CASE STUDY: Batch Record Translation Under Pressure

Translating registration documents for the world’s second-largest pharma company.

Translating registration documents for the world’s second-largest pharma company.

This case study highlights how an innovative approach to translation workflows and cutting-edge tech supported the delivery of a month’s worth of translation in just six days. 

Together, we achieved:

  • 200,000 words translated
  • 6-day turnaround time driven by TM & NMT
  • A follow-the-sun approach across Asia, Europe, Latin America
  • Automated workflows & live feedback environment

“The overall understanding of the market’s needs, especially deadlines related to national administrations, is excellent. Collaborating and partnering with Welocalize ensured matching our deadlines with the highest quality…” 

– José G. Martinez-Figueroa, Director of Regulatory Affairs – Latin America & Canada, Global Drug Development

Scroll 👇 to read the full case study. If you have any questions, reach out to us here.  


The Challenge

Brazil has a challenging and complex regulatory environment. Anvisa, the Brazilian Health Regulatory Agency, is among the strictest and most demanding in the world.

Since 2012, our client – one of the largest pharma companies in Brazil – has turned to Welocalize as its collaborating partner to translate registration documents, with over 500,000 words translated each month into ten languages.

In this case, the client required 200,000 words, including handwritten documents, to be translated from English to Portuguese-Brazil for the Brazilian market. 

This translation usually takes a month to complete. The client needed it in six days. 

The Solution

Agile Methodology

To deliver this work as quickly as possible, we adopted an Agile methodology, breaking the project into manageable chunks enabling our teams to work simultaneously.

Preparation was key. Our teams achieved terminology coherence through glossaries and automated QA checks. All content was pre-translated using a combination of translation memory (TM) and neural machine translation (NMT).

Then, we divided the content into batches, assigning them to teams working on them simultaneously. For instance, we worked with multiple formatting teams in parallel. We used native speakers to retype handwritten text. We assigned the content batches to a pool of dedicated linguists for post-editing and editing work with rolling deliveries.

Finally, instead of waiting for all the documents, we worked with several quality control (QC) specialists in parallel to prepare the final deliverables.

Tech Enabled

None of this could have been possible without our approach to technology and machine automation to create, annotate, and augment high-quality training data for machine learning (ML) and artificial intelligence (AI) initiatives.

We used innovative technology, including an automated central workflow system, for the client’s project in the Brazilian market. This innovation enabled our teams to streamline the translations. Close collaboration with our client allowed live feedback and immediate error correction.

We also used machine translation (MT) to speed up the process and global terminology libraries and TM to keep translations consistent across all content.

The Results

Ultimately, the client was able to submit batch records to Anvisa just six days later…

“The overall understanding of the market’s needs, especially deadlines related to national administrations, is excellent. Collaborating and partnering with Welocalize ensured matching our deadlines with the highest quality…” 

– José G. Martinez-Figueroa, Director of Regulatory Affairs – Latin America & Canada, Global Drug Development