Marketing Operations Practical insights
Keep campaign names and UTMs consistent with AI
Validate campaign identifiers generated by AI against a shared naming standard so analytics stay interpretable across channels, markets and creative variants.
AI-generated campaign names can look sensible while breaking reporting conventions. Small differences in capitalization, separators or source values can split one campaign into several rows. Generate identifiers from a defined structure and validate them before activation.
Define the fields people actually use
Agree the purpose of each campaign field and tracking parameter. Keep controlled values for channel and medium, and establish a stable campaign identifier. Document whether market, objective and creative variant belong in the name or in separate metadata. Avoid packing every possible attribute into an unreadable string.
Provide examples and validation rules
Supply accepted examples and common errors. A model can propose a readable label, while a deterministic check verifies required fields and allowed values. Adobe’s GenStudio concepts provide context for organizing campaign content and metadata. [1] The organization’s reporting definitions still need their own explicit standard.
Test the destination as well as the string
For an illustrative EN/NL launch, check that both localized links retain the campaign identifier and point to the correct language page. Open the final destination and inspect whether redirects preserve necessary tracking. Do not place personal information in URLs or confuse tracking parameters with canonical page addresses.
Handle exceptions deliberately
If a partner channel requires a different format, document the mapping to the internal campaign ID. Keep a register of approved exceptions and reconcile the first reporting data after launch. A naming standard is useful when marketers can apply it consistently and analysts can explain the resulting records without guessing which generated spelling refers to the same campaign.
Sources and evidence
Sources checked on 4 October 2026. Proposed workflows and hypothetical examples are editorial analysis.
From insight to practice