects. For example, {"sku": "ABC123"}. Regex patterns here match against the key name directly. If you define an exclusion for sku, the engine checks the string "sku" exactly. Case sensitivity matters here. If your feed uses Sku instead of sku, a rule for sku will fail unless the system supports case-insensitive matching or you adjust the rule.

XML sitemaps use tags and attributes. A product might be represented as <product><sku>ABC123</sku></product>. Here, the regex must account for the tag structure. Some systems allow excluding based on the tag name (sku), while others require excluding the content inside the tag. If the translation engine treats the entire XML node as translatable text, excluding just the tag name might not stop the content from being processed if the content itself looks like translatable text.

When using API calls to trigger translations, ensure that exclusion rules are passed in the request payload. If the API does not support rule inheritance from the dashboard, you must explicitly include the exclusion logic in every API call. Failure to do so results in standard translation behavior, potentially corrupting identifiers.

Common Pitfalls in Field Mapping

Even with correct rules, mapping issues can cause unexpected translations. Two common pitfalls are case sensitivity and whitespace handling.

Case Sensitivity: Many systems treat SKU and sku as different fields. If your data source uses uppercase headers but your exclusion rule uses lowercase, the rule will not match. Always verify the exact casing in your source data. If unsure, use regex patterns that ignore case, such as (?i)^sku$, if supported.

Whitespace Handling: Hidden spaces can break exact matches. A field named sku (with a trailing space) is different from sku. Trim whitespace in your data before applying rules. Alternatively, use regex patterns that allow optional whitespace, such as ^sku\s*$.

Nested Structures: In complex JSON, fields may be nested. For example, {"variants": [{"sku": "A1"}]}. Simple field name matching might only look at top-level keys. You may need to use dot notation (e.g., variants.sku) or recursive regex patterns to reach nested fields. Check your platform’s documentation for supported path syntax.

How Field Exclusion Works in SeaText

SeaText’s Translation Agent scans your product data before translation. When a field matches an exclusion rule — either by exact name or regex pattern — it is copied directly to the target language without modification. This process happens in real time and applies to all 125 supported languages.

The system does not translate, truncate, or re-encode excluded fields. Instead, it treats them as protected metadata, ensuring consistency in inventory feeds, marketplaces, and downstream systems that rely on exact identifier matching.

Key Facts About SeaText’s Field Exclusion Feature

Feature	Detail
Supported exclusion methods	Exact field name matching and regex patterns
Applies to	Product fields, meta tags, custom attributes, and structured data
Languages covered	All 125 languages supported by SeaText Translation Agent
Real-time application	Yes — exclusions are applied during live translation jobs
Reversibility	Rules can be edited, disabled, or removed at any time

Why This Matters: Risks of Not Excluding Identifier Fields

If SKUs, GTINs, or similar codes are accidentally translated:

• Marketplaces like Amazon or eBay may reject product listings due to invalid or duplicate identifiers
• Inventory sync systems can fail, leading to overselling or stock discrepancies
• Barcode scanning in warehouses or retail stores may fail if the translated code doesn’t match the physical label
• Customer trust erodes when product numbers don’t match across regions or documentation

These issues can trigger compliance problems, increase return rates, and disrupt automated fulfillment workflows.

Options and Trade-Offs: Field Exclusion vs. Whole-Product Exclusion

You might consider excluding entire products instead of individual fields. However, field-level exclusion offers better control:

Approach	Best For	Trade-Off
Exclude specific fields (e.g., SKU, GTIN)	When only identifiers need protection but descriptions, titles, and benefits should be translated	Requires accurate field mapping; regex needs testing
Exclude entire products	When a product should not appear in certain languages at all (e.g., region-locked items)	Overly broad — prevents translation of all content, even translatable fields

For most ecommerce use cases, field-level exclusion is the preferred method because it preserves translation benefits while protecting critical data.

Practical Scenarios Where Field Exclusion Is Essential

• Multilingual Amazon sellers: GTINs must match the global product catalog; any translation causes listing suppression.
• Shopify stores using Google Shopping: Feed validation fails if gtin or mpn fields are altered. Shopify exports often use specific field names like variant_barcode. Ensure your exclusion rule matches this exact name.
• WooCommerce integrations: WooCommerce uses meta keys like _sku and _gtin. The underscore prefix is part of the field name. Your exclusion rule must include the underscore, e.g., _sku, otherwise the translation engine may process the value.
• Magento setups: Magento attributes often have unique codes. An attribute labeled "SKU" might have a backend code like sku or custom_sku. Check your attribute configuration to find the exact code used in the data export.
• Enterprise ERP integrations: Systems like SAP or Oracle rely on exact SKU matches for order processing and invoicing.
• Marketplaces with strict feed requirements: Platforms such as Walmart or Cdiscount validate identifiers and reject feeds with modified codes.

Limitations and When This Advice Does Not Apply

Field exclusion may not work as expected if:

• Your product data uses nested or dynamically generated field names that aren’t exposed in the translation interface
• Identifiers are embedded inside translatable content (e.g., “Model XYZ-123 is now available”) — in such cases, consider using inline protection tags or restructuring content
• You’re using a custom data feed without standardized field naming — verify field names with your developer or platform documentation
• The translation job is running via API without rule inheritance — ensure exclusion rules are passed in the request payload.

In these cases, consult your integration specialist or SeaText support to validate rule application.

Terminology Clarification

SKU (Stock Keeping Unit)

A unique identifier assigned by a retailer to track inventory internally.

GTIN (Global Trade Item Number)

A globally unique product identifier, including UPC, EAN, and ISBN formats.

Regex (Regular Expression)

A pattern-matching language used to define flexible exclusion rules for variable identifiers.

Field Exclusion

A setting that prevents specific data fields from being processed by the translation engine.

FAQ: Next-Level Questions About Field Exclusion

• Can I exclude fields based on their value rather than their name? Yes — use regex rules to target values matching specific patterns (e.g., all 12-digit numbers) regardless of field name.
• Will excluding a field affect SEO or search visibility? No — excluded fields like SKUs are not typically indexed for search, so preserving them has no negative SEO impact.
• Can I apply different exclusion rules per language? No — exclusion rules apply universally across all target languages to ensure global consistency.
• What happens if I make a mistake in my regex? The rule may fail to match or over-exclude content. Always test rules on a sample batch before full deployment.
• Is there a limit to how many fields I can exclude? No practical limit — you can exclude as many fields or patterns as needed.
• Do I need to reapply rules after updating my product feed? No — rules are based on field names or patterns, so they persist through feed updates unless the underlying structure changes.
• Can I exclude HTML attributes like data-sku or itemprop="gtin"? Yes — if these attributes are part of the translatable HTML, you can exclude them using field names or regex patterns that match the attribute context.

Further reading and comparison sources

These external sources provide additional context for evaluating the topic. Their inclusion is not an endorsement.

• Automatic translation - WPML
• I want to use automatic translation but not for products. Products ...
• How to exclude content from being translated in my store : Transtore ...

Learn more

Visit the website for more information.