Research report · The Edit Distance Report
Where your reviewers’ time matters most, in English and French
We measured 27 million sentences, including 760,000 from English into Canadian French.
- Free, no form
- 14 sections
- Report in English
- PDF, 0.8 MB
Download the report (PDF)Read it online
Key findings
| Measure | Result |
|---|---|
| Plain machine translation, share of text changed | 15.5% |
| Context-sensitive AI suggestions, share changed | 8.1% |
| AI suggestions that go out unchanged | 84 in 100 |
| Sentences review leaves as they are | 98 in 100 |
| Editing, month one to month ten | 42% to 16% |
Context-sensitive suggestions draw on the customer’s translation memory, glossary and past corrections.
English to Canadian French
24%of the text changed
People change about 24% of the text from English into Canadian French, more than the 17% average. French is where your reviewers’ judgment does the most work.
- Spanish (United States)9%
- Latin American Spanish13%
- Brazilian Portuguese14%
- Korean (South Korea)18%
- Spanish19%
- Japanese (Japan)20%
- Chinese (China)20%
- European Spanish21%
- Dutch (Netherlands)21%
- French (France)22%
- German (Germany)22%
- Canadian French24%
- Brazilian Portuguese → American English25%
- Italian (Italy)26%
The person who runs localization
Your role matters more than ever
Editing a suggestion is only 7% faster per word. The time comes back in the reading, so the work moves to deciding what gets read, reviewed and kept. Whoever runs that flow decides how much time AI saves.
You run the flow
- 1SuggestAI suggests from your memory and glossary. 84 of 100 go out as written.
- 2ReadYou choose what gets a full read. Smells flags where meaning may have drifted.
- 3ReviewYou choose which passes run. Keep them for legal, regulatory and public-facing French.
- 4KeepYou decide what the memory and glossary keep. As they fill up, editing falls.
Editing falls as the memory fills up
How it was measured, and your data
- Sample
- 27,040,017 sentences on the wxrks platform, August 2024 to September 2026.
- Change
- Words replaced, added, deleted or moved before a person confirmed the sentence.
- Privacy
- Group numbers only, each from at least five accounts. No client named, no text read.
- Your translations
- The learning stays in each customer’s memory and glossary, never in an AI model.
What the data can’t tell you
- It counts changes, not quality. Quality is the next study.
- It is production data, not a lab test.
- The learning curve follows accounts that stayed.
Read the full report
Date modified