ModalB

Case study

Business software for managing translation workflows

A single workspace for running the translation of content into dozens of languages: importing text, allocating work, tracking progress, delivery and invoicing.

Client
Localisation company
Sector
Translation and localisation

The context

Translating a game, an application or a catalogue into fifteen languages means thousands of sentences, dozens of contributors and an endless back-and-forth of files.

We designed and built, end to end, a web application that replaces all of that with a single workspace.

What was going wrong

The difficulty is not the translation itself, but everything around it.

Volume. A project means thousands of sentences to track, language by language. A spreadsheet quickly reaches its limit.

Coordination. Translators and reviewers work on the same content at the same time, with the constant risk of overwriting each other's work.

Files going back and forth. Every exchange is an opportunity to lose a version, forget a language, or reintroduce a correction that had already been handled.

Invoicing. Each supplier's word counts were recalculated by hand at the end of the month, from scattered sources.

Who uses it

Four roles, each in a space that shows only what concerns them: the administrator, the project manager who allocates the work, the client who follows progress and collects their files, and the translators and reviewers, who only see their own assignments.

What the platform does

Text is gathered into a clear grid: one row per sentence, one column per language. Existing files are imported in a few clicks, whatever their format.

The project manager then allocates work between translators and reviewers, and follows progress in real time: who is translating what, what is finished, what is still to be approved. Two people cannot modify the same sentence without knowing it.

Once translations are approved, the platform regenerates the files automatically in the expected formats and delivers them to the client. It also computes word counts and prepares the cost summaries and supplier invoices.

Translator queries, change history, notifications: everything is centralised and traced in one place.

What it changes

File exchanges between contributors disappear from the process. There is a single reference text, and everyone works on it within the limits of their role — which removes diverging versions and lost corrections at a stroke.

A project's progress is read directly in the platform instead of being reconstructed. And because word counts are computed as the work goes along, cost summaries and supplier invoices are produced from production data rather than re-entered at the end of the month.

Technologies

  • Java 21 / Spring Boot
  • React / TypeScript
  • PostgreSQL
  • Temporal
  • Redis
  • MinIO
  • Docker
  • REST APIs

Expertise involved

Other case studies

Working on something similar?

Our teams can support you on scoping, architecture and delivery.