Embracing an agile localization approach

Software localization teams face increasing pressure with agile development processes. In a continuous release cycle, new software features can be released daily, sometimes even multiple times per day. However, traditional localization workflows struggle to keep up with this accelerated pace if the goal is to only ship localized versions that are 100% reviewed and approved. A comprehensive localization process simply takes time. Therefore, the dilemma arises: either compromise on quality standards or find a way to give more time for localization without delaying new releases. But how?

Traditional flows for translating and reviewing software

Traditionally the translation and review process begin after the first development release. That is when new features were implemented, the string list is completed, and the source files can be sent to the translators. Unfortunately, this often results in an agile development sprint followed by a sequential waterfall localization process. Most conventional computer-assisted translation (CAT) tools require a complete import and alignment of the updated source file, using translation memory to ensure previously translated strings keep their correct status. Subsequently, translation tasks are assigned to internal or external translators. Once the translations are complete, in-country reviewers conduct a thorough review of the work. Any reported issues are processed by the localization team, and a set of validated target files can finally be returned to the developers for compilation into a new application build.

Essentially, this approach transforms the entire process into a waterfall model, as the production release must wait until the translation and review stages are finalized, often necessitating intermediate application builds. Although in-country reviews could be done without an intermediate build by examining language pairs, having the translations in the context of the actual user interface significantly improves the ability to make correct judgments. You can imagine that the initial speed of the agile development sprints will be slowed down if we adhere to this traditional flow.

Embracing an agile localization approach

What if we could expedite this process without compromising quality? How can we establish a seamless environment where localization and development can collaborate on the same project, independent of each other? Envision a scenario where localization can embrace an agile approach, enabling a continuous localization flow. However, achieving this reality requires fulfilling several critical conditions. First and foremost, the strings should be readily accessible for localization as soon as they are created. Secondly, it is essential to provide clear context for each string, avoiding the inundation of development teams with numerous queries. By providing comprehensive context, localization teams can accurately understand the purpose and usage of each string, enabling them to produce precise translations without constant back-and-forth communication.

This is precisely where Rigi.io comes into play. Rigi.io empowers app teams to create parallel sprints for both development and localization. By doing so, it facilitates a continuous localization and integration flow, significantly reducing time to market. At the heart of Rigi.io lies its remarkable ability to generate interactive previews for web and mobile applications. These previews seamlessly integrate with most conventional CAT tools, ensuring compatibility and ease of use. The primary objective of Rigi.io is to enhance translation quality and speed by minimizing quality-related issues, associated delays, claims, communications, discussions, and research. Additionally, it promises to reduce the effort and time required by development and localization teams to initiate and manage translation sprints.

Rigi.io achieves this by decoupling localization from development through the creation of dynamic previews for your software application. These previews serve as guidelines for translators while also providing visual feedback for in-country reviewers. Simultaneously, Rigi.io automates the exchange of language updates. Combined with HTML previews, this results in an interactive mockup that faithfully resembles the software application. The process can run continuously, ensuring both parties are consistently updated with the latest information. The image below depicts this entirely new process, benefiting all stakeholders involved:

Push, parse, preview, repeat

To support this process, Rigi.io utilizes an artificial language. Consequently, Rigi must parse every resource file, which is an automated process. As such, whenever a developer pushes new code to the feature branch, the related source strings become immediately available to the localization team. At this point, HTML previews can be generated from the application in development, granting the localization team complete independence to execute the entire localization roundtrip until review and validation. All of this is bolstered by the interactive visual context of the application. Repeat this flow every hour of the day by exporting delta strings during feature branch updates, capturing previews, localizing in context, and returning everything to the developers on time, eliminating any delays in release due to missing or incorrect translations.

Benefits of automation and visual context

While Rigi.io automates numerous processes and proves to be a powerful tool for everyone involved, it is crucial to acknowledge that dynamic previews are essential to ensure the continuous localization flow yields high-quality translations. With visual context in place, the following advantages are realized:

• There is no longer a need to inquire about the location of a string from developers since everyone can see the string in its context.
• Intermediate builds are unnecessary to showcase translations within the application because the previews are dynamic, ID-based HTML files. This means that every target language can be displayed in the same preview instantly.
• Translators and reviewers no longer require extensive onboarding to understand the application. They can simply refer to dynamic previews for translation or testing. Furthermore, because the previews are not only dynamic but also intelligent, Rigi.io precisely identifies the screens where a string is used and presents them alongside the selected string within the preferred CAT tool’s translation editor.

This streamlined flow benefits everyone involved. Development teams can concentrate on completing their development sprints without having to address linguistic inquiries. For instance, Degreed reported a reduction in question response time from two days per release to only two questions after adopting Rigi.io. Another Rigi.io subscriber experienced an 87% decrease in localization-related development efforts. Simultaneously, localization teams can keep pace with development. LexisNexis, one of our customers, enjoys a predictable and efficient process with guaranteed translation quality. With Rigi.io, the traditional waterfall approach transforms into a reliable continuous localization flow.