The Human in the Loop
Respecting the emotional toll of LangOps

By Didzis Grauss

Localization is a constant reflection of how far the simple idea of people communicating with each other in their native language has come. As a business, we’ve gone from managing tasks in spreadsheets and translating word by word to workflows of the highest complexity accompanied by productivity of unseen measure. Technology is what makes localization one of the fastest growing and most exciting industries to be in. However, for an industry that has fallen in love with technology quite recently, there are investments beyond money to be paid. The people involved in operations and production are paying a price to keep up and coexist with the rising impact of artificial intelligence and machine translation.

We are living in the most innovative era in the history of the language industry. Nimdzi’s recent Language Technology Landscape analysis revealed an astounding increase from roughly 500 language technology solutions in 2017 to more than 800 in 2022. This is most likely helped by the increased investor interest in the industry in recent years, and if history is any guide, this trend is not going away anytime soon.

Alongside the massive initiative to create more solutions, the need for a qualified workforce to orchestrate the increasing number of systems is growing as well. Although there are estimations by the World Economic Forum that by 2025 technology will create 12 million more jobs than it will destroy, there is a looming question about the nature of these jobs and what emotional toll these jobs are going to create on people that choose to do them.

Enter LangOps, a concept that is quickly gaining wind in its sails. While it is a very future and tech-centric summary of ideas, its manifesto, which was published in MultiLingual’s December issue, stresses the importance to respect the human in the loop. The concept recognizes that human knowledge will be the driving force of LangOps, but also acknowledges the importance of supporting workers emotionally. Because the LangOps operators or orchestrators are at the very center of the technological meatgrinder, employees in these positions are subject to increased levels of stress, leading to burnout.

Technologies perform scaling very well, which is not necessarily true for the human brain. If overlooked, the imbalance of technology versus human capacities can create an extreme toll on the people on the execution end. Therefore, it is exciting to see tech companies recognizing the importance of providing language operators with the tools to sail through the vast ocean of language technologies. Companies such as Phrase and are spearheading ideas that give the power of workflow management back to managers in charge of localization. These solutions, while focusing on benefits to the end users, are crucial to balance the scales between technological advancements and the increasing levels of stress they put on the people in the middle.

A great quote started off the Custom.MT’s very successful panel on “ChatGPT in Localization” recently. It was shared by the keynote speaker, Marco Trombetti of Translated, who said, “Language is the most human thing out there,” later continuing, “Language is the driving force of human evolution.” I cannot help but wonder what lies ahead and how will the marriage between technology and language evolve further.

Looking beyond the aspects of mental health, with so much technology at our fingertips, we’re subject to the potential risk of technology overuse. There are new solutions created every day, helping companies create, localize and publish content. It is now true that there is a spectrum of human involvement in content creation, something that we did not see on large scale prior to the introduction of ChatGPT. As more developments in AI writing come to fruition, the need for human involvement on the content creation spectrum is subject to decrease. If the language singularity is indeed upon us, does that mean that future content will gravitate toward a recycled version of itself?

There is a lesson to be learned here. Similar to how translators and content writers are worried that AI is bound to take their jobs in the future, companies choosing to artificialize their content can suffer from equal outcomes. A recent survey analysis published by Riteba McCallum in MultiLingual’s December issue asked how close MT is to reaching human parity. It highlighted a very crucial understanding to what “human level” means in terms of content creation and localization. The survey showed that, in a way, we’re already here – AI and MT can generate and localize content on a level that can be compared to a person being average at their job. Do companies choose average outcomes? It depends, if they are free, they might.

We’re landing on the basic question of content purpose and strategy. Is it for a company to engage their audience, help engagement, or encourage purchasing? It all lines up with the need to create an emotional response. No one is better at emotions than a human. So, the best bet is to match the intensity and passion that goes into the product’s creation and pass it on to storytellers on a local and global scale.

It’s important to remember that language is fundamental to our growth. Therefore, it must remain the distinguishing factor in how companies tell their stories and create interest.

Stress has always been a killer of all sorts. Employees suffering from increased levels of stress often experience reduced work performance, lack of creativity and in more extreme cases are forced to change jobs. A study by the Society for Human Resource Management found that stress was the top cause of employee turnover, with 41% of employees citing stress as the reason for leaving their job. The localization industry is no different in that regard. We see an increasing number of high-level localization professionals leaving work that they love doing, just because it is not sustainable in the long term.

Scaling to multiple markets can highlight underlying issues in a company’s go-to-market strategy. It is not about finding weaknesses in this case; you put enough pressure on any single point of a stick, and it will break. And what is scaling if not an exponential pressure multiplier for a professional tasked to manage it?

To counteract and reinforce these positions, we must first recognize and respect the effort it takes to be a localization manager. Companies must measure the work of localization against the correct criteria, which means understanding that there are functional KPIs for localization that do not correspond to revenue, but rather affect the client-facing indicators, such as market presence, number of users, customer retention and sales support.

Attitudes toward localization might not change overnight, so let’s look at other ways to cope. While technology can be a significant source of stress, there are tools that can help enterprises solve problems and introduce structure and process automation in the pipeline. Custom workflow orchestration, no-code integrations, and automated pipeline movement are some of the things that seemed out of reach for localization professionals just a few years ago. I am confident that the key to future language technology success lies in its ability to not only solve client-facing issues but also empower the highly competent people managing them.

If we look at stress as not just an external condition, there are ways to reduce stress from within. A study published in the Journal of Occupational Health Psychology found that workers who had more knowledge about their job and the demands of their job experienced less stress and burnout than those who had less knowledge. This supports the idea that knowledge can help reduce stress by increasing one’s sense of control and improving decision-making abilities. After all, having a more complete understanding about the processes equips the managers with better coping mechanisms and improved crisis-thinking when necessary.

Stress can also be managed by transferring a part of it to someone else. That is where outsourcing comes in. Having the right strategic partner in localization that hears and adapts to you can elevate stress and reduce friction associated to scaling on a global level.

A big part of resolving stress in localization project management is understanding where it comes from. While understanding measures and correct setting of KPIs is a key part, we should also consider management compassion to play a huge role in managing stress.

Burn out is a common problem in the localization industry, and it is often caused by a product-centric approach, where localization is viewed as a support function, and its goals and challenges are often overlooked in favor of keeping the product cycle moving forward. This can create a strained environment where supporting roles feel undervalued, particularly when the credit for achieving goals is attributed solely to the product or sales teams. Additionally, the pressure to optimize budgets often leads to understaffing, which can contribute to burnouts.

To address this issue, it can be beneficial to adopt a “little picture” approach that focuses on identifying and managing stress in procedural positions, such as localization management. This involves recognizing the importance of localization and valuing the contributions of all team members. Creating an inclusive work culture that fosters a healthy work-life balance really is the only immediate way to reduce burnouts and promote sustainability in the workplace.

Another great way of making the marriage of software and people work in localization would be to use the freedom gained by optimizing and automating processes and devote it for creativity and relationships. Rather than squeezing every last drop of productivity out of people, we should use technology to enhance and support their work, not replace it.

Recently, on a podcast by LocReady, Jeff Beatty, executive director of product management at Disney, shared great insights on the topic of where localization is going. One of his observations really resonated with me. He said that localization has become increasingly focused on cost-cutting and efficiency, rather than taking risks and exploring new possibilities. Beatty also expressed concern about the overuse of AI and machine translation in localization, and the potential negative impact on creativity. But he also suggested that we can find ways to work with technology without losing sight of the human aspect of localization. These ideas really sum up the challenges and opportunities facing the human side of localization. While there have been many advances in automated translation, there’s still a lot of room for improvement in how we manage and execute localization projects. By embracing technology that focuses on the managers in the middle, we can empower both the managers and translators to do their best work and avoid having them pay the emotional price for it.

The localization evolution is just starting to take off its training wheels, driven by technological innovation and the companies that push its boundaries. While these advancements can sometimes be seen as a threat to jobs, it’s important to reframe them as an opportunity to refocus on the core values of the industry. As professionals, we should be leveraging technology to enhance human communication and connection. By prioritizing the well-being of our people, we can ensure the long-term sustainability of the industry, and continue to attract and retain the skilled and passionate talent who make it thrive. This is the perfect time to prove that innovation is not just about being passionate about technology. It is also about being compassionate towards the people in the loop.

Didzis Grauss is passionate about design fundamentals and execution for software companies, opening global markets through localization. He is a co-founder of Native Localization.



Subscribe to stay updated between magazine issues.

MultiLingual Media LLC