About quality models

You can use quality models in multiple ways, depending on how your company approaches the review process.

In the most common use case for quality models, a language service provider (LSP) or a company receives a finished translation and decides to perform a quality audit on it. This involves taking random samples from the translated assets and sending them out to a separate reviewer. This reviewer is usually a subject matter expert. The reviewer flags segments with errors and returns the asset to the LSP. The LSP then scores the asset based on the types of errors in it and decides whether to keep the translation or send it back to the translator.

A second use case is one where reviewers use the error types and severities during the regular review process to provide more specific information about why they rejected certain segments. In this use case, reviewers have more information they can use when rejecting segments.