Case Study Migros Bank

Transparent Handling of Translation Projects with Across

During the introduction of Across, we were very satisfied with the care and support provided. The members of the Across team were always eager to deliver solutions, which they then successfully rolled out.

Beatriz Gonzalez, Senior Business Analyst at Migros Bank AG

Migros Bank, headquartered in Zurich, Switzerland, is a wholly-owned subsidiary of the Federation of Migros Cooperatives. The Swiss universal bank focuses on domestic private and SME clients. Established in 1958 by Gottlieb Duttweiler, the bank rapidly grew thanks to savings accounts with attractive interest rates and inexpensive mortgages. With more than 800,000 client relationships, Migros Bank has become one of Switzerland's leading banks. The internal and external communication takes place in the three national languages German, French, and Italian. Various content types, such as marketing texts, client letters, Internet and intranet texts, internal instructions, processes, etc., are translated into these languages. Most of the source texts are composed in German and mainly need to be translated into French and Italian. This task is handled by four team members, a project coordinator, and several language providers. The yearly translation volume amounts to approximately 5,800 A4 pages. To deliver the translations efficiently, the newly formed internal translation team was in need of a CAT tool. Migros Bank especially wanted to facilitate the placement of orders and enable transparent project management. However, as the bank lacked expertise in terms of the needed software capabilities, the project managers asked the language technology provider TextShuttle for assistance. Together, the requirements for the future solution were defined. The solution was to include reporting functionality and be equipped with a language portal that would enable the team members to directly place translation orders. Another requirement was a central translation memory to which existing translation memories of external translation service providers could be imported. Future translations, too, were to be integrated in this translation memory, regardless of whether the respective translations would be prepared directly in the CAT tool and could thus be adopted automatically or be created by external service providers. Moreover, the bank wanted to be able to integrate machine translation systems.

Compelling Features

The elaborated proof of concept (PoC) pointed to the Across Language Server as the solution that would best meet the requirements of Migros Bank. The bank was especially impressed by the order overview and management as well as the preview function. Apart from the solution, the bank was also pleased with the manufacturer's cooperation. The professional, timely care provided by Across Systems GmbH during the PoC was very helpful. In response to questions, the software manufacturer promptly provided plausible examples or parameterization adjustment proposals. In this connection, it was helpful that all contacts worked at the Across headquarters and were thus able to discuss the various issues directly and speedily with the decision-makers of Migros Bank. Another aspect that ultimately contributed to the decision in favor of Across was that the Active Directories can be connected to the order portal, enabling the client information to be sent to the tool.

Rapid Implementation

The rollout until the go-live in September 2017 merely took six months. This was mainly because the Across Language Server version that had been installed for the evaluation could already be populated with the existing translation memory during the test phase. Subsequently, the bank merely had to make some minor adjustments. Shortly after the introduction of the Across Language Server, Migros Bank additionally connected to neural machine translation systems. These systems have been in operation since mid-2018.

Transparent Overview of the Project Status

Today, Migros Bank uses the Across Language Server for the entire translation process from the order placement in the Language Portal to the translation to the reporting. The members of the internal translation team translate directly in Across. External service providers receive a package containing the order data, the translation memory hits, and the results from the machine translation systems from Across. Almost all orders are placed via the Language Portal, which is integrated in Across. In the order management, the translation team always has an overview of the current orders. For example, they can see the status of the individual orders and where they are currently being processed. Moreover, the order management provides an overview of the workload of the team members involved.

Get Case Study by E-mail

By submitting your data, you consent to our processing the data in the scope specified in the Privacy Policy for the purpose of handling your request and to our contacting you via the same communication route. Moreover, we will process the data on the basis of our legitimate interests. We will delete the data as soon as your request has obviously been completed or if you effectively object to the further processing of your data by us. For more information on this subject, please refer to our Privacy Policy.

Migros Bank AG

  • A leading Swiss bank
  • Founded in 1958
  • 1,300 full-time employees
  • Around 70 branches in Switzerland
  • Balance-sheet total: CHF 44.7 billion (as at December 31, 2018)

Initial Situation

  • No central translation memory
  • Translation volume: 5,800 A4 pages/ year
  • Translations mostly from DE to FR and IT
  • Plans to connect neural machine translation systems

Across components employed

  • Across Language Server
  • Across Language Portal
  • Across Data Cube
  • crossTerm Now


  • Placement of 99 percent of the translation orders via the Across Language Portal
  • Transparent project management and status overview
  • Overview of the workload of the parties involved in the process
  • Efficient translation process thanks to the use of translation memories and neural translation systems