BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Using LangOps To Improve Multilingual Customer Service (And Beyond)

Forbes Technology Council

João Graça is the Co-Founder & CTO of Unbabel. 

It sounds too good to be true: multilingual customer service teams distributed across the globe, helping clients in their native tongues and remaining responsive to the nuances of the cultures they're serving.

However, by harnessing the power of Language Operations (LangOps), many companies today have been able to do exactly that, improving global customer experience, accessing new markets and increasing revenue along the way. 

LangOps is an AI-powered, human-refined approach to language translation that ties together all multilingual operations and tools throughout the departments in an organization. It removes language barriers that stand in the way of a company's expansion, typically starting with customer service — the area where it provides the quickest ROI. Let's take a look at what LangOps is and how your organization can get ready to start using language as a growth driver.

Solving A Pervasive Problem With AI

Prior to LangOps, resources for language translation were difficult to manage — and they weren't cheap. A team of full-time translators would be dedicated to each language served, limiting company growth by translator availability. Translators would also be cross-functional; your Spanish speaker would address everything related to that language, whether marketing, sales or customer service.

LangOps leverages AI and advanced analytical capabilities to increase efficiency and reduce the amount of time it takes to start servicing a new language. In addition to providing near-real-time translation and quality estimation, a centralized LangOps platform — a solution that integrates content, communications and data across an organization's entire tech stack — can be used to deliver insights surrounding customer-centric KPIs such as first contact resolution rate (FCR) and average customer satisfaction score (CSAT) per language.

Getting Started With LangOps

To prepare for LangOps implementation, there are two key steps that your organization must take. First, you need to appoint someone to take ownership of this process. I frequently see customer service leaders, such as a vice president of customer service or a director of customer operations, step forward to take charge of a strategic LangOps initiative. That's because, as a cost center, the customer service department is often interested in evaluating AI technology as a solution for driving process and cost efficiencies.

The second step involved in getting started with LangOps is to take stock of the current translation and localization efforts that are happening throughout your organization. Use the people, process and technology framework to ask relevant questions such as:

• People: Who owns our translation/localization efforts? Current localization efforts may be carried out in-house as part of an existing team or standalone department, or they might be outsourced to a translation agency.

• Process: How do we go about delivering translations for customer service inquiries, marketing communications, etc.? At this stage, it's important to identify process bottlenecks, duplicate efforts and time-consuming tasks that could be eliminated or consolidated.

• Technology: What does our current tech stack for language translation look like? This can include translation tools themselves as well as the programs they integrate with, such as customer relationship management (CRM) software, content management systems (CMS) and conversational messaging platforms.

After assessing those three pillars, you might come to the realization that your translation and localization efforts don't currently have a single owner or that they suffer from disconnected, decentralized processes that waste time and resources. You might also discover that you don't have the right technology in place or that you're juggling too many software vendors.

Once you have defined who will own the LangOps initiative as well as the footprint of your current translation and localization efforts, it's time to start thinking about the scope of what LangOps can do for your business.

Maximizing The Potential Of LangOps 

To pave the way for a smooth and effective LangOps deployment, it's crucial for your LangOps leader to have a clear idea of the results they're looking to get from the solution — in other words, the desired ROI.

Every company uses a different measuring stick, which means your organization will have its own unique blend of KPI improvements to define success. Here are a few goals we have seen organizations set (and meet) with LangOps:

• Expand multilingual customer support.

• Transition from set hours to always-on global support.

• Decrease contact center first reply time.

• Improve service level performance.

• Reduce cost per agent.

• Increase net promoter score (NPS) for specific languages.

One pitfall I see some organizations struggle with is the tendency to think too narrowly about what LangOps can do and the scope of problems it can solve. For instance, a company might take a LangOps strategy to a tactical level and say: "We have a large backlog of inquiries in French and German, so let's use this to enable our English-speaking agents to tackle those tickets."

That is certainly a valuable use case, but to really deploy this technology in a successful way and get the most bang for your buck, you have to think bigger. Don't be afraid of the change that LangOps can invoke; instead of just focusing on reducing ticket backlogs, how might LangOps allow you to centralize operations and realize cost efficiencies across the board?

For example, a localization team could use a LangOps strategy to become more efficient and scale beyond its own headcount. Instead of spending all of their time focusing on repetitive translation, some employees would provide feedback on translations to help iteratively train a machine translation solution so it can more accurately reflect brand-specific language. This shift could also give the human resources department more freedom to make hires based on technological and subject-matter expertise instead of only language fluency. 

We have evolved to a point where language is an asset to doing business on a global scale, not a barrier. Planning a LangOps strategy could help organizations get ahead of competitors that are still using a decentralized, siloed approach to language translation and localization.


Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?


Follow me on LinkedInCheck out my website