Introduction #️⃣

Use the glossary feature in Localazy to keep your translations correct and consistent. Add key terms, product names, or phrases that should alwways be translated the same way or rather not translated at all.

Translators and some machine translation providers (including Localazy AI) will be instructed to follow your rules, so you don’t get mixed terminology across your texts.

Incorrectly used glossary terms are also flagged in the validation checks to help you spot errors across your translations. Set it up once before you start translating, and you’ll save time searching for inconsistent wording later.

Glossary languages #️⃣

Source language for your glossary must match your project’s source language. The glossary is currently not automatically migrated when a project’s source language is changed.

The target languages used in your glossary are independent of your project’s languages and you can add/remove the languages at any time.

When adding target languages to your glossary, you can select all languages that are already in your project or choose completely new ones. Languages that are currently present in your project and in the glossary are highlighted in green.

Removing a target language from the glossary keeps its translations. They’ll be restored if you add it back later.

Glossary terms attributes #️⃣

The terms you add to your glossary will be matched and translated (or not) according to your preferences.

• Term in source language:
  • The word or phrase that will become the glossary term in the project’s source language
• Description:
  • Use this to provide additional context, is it a verb or noun? What does it mean? If it’s an acronym, write what it stands for etc.
• Translations:
  • Define translations of your term in other languages, if the term is translatable.

There are also three switches that specify the glossary term behavior:

• Case sensitive:
  • Enable to only match terms with the same casing, for example the term Apple would not be matched if only the word apple appears in the source text.
• Match only whole words:
  • Enable this when you want only the whole word or full phrase to match. For example, the glossary term “book” won’t be highlighted inside the word “booking” when it appears in the source text if this is enabled.
• Translate to other languages:
  • Disable this if you want to keep the term as-is in the source language. For example, brand names using common words such as Apple should not be translated and can be protected by the glossary.

Creating glossary terms #️⃣

Navigate to the Context ➡️ Glossary screen in your project and click on “+ Add glossary” button.

You will be able to set up all glossary attributes as needed on the screen.

Importing glossary terms #️⃣

You can import your glossary terms and their translations in one of the supported formats (.csv, .xls, or .xlsx)

To prepare your table for proper parsing, you need to follow a few rules:

• Rows are the individual glossary terms, columns are either a language or a term attribute (see below)
• A header row is mandatory, as it is used for proper column mapping of values
• Up to 1000 terms can be imported (this is the maximum number of glossary terms in any given project)
• Maximum file size cannot exceed 10MB
• The file must contain a header row and at least one data row item, otherwise, it is considered empty

Example of a well-prepared file (source language: en):

Batch import is currently allowed only when the glossary is empty. To re-import, existing terms must be deleted first.

Column & value mapping #️⃣

Each column is mapped to one of the following:

• A language
• An attribute:
  • Term description (up to 2000 characters)
  • Case sensitive (true/false)
  • Whole word only (true/false)
  • Translatable term (true/false)
• Or marked as skipped (Do not import).

Your file will be analyzed and if it’s prepared correctly, no manual column and value mapping will be necessary. If some headers or values are not recognized automatically by the system you must assign them. You can also override the automated mapping before import if something is mismatched.

• Languages match by locale code, BCP-47 tag, English name, or localized name. Case-, whitespace-, hyphen-, and underscore-insensitive (e.g. en-US, en_us, English, english all match).
• Attributes match by keyword in the header:
  • desc* → description,
  • case* → case sensitive,
  • whole* / exact* → whole word only,
  • trans* → translatable.
• Unrecognized headers remain unmapped and must be assigned manually.
• True/false (boolean) values are also auto-resolved:
  • True: true, yes, 1, on, match
  • False: false, no, 0, off, nope

Where is glossary used? #️⃣

The glossary is available when translating strings or reviewing translations in the UI. Matched terms are highlighted and validated for every translation.

Glossary terms are used as context when generating translations with Localazy AI and other machine translation providers that support it.

You can also share glossaries across projects with Connected Projects feature.