HOME → TQE
Assess translation quality with a unified method for human and machine translations. Ensure your content meets standards across all languages and vendors.
Send your translations for evaluation—full text or a randomized sample.
We provide a TQE template, customizable if needed (error types, penalty points).
Our linguists assess issues and calculate an overall quality score.
QA specialists re-check the scoresheet for compliance.
You get a final scoresheet with described errors and overall score.
| TQE | Editing | LQA | |
|---|---|---|---|
| Focus | Classify & measure errors | Silently fix errors | Test product + technical issues |
| Scope | Text samples or full text | Full text only | Product-wide testing |
| Output | Detailed scoresheet with metrics | Corrected text + general feedback | Bug reports + technical fixes |
| Best for | Quality benchmarking, vendor tracking | Final polish before release | End-product validation |
Editing focuses on fixing errors in the full text. TQE focuses on classifying and measuring errors, and can be done on samples—making it more cost-efficient for large projects.
Yes, TQE is applicable to both human and machine translations, including AI/MT engine evaluation.
We use the MQM framework and comply with ISO 5060:2024 for unified yet flexible evaluation.
No, TQE can be performed on a randomized sample, which is especially cost-effective for large projects.
A final scoresheet with described errors, error classifications, and an overall quality score. Corrections are also provided where applicable.