Ali Fakhar
فارسیGet in touch

How to translate and localise marketing content with AI

  • Level: Intermediate
  • A few days to set up

Ali Fakhar · Updated: Sept 2026

The short answer

Build a glossary and style notes for the target market first, translate with a model that is given both, and ask it to flag idioms, claims and cultural references that need a local decision. A native-speaking reviewer edits for meaning and tone, and their corrections go back into the glossary.

On this page

At a glance

The marketing problem
Direct translation keeps the words and loses the message: idioms, examples, units and claims often do not travel. AI makes a first draft fast, but without a glossary and a native reviewer the brand sounds different in every language.
Who this is for
Marketing teams publishing in more than one language or entering a new market.

What you need

Before you start

  • A native speaker of the target language who knows the market and can review
  • A glossary: product names, terms you never translate, preferred translations
  • Agreement on which content is translated and which is written locally

What goes in

  • The source content, final and approved
  • The glossary and style notes for the target language
  • Local examples: competitor pages or your past content in that language

How it works

  1. 01 · You decideBuild the glossary and style notes
  2. 02 · AI stepTranslate with the glossary
  3. 03 · AI stepFlag what needs a local decision
  4. 04 · You decideNative review
  5. 05 · OutputPublish and update the glossary
  • AI step
  • You decide
  • Output

Step by step

  1. 01 You decide

    Build the glossary and style notes

    List product names, terms that stay in the source language, preferred translations, the level of formality and anything the market is sensitive to. The reviewer owns this document.

  2. 02 AI step

    Translate with the glossary

    Give the model the glossary with every request, so terms stay consistent across pages.

    Prompt to copy

    Translate the text below from [source language] into [target language] for [audience in market]. Follow the glossary exactly and keep the level of formality described in the style notes. Keep product names and terms marked “do not translate” in their original form. Glossary and style notes: [paste] Text: [paste]

  3. 03 AI step

    Flag what needs a local decision

    Ask the model to list what may not travel, before anyone polishes the wording.

    Prompt to copy

    Compare the source text and your translation. List every idiom, example, unit, date format, legal or pricing claim and cultural reference that may not work in [market]. For each, explain why and suggest one or two local alternatives. Do not change the translation yet.

  4. 04 You decide

    Native review

    The reviewer decides each flagged item, edits for meaning and tone, and checks the layout, especially for right-to-left languages.

  5. 05 Output

    Publish and update the glossary

    Publish, then add the reviewer’s recurring corrections to the glossary, so the next translation starts better.

Where a person decides

A native-speaking reviewer approves the meaning, tone and every claim before publication.

What to watch for

  • Models are weaker in some languages and dialects than in English.
  • Legal, pricing and regulated claims need local checking, not translation.
  • Right-to-left languages also need layout checks, not only text.

How to tell it is working

  • The reviewer’s edits shrink over time as the glossary grows.
  • Local readers do not flag the text as translated.
  • Local performance is compared with local benchmarks, not with the source market.

An illustrative example

Input

Illustrative: an English landing page for a fictional HR tool, to be published in Persian for teams in Iran and the diaspora.

Output

A Persian draft that keeps the product name in English, replaces a baseball idiom with a local expression, flags a pricing line for local review and checks the right-to-left layout.

Want this running in your team?

AI marketing advisory. An experienced marketing partner for the decisions your team needs to make.

View service
All AI in marketing workflows ↗
Analytics preferences

Current preference: off