How do you manage multilingual content across ten languages, process over 6 million words annually, and still keep stress levels low? That’s a question Edwin Trebels of the Philadelphia Church of God seems to have mastered.
In a recent episode of The Agile Localization Podcast, Edwin joined host Stefan Huyghe to share how his team leverages automation, collaboration, and cutting-edge tools like knowledge graphs to streamline workflows and maintain a culture of transparency.
Listen to this episode of The Agile Localization Podcast on:
Streamlining Localization Through Automation
One of the key challenges in localization is managing repetitive, time-consuming tasks. Edwin identifies translation and editing work, along with communication and process management, as primary areas that can benefit from automation.
Automating processes such as content classification and workflow initiation allows his team to reduce the manual workload and ensure a smoother, faster localization process.
Nevertheless, Edwin is quick to point out the risks of over-automation. “You can also get into some system fragility where you have issues, especially when you’re dealing with various components. One might be updated and, all of a sudden, the whole process breaks because of that.”
To mitigate this, Edwin’s team builds notification systems via Slack to monitor workflows and troubleshoot issues in real time.
Localize your product with Crowdin
The Human Element in Automation
While automation is powerful, Edwin stresses the importance of preserving human oversight. Certain decisions, such as determining whether content should follow one workflow or another, still require human judgment.
“We are moving towards a world where more of that can be automated, where there is more further classification that can be automated through intelligent use of AI. But a lot of care is needed before fully automating these steps.”
The balance between automation and human input ensures that the workflow remains efficient without compromising on quality or flexibility. For Edwin’s team, the goal is not to replace human expertise but to enhance it, allowing translators and editors to focus on higher-value tasks.
Innovating with Knowledge Graphs
An exciting development in Edwin’s workflow is the use of knowledge graphs. These tools use semantic techniques to structure unstructured knowledge, enabling more accurate and context-aware machine translation.
“Instead of having all these documents and all sorts of content, you can create a structured knowledge graph of all of that utilizing several semantic techniques.”
For example, a knowledge graph can operationalize language across departments, supporting tasks from content creation to customer support. Moreover, by integrating knowledge graphs with AI-powered tools like AI proofreaders, Edwin’s team achieves a level of precision and efficiency that traditional workflows simply can’t match.
Maintaining Transparency and Accountability
Automation and innovation often come with the risk of losing visibility into workflows. To counter this, Edwin’s team prioritizes transparency through detailed visualization and process management. “We use templating in Trello, where a card moves semi-automatically across the process.”
For example, if one step in the workflow is slower than others, it becomes immediately evident, allowing the team to address the bottleneck proactively. Transparency is also crucial for managing the expectations of volunteer translators and editors. Edwin’s team maintains open communication with their volunteers, ensuring that deadlines are realistic and workloads manageable.
Lessons for Localization Teams
For organizations looking to optimize their localization workflows, Edwin offers this advice: start small and focus on incremental improvements. “Always look at your objectives first. What are you trying to accomplish?”
From there, break down the process into manageable steps and introduce changes gradually. Focusing on one improvement at a time helps teams to avoid overwhelming their resources and achieve sustainable progress.
Explore Localization Resources
Get industry insights, practical tips, and draw inspiration. We prepared webinars, e-books, and other content for you.
Check it Out
Conclusion
With advancements like knowledge graphs and AI agentic workflows on the horizon, the future of localization is undoubtedly exciting.
However, as Edwin’s experience shows, the key to success lies in thoughtful implementation and a deep understanding of both technology and human collaboration.
“Change is good, but it has to be managed wisely.”
Edwin’s Background
As the Director of LangOps at the Philadelphia Church of God, Edwin leads the Language Operations Department, overseeing localization from English into ten languages and providing a range of internal language services to support multilingual communication. With 20 years of experience in localization, Edwin’s leadership in 2023 enabled the team to translate and publish approximately 4.5 million words and interpret around 2 million words.
Listen to this episode of The Agile Localization Podcast on:
