3 Tips for Customer Support Leaders in the Travel Industry


Share on LinkedIn

You’d think the economic downturn would’ve put a damper on travel plans this summer. But that’s not how it’s playing out in the real world.

Call it “vacation deprivation” or “revenge travel”. But in spite of high inflation and a looming recession, consumers are hitting the road now that the pandemic and travel restrictions have subsided.

The TSA is routinely screening more than 2 million people daily at airport checkpoints — levels not seen since before the pandemic. Hotel bookings in two of the world’s most popular holiday destinations, France and Spain, are at 109% and 117% of pre-pandemic levels.

With the world’s citizens moving again, travel and hospitality brands are tasked with managing surges they haven’t seen since 2019. In travel, providing outstanding customer support means having to navigate frequent language barriers. However, staffing your teams with dedicated native-speaking customer service agents in every market is impractical and costly.

To overcome language challenges, travel brands around the world are turning to AI-based technology that enables agents to speak to global customers in their native language.

Here are three best practices for travel companies looking to utilize cutting-edge technology to deliver worldwide customer support while maintaining lean teams.

Manage demand spikes with language translation

In an ideal world there would be no language barriers. But in this world, that’s not the case, so travel brands need to provide multilingual customer support at scale to handle spikes in demand. In Unbabel’s 2021 Global Multilingual CX Report, 92% of respondents said that poor quality customer support in their native language will negatively impact their loyalty toward a brand. In a multilingual world, the stakes are high for language accuracy.

Luckily, machine translation can help.

Machine translation services sit invisibly between your customer support or service team and customers. The translation engine instantly translates chat support conversations, emails and FAQs. So rather than hiring three different people to support German, French and Spanish customers, you can equip one agent with multilingual superpowers to accommodate many languages at once.

It’s worth noting that while fully-automated AI-based translation services are fast and cost-effective, the process needs human input to achieve the highest quality translations.

For example, live chats about a support issue can be translated in seconds by autonomous AI, but email correspondences require more attention. At Unbabel, we use quality estimation technology to determine which FAQ and email translations are of lower quality and send them to our community of editors to review for overall translation quality and regional language nuances.

We’ve seen our customers adopt this human-in-the-loop AI language strategy to deliver translations to international customers and receive exceptional CSAT (customer satisfaction) scores. And they do it at one-tenth the cost of hiring full-time native speakers.

Integrate AI into the CRM

CRMs or platforms such as Salesforce, Zendesk, Kustomer and others — help power interactions with customers. Chatbot tools within CRMs are the most widely implemented use of AI in the customer support space.

When AI-based tools are integrated into CRMs — for automation and data analytics as well as language translation — the AI can handle lower level inquiries via chatbots, freeing up agents for customer issues that require emotional intelligence and empathy. During seasonal surges, your already lean customer support team won’t be slowed down by routine requests better left to an AI-based chatbot.

On the AI language translation front, travel app Hopper, has integrated Unbabel’s AI-powered Language Operations Platform into its CRM to expand the number of languages for which Hopper provides customer service. As the Hopper app becomes available in a new language, native-language support becomes instantly available via its AI-enhanced CRM. Given that Hopper did not have any native-language agents, this allowed them to grow by keeping operations in one language. Hopper’s KPIs such as CSAT score, AHTs (average handle times) and translation quality have all gone up since integrating AI language translation into its CRM.

Organize teams around their expertise

While AI can empower agents to support multiple languages during demand spikes, language translation alone isn’t a cure-all for customer support challenges during surges.

Travel brands must also organize customer support teams based on their expertise of different product lines so that the right agents are working in the right product areas at the right times.

After all, even if agents are given multilingual powers via AI, they’re not going to be much help if they don’t understand the particular issue frustrating their customers. If they’re not an expert in the customer queries they’re receiving it’s going to cause a major backlog and frustrate customers being bounced from agent to agent.

For instance, there should be enough specialist agents in key travel support areas such as “VIP benefits”, “reward points” or “hotel refunds” based on the typical query volume for each area. If only one or two agents with little expertise in “VIP travel benefits” are stuck managing hundreds of queries, then the language super powers afforded each agent by machine translations will go to waste.

Keeping up with the travel rebound

Travel brands must deliver high-quality customer service, and do it faster than ever. Companies that capitalize on AI-based language translation, integrate AI within CRM platforms and organize agents around their expertise will be in a position to satisfy global customers – no matter the language — at the scale and speed the market demands.

Vasco Pedro
Vasco Pedro is a co-founder and chief executive officer of Unbabel, a company that removes language barriers by blending artificial intelligence with real time, human translations. A serial entrepreneur, Vasco has led Unbabel since 2013, taking it through Y Combinator and raising a total of $31 million in funding.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here