Analysing terminology

Make sure you’re clear about your terminology before taking the next steps.

Why carry out a terminology analysis?

You have a terminology database – but are you sure it’s still doing what it’s supposed to? Many databases grow over the years: new entries are added, processes change, and service providers change. What remains is a database that promises more than it delivers, containing duplicates, missing usage information and terms that hardly anyone uses anymore.

A structured terminology analysis provides clarity. It shows which entries are actually being used and where there is room for improvement. This is because reliable terminology data reduces the amount of post-editing work required for translation output and is essential for AI systems to deliver accurate results.

Would you like to know where you stand with your terminology?

Arrange a consultation now

Typical problems that terminology analysis brings to light:

Duplicates and contradictory entries
Synonyms without usage information (preferred/permitted/prohibited)
Entries that are no longer in use
Comments and notes that should actually be separate terms
Isolated entries with no reference to the source language

Anyone who is aware of these weaknesses can take targeted action and ensure that further clean-ups, expansions or artificial intelligence integration are not undertaken on an uncertain or unnecessary basis.

What we analyse

Frequency of occurrence and active use

Not every entry in a terminology database is actually used. By carrying out a database analysis, we identify which terms appear regularly in your documents and which have been used very little or not at all since they were created. The result: a clear set of priorities that shows where it is worth investing effort.

Structural testing and quality assessment

Have the fields been set up and populated in the most useful way? What information is available on how to use it? Have synonyms been created correctly at term level? We check your database for adherence to terminological guidelines, a logical structure and form consistency, and document any areas where action is required.

Potential for cleaning up with oneCleanup

Based on the terminology analysis, we identify specific areas for improvement: Duplicates, orphaned entries or comments that should actually be terms in their own right. Our oneCleanup service combines automated checks with human expertise to highlight the potential for cleaning up your database.

The basis for machine translation and artificial intelligence

Terminology can significantly improve the quality of machine translation, but only if the underlying data is reliable. Analysis and clean-up are therefore also crucial steps prior to glossary creation for machine translation and AI systems.

This is how we analyse your terminology:

Initial consultation and assessment

We’ll get an overview of your situation: which tools you use, how your database is structured, and what you hope to achieve from the analysis.

Data preparation

We systematically process your terminology data, regardless of the file format or the terminology tool used.

Analysis of the existing data and structure

We identify which entries are actively used, where structural problems are present and where there is a need for clean-up. The result is a set of specific key performance indicators and a clear set of priorities.

Presentation of results

We will present the findings to you, explain the options available and recommend specific next steps, tailored to your resources and objectives.

Our services at a glance.

When is a terminology analysis useful?

An analysis is the best place to start if the database has been populated over the years by different people and nobody has a clear overview of it any more. Similarly, if a change of translation service provider or CAT tool is imminent and the database needs to be cleaned up for this purpose. It is also a necessary step prior to introducing machine translation or AI-supported processes. Reliable terminology can only be based on verified data. The same applies if a consistently high volume of corrections to translation output suggests there are terminology issues.

FAQs

What is the difference between terminology analysis and terminology clean-up?

Analysis is the first step: it highlights where problems lie, how actively individual entries are being used, and where action is needed. The clean-up process then begins and specifically addresses the identified vulnerabilities. Both can be commissioned together.

Which formats and tools are supported?

We operate independently of any particular manufacturer and support standard terminology tools and export formats. We will discuss what information you can provide us with during our initial consultation.

What happens after the analysis?

You will receive a structured evaluation with specific recommendations for action. On this basis, you can decide which next steps you wish to take – whether to clean up the terminology, expand it, or prepare it for machine translation and artificial intelligence.