Technology now forms the backbone of every modern business and its processes—and in our case, its partners’ localization processes. Let’s take language quality assurance (LQA) technology as one example. Every client wants their global content cheap, fast and of a high quality—but we wouldn’t have been able to tick all three boxes without the technological advances of the last decade.
So, what’s new?
Considering we can no longer avoid technology in LQA, it’s now a question of how we as (human) localization professionals can leverage it to stay competitive by solving the problems quality assurance tools themselves create as they evolve (think false positives—more on this below) and finding new ways to drive down our clients’ costs.
Here are two ways in which tech and language quality are merging to produce more efficient localization workflows and better-quality results.
Customized rules, fewer tools
In recent years, we’ve noticed more and more localization teams adopt CAT tools to perform quality checks. But there’s still room for improvement.
Yes, the move from desktop to the cloud has been instrumental in helping translators check for errors from right within the CAT tool in real time. Yet in many cases, even cloud-based CAT tools aren’t efficient enough. It’s one thing to be able to detect errors as you work, but what if you get the same types of errors over and over again?
This is why it’s increasingly important to use tools we can customize to our clients’ needs—such as to their specific spacing and punctuation guidelines. Since translation management systems and CAT tools focus more on the translation interface, their ability to customize linguistic checks to client-specific requests or instructions is limited. As a result, translators have had to rely on separate tools like Xbench or Verifika to run checks on their work. Instead, we recommend implementing fully customizable QA tools that can catch all errors and help clients consolidate their tech stacks.
What will we see in the future? At present, quality automations are designed to detect the errors we tell it to detect. What if, in the future, they become intelligent enough to tell us what we need to correct? In this era of fast-evolving artificial intelligence and machine learning, the possibility isn’t far out of reach. Just look at the rate at which natural language processing has advanced—similarly, quality automations should soon be smart enough to correct style and tone in translations as well as basic errors like typos.
External technology ‘coaching’
What the client wants isn’t always what the client needs. As you know, a large part of our job at the quality stage is helping clients and their translation teams understand what’s best for them.
That starts with “coaching” on specific tools. Clients and translators need to understand that QA tools produce false positives, for example. Without that knowledge, translators might assume their tool isn’t working properly when it catches errors that aren’t really there—and if that leads to distrust of the tool, they might overlook legitimate issues. The idea of technology coaching is to train linguists how to be self-sufficient and not put all their faith in any one QA check tool. You never want to be tied down to one—again, in case needs change.
But more than that, as we’ve found at RWS Moravia, it’s becoming important to take a consultancy approach to solving clients’ quality needs and providing customizable solutions in relation to quality assurance. If a client’s team uses a TMS, CAT tool or rigid third-party QA tool, we as their LSP might have to consolidate or switch their QA tools as their needs change and program evolves. Then it’s a scramble to quickly teach the client’s translators how to run and customize checks with the new tool for a seamless transition.
This is why we encourage a consultative approach to test, analyze and make decisions on the best tool for each unique client at the onset.
What will we see in the future? To be clear, “technology coaching” is a term we coined ourselves to create a culture of educators at RWS Moravia. Who knows if it’ll catch on? The point is, those of us who can keep on top of technological changes—and convey those changes clearly to clients—will be best positioned to help companies avoid costs on an ongoing basis.
What do these advancements tell us? That technology has changed the very meaning of “quality.” It’s not just about quality assurance anymore—it’s about the overall quality of service we provide our clients and our flexibility with the QA tools we use.
Which brings us to why we still need humans. Helpful as localization technology continues to be, it can’t work without the quality frameworks we put in place, and it won’t always solve all our problems. Even if QA tools become smart enough to detect and fix errors for us, they’re no substitute for human agility. In other words: it’s okay to be objective about what each new advancement means for you and your customers.
Questions about how to integrate customization and coaching into your own quality programs? Ask away!