Localization breaks most engineering teams at scale, especially when it’s bolted on after the fact. We design localization architecture, translation pipelines, and internationalization systems that integrate directly into your codebase and CI/CD workflows.
Common Traps
Teams often choose a translation management system (TMS) for a proof of concept and later find out that it does not integrate with other platforms such as CMS, customer success or lifecycle communications. This results in costly and time consuming rework to migrate onto a consolidated platform.
Even worse, some companies build a DIY translation management system thinking “it’s just a database of translations”, only to discover how complicated these systems are. There are a lot of edge cases in localization and internationalization.
Weak integrations with the localization pipeline create bottlenecks when translating content at scale into many languages. Shipping cadence drops and localized versions of the product are delayed or look lazily translated.
Cultural assumptions are often baked into the product itself, which sets it up to fail in other regions and languages, no matter how well it is translated. This is especially common with US based startups that assume everyone speaks English or shares the same habits and customs.
Poor or inconsistent executive sponsorship and cross-function coordination results in localization and internationalization being under-resourced, which creates a self-fulfilling prophecy of mediocre outcomes.
Who We Are & What We Do
Localization Technology Partners was founded by Brian McConnell, a 20 year localization and startup veteran. He built and scaled localization systems at companies including Notion, Lyft, and Medium, supporting global product launches across dozens of markets
We work with clients to build out localization infrastructure that:
- Has a low cost of entry, but can scale as the company enters many regions and languages. This where a lot of companies get stuck.
- Pivot from outdated batch oriented, build time translation to runtime localization infrastructure that is in line with modern platform architectures.
- Enables customer touch points to be continuously localized in near real-time. Eliminates most manual processes, reduces release friction and enables companies to ship across languages simultaneously.
- Combines AI and human translation processes, to optimize for speed, quality and cost based on content type and risk tolerance.
- Allows companies to scale while running a lean localization team.
- Proactively identify, avoid or mitigate technical and cultural risks, while avoiding localization related regressions.
Most localization efforts are managed as workflows. We treat them as infrastructure. If you are starting your localization path or refactoring an existing implementation, we can help.
Reach out to brian@loctechpartners.com
Recommended Reading
We’ve compiled a library of articles that address most of the issues that engineering teams will encounter related to localization and internationalization.
Foundations
Building For Runtime Localization – traditional localization assumes the content is known ahead of time, an assumption that is out of date in a world of dynamic, AI first platforms. In these systems, localization must operate as infrastructure: resolving, generating, and improving translations in real time.
Translating At Scale – this article is a companion piece to Building For Runtime Localization. It discusses the challenges of building localization infrastructure at scale, and how to directly couple translation to ROI, something which few localization platforms consider.
Future Proofing Your Codebase – These low cost, no cost techniques will save you enormous amounts of time and money caused by tech debt (spoiler alert: hardly anybody does this).
If you don’t read anything else on this website, be sure to read these articles and share them with your EPD teams!
Systems and Pipelines
Choosing A Translation Management System For Scalable Localization – one of the most important decisions you will make is to choose a TMS that works well with your CI/CD pipeline, CMS and other platforms. This article explains what TMSs do and how to evaluate them.
Building A Dynamic Translation Pipeline – in many applications, user facing content lives outside of message catalogs defined in code. Static message catalogs are also outdated in an era of AI first software development. This article explains how to build a dynamic translation pipeline around popular TMS platforms.
Choosing The Right Content Management System (CMS) – just as it is important to follow coding best practices, it is also important to choose a CMS that can support multiple languages and that integrates well with popular TMS platforms.
REST APIs And Localization – if you provide a web (REST) API, you may also need to partially or fully localize the content it serves, depending on how it is used.
Scaling Challenges
Analytics : Prioritizing Localization By User Impact – ranking texts by their relative visibility enables translators and reviewers to focus on high visibility, high risk texts first before proceeding to lower visibility content that is handled by more automated AI-first workflows.
Localizing AI Based Services – it is becoming increasingly common to build applications on top of AI based services. This raises a new set of challenges because these services are overwhelmingly trained in English and may not function as well in other languages.
Localizing Legacy Websites And Services – most companies have legacy systems that are difficult and expensive to localize. Translation proxies and Javascript embeds enable you to render these in other languages with little or no coding.
Global Product Considerations
Beyond Localization : Currencies And Payment Methods – translation is just one component of international expansion. It is also important to support other currencies and payment methods that are relevant to the markets you are targeting.
Hiring Strategically – as you hire, look for EPD (engineering, product and design) staff who speak other languages. They can help find and fix language related defects that might otherwise go unnoticed.
Want to learn more? Reach out to brian@loctechpartners.com or book an appointment for a meeting.