A Glossary stores the approved terms and phrases your organization wants translated the same way every time — product names, domain-specific vocabulary, and words that should never be translated at all. This article explains what a Glossary actually is, how it differs from a Translation Memory, the Terminology Board, and Learning Terms, and how Glossaries are scoped across your Organization, Organizational Units, and projects.
💡 Who is this for? This guide is for Account Admins and Project Managers who need to understand how Glossaries fit into wxrks's terminology system and connect them to the right Organizational Units and projects.
What is a Glossary?
A Glossary is a structured collection of concepts. A concept groups the same idea across every language it's defined in, so "customer" (EN) and "cliente" (PT-BR) sit under one concept as two terms. Each term can carry its own metadata: part of speech, whether it's case-sensitive or must be an exact match, whether it's forbidden (it should never appear in a translation), a definition, usage notes, and any custom attributes your organization has defined.
Terms reach a Glossary in four ways: added manually, imported from a .TBX or .xlsx file, requested and approved through the Terminology Board, or added automatically by Learning Terms — the next two sections cover how those last two actually work.
Glossary vs. Translation Memory
Glossaries and Translation Memories both help reuse prior work, but they operate at a different grain and enforce a different level of strictness:
Glossary | Translation Memory | |
Stores | Individual terms and phrases, grouped into concepts | Whole translated segments (source paired with target) |
Matching | Exact word/phrase matching, with rules like case-sensitivity and forbidden terms that must be honored | Fuzzy percentage matching against full segments |
Purpose | Enforce consistent terminology, everywhere that term appears | Reuse previously translated content to reduce cost and turnaround |
Both follow the same Organization → Organizational Unit → project scoping model — see the section below for how that works for Glossaries specifically.
The Terminology Board and Learning Terms
A Glossary is where approved terms live, but two separate mechanisms are what actually add new terms to it beyond typing them in by hand:
Terminology is an approval workflow: a kanban-style board where a proposed concept moves through review lanes before it's published into a Glossary. Use it when new terms should be checked by someone before they take effect.
Learning Terms is an automatic, AI-assisted way of populating a Glossary — it extracts likely key terms while a file is being parsed, and adds more as segments are translated with context-sensitive translation. Depending on how a project is configured, a learned term is added straight to the Glossary or routed through the Terminology Board for review first, the same as a manually requested one.
Whether Learning Terms is active for a given Glossary in a given project is its own setting — covered in Scoping below.
How Glossaries are scoped
A Glossary is created inside one Organization — required at creation, and it cannot be changed afterward, so choose the right parent Organization before creating one.
To actually be usable in a project, a Glossary then needs to be linked ("bound") to one or more Organizational Units, and from there to individual projects — the same three-level model Translation Memories use.
Organizational Unit level
Open the Unit's Context tab (Settings > Organizational Units > pick a unit) and use Add in the Glossary table to link any Glossary that belongs to the Unit's Organization. Marking a linked Glossary Default attaches it automatically to every new project created in that Unit.
That same table has an Enable Learning Terms toggle — whether this Glossary accepts automatically learned terms by default — and, when your account has the Augmented Term Lookup feature enabled, an Augmented Term Lookup toggle. See Augmented Term Lookup for what that changes in the Editor.
Project level
A project's own Context tab lists the Glossaries actually attached to it. Add Glossary links one manually — only Glossaries already linked to the project's Organizational Unit are selectable — or a Default Glossary attaches automatically when the project is created. Each Glossary linked here has its own Enable Learning Terms toggle and, if applicable, its own Augmented Term Lookup toggle, independent of the Organizational Unit's setting for that same Glossary.
See How to create a Glossary for the field-by-field walkthrough of creating a Glossary and linking it at both levels.
How Glossaries are used during translation and QA
While translating, any Glossary attached to a project surfaces matching terms in the Editor's sidebar as you work through a segment — exact matches appear right away, and inflected variants (a Glossary entry for "play" matching "playing" in the source) appear too once Augmented Term Lookup is enabled for that Glossary.
During quality checks, wxrks cross-references your project's Glossaries automatically: a term marked Forbidden is flagged if it shows up in a translation, and a valid Glossary synonym is recognized so it isn't reported as a false terminology mismatch.
Managing Glossary content
This article focuses on what a Glossary is and how it fits together with the rest of wxrks. For the step-by-step of actually working with one, see:
How to create a Glossary - creating a Glossary and linking it to an Organizational Unit and a project.
How to add new concepts to glossaries - adding and editing concepts and terms, from the Glossary page or the Editor.
How to export the Glossary - downloading a Glossary's content.
How to Format Your XLSX File for Glossary Import - preparing a spreadsheet for bulk import.
Related Articles
Terminology - the Terminology Board's request, review, and publish workflow in full.
Learning Terms - configuring automatic term extraction during parsing and translation.
Augmented Term Lookup - matching inflected forms of Glossary terms in the Editor.
All about Translation Memories - the segment-level counterpart to Glossaries.