OTAs, AI and ML
Fraud Prevention

How Emerging Travel Group Uses AI To Weather Fraud Turbulence

With the global travel industry expected to lose $11 billion to fraud this year, online travel agents (OTAs) must safeguard against threats across multiple channels. In the latest Merchant Fraud Decisioning Playbook, CEO Felix Shpilman, of global travel services provider Emerging Travel Group, discusses how artificial intelligence (AI) and machine learning (ML) can help break data silos and unlock insights for omnichannel protection.

The global travel industry has seen its fair share of turbulence over the past decade. The industry is projected to reach approximately $11.4 trillion in value by 2025, but fraud losses in the space are estimated to exceed $25 billion this year alone as cybercriminals employ sophisticated tools and technologies to steal funds and valuable customer details.

Fighting travel fraud is particularly challenging, however, especially for online travel agencies (OTAs) and hotels, according to Felix Shpilman, CEO of global online hotel accommodations and travel services provider Emerging Travel Group. These businesses are often hit hard when attacked, prompting many to employ artificial intelligence (AI) and machine learning (ML) when building their defenses, he explained. Such technologies safeguard the onboarding and payment processes without adding unwanted frictions.

Battling Fast-Growing OTA Fraud

OTAs often have diverse sets of travel products that can be sold across various channels, making fraud more difficult to detect unless automated processes are involved. Such tools can easily sift through transactional and user data, but many online travel merchants instead rely on systems that focus on purchase order (PO) data to find possible threats — a process that often results in false positives. OTAs are also seeing a rise in fraudulent chargebacks, which are estimated to increase 20 percent per year, taking a huge bite out of merchants’ profits.

“All of the large online travel companies spend a significant amount of resources fighting fraud,” Shpilman explained. “Fraud has evolved over the years, and chargeback fraud, which used to be the primary type of fraud that people fought with, is just one of the many things we have to fight with [now].”

Chargebacks and false positives can also take heavy tolls on online travel merchants’ bottom lines.

“Travel, in principle, is a fairly low-margin business [in which] the majority of the companies make anywhere between 2 [percent] and 15 [percent] — maybe 20 percent — of the turn- over [on] their net revenue,” he said.

Online travel platforms looking to stay competitive need to continually balance robust safety protections with seamless experiences for customers.

“Fighting fraud is an important priority for most of the modern travel companies,” Shpilman said. “That said, the balance between fighting fraud and having a consumer-friendly user experience is also very important.”

The company reexamined its in-house operations and the merits of working with third-party solution providers to achieve this balance. It also revamped its authentication procedures, ensuring that only verified customers can transact on its platform.

Deploying An AI-Based Fraud Solution

Emerging Travel Group had originally built a flexible, in-house, anti-fraud system, Shpilman said, but the rules-based product was difficult to scale in unknown markets. The company wanted a solution that could unlock greater access to data from multiple databases, companies and industries so it could provide deeper protection and accuracy. It thus sought outside help to acquire the right anti-fraud offering.

The company implemented a third party’s AI- and ML-powered system in 2018 that works in tandem with its internal system to better detect fraudulent behaviors. The product has largely made customer experiences frictionless by keeping authentication on the back end, where customers cannot experience it, according to Shpilman.

“The customers have no idea they are being evaluated by the third-party solution,” Shpilman explained. “The idea is to ensure that the customers’ experiences are not diminished by the fact that we have this solution [while still providing added security to protect their transactions].”

Emerging Travel Group only authorizes credit card payments, he noted and presently does not employ additional protocols like 3D Secure or two-factor authentication (2FA) to thwart CNP fraud at checkout. It does rely on automation to improve verification outcomes, however.

“All of the risks are covered by the analysis made by the anti-fraud system that uses AI and ML to analyze the customer and make a decision on the order,” Shpilman said. “There is no human involvement in the process. All of the relevant data is sent to our third-party tool via an [application programming interface, which returns its] decision [on] whether we should or should not approve the order.”

The AI-based system has largely eliminated Emerging Travel Group’s fraudulent chargebacks. If the chargebacks do happen, however, the system also helps to contest them. Affected banks are also contacted and given as much information as possible, including evidence from the fraud prevention system, partners and hotels.

Challenges With Platform Integration

Shpilman acknowledged that integrating third-party systems often requires extensive work and support from development teams, especially when firms have their own in-house software. Application programming interfaces (APIs) can feed the necessary data into the new AI-based systems once the integration is complete, enabling the product to analyze users’ behaviors and determine fraudulent orders.

Having access to data insights and implementing the right tools is critical to accommodating the emerging payment trends that impact OTAs, such as mobile payments and virtual cards. Some organizations use the latter to provide business travelers with 16-digit credit card numbers for one-time purchases, including hotel stays, rental cars and other travel-related services. Virtual cards act as anti-fraud tools on their own, preventing permanent credit card and personal information from entering OTAs’ databases.

The global travel industry has had its share of ups and downs, but AI- and ML-based anti-fraud systems can safeguard everything from onboarding to payments while easing frictions on customers. OTAs and hotel management teams looking to remain prominent in the space will need to stay on top of guests’ payment preferences and ensure the anti-fraud solutions they develop travel well — both today and for years to come.

——————————

PYMNTS STUDY: THE CROSS-BORDER MERCHANT FRICTION INDEX – JUNE 2020

The PYMNTS Cross-Border Merchant Friction Index analyzes the key friction points experienced by consumers browsing, shopping and paying for purchases on international eCommerce sites. PYMNTS examined the checkout processes of 266 B2B and B2C eCommerce sites across 12 industries and operating from locations across Europe and the United States to provide a comprehensive overview of their checkout offerings.

TRENDING RIGHT NOW