Terminology must first be consistent in the source language

What it is

Source-language terminology consistency gives a product concept a stable name before content enters translation: one term maps to one defined concept, while synonyms, abbreviations, and legacy names appear only for an intentional audience or state distinction. This is upstream control of multilingual production, not merely consistent target-language wording. If source copy uses “workspace,” “space,” and “team area” for one object, each language can amplify that naming drift into several translations.

Why it happens

Translation memories retrieve prior segments by similarity, termbases recognize terms and supply concept metadata, and machine translation models infer output. Their mechanisms differ, but without a product concept identifier or sufficient context none can reliably recover the team's internal concept identity. Several source terms for one concept reduce TM leverage, activate different term candidates, or produce divergent translator judgments. One source term for several concepts makes mechanical unification equally unsafe. As language count grows, each source variant can form another target-term branch, multiplying naming drift across localized versions.

Where it stops holding

Consistency does not require identical wording everywhere. A constrained button may use an approved short label, and novice and expert content may explain a concept differently, provided the variants are recorded against the same concept. Distinct states, permissions, or objects must not be collapsed for superficial uniformity. Deliberate campaign language and necessary pronoun use are not automatically terminology failures. Nor should a source termbase force every language into the same grammatical form: inflection, order, gender, and regional usage remain locale-specific.

Applying it

• Lint source terminology before translation handoff. When a synonym appears, determine whether it is accidental drift or evidence of a genuinely different concept.
• Pass the concept identifier, short definition, preferred source term, approved variants, and deprecated names with each translation unit so tools and translators can distinguish homonymous product meanings.
• On a source-term rename, mark or invalidate affected TM candidates and state the legacy name, effective version, and migration scope so high-similarity matching does not reintroduce the old translation.
• Sample target-term branches by concept across locales and locate cases where one source concept has split into unnecessary target variants in a market.

Related

• Same group: S1.07.1 Avoid idioms, puns, and culture-bound jokes · S1.07.2 Simple sentence structure can substantially reduce ambiguity · S1.07.4 Avoid embedding text in images
• Adjacent: S1.05 Translation context and string reuse · T1.04 Terminology consistency and glossaries
• Search terms: source terminology · concept-term mapping · terminology governance