What is collusion fraud?
Collusion fraud occurs when two or more individuals work together with the goal of manipulating a service or platform for their gain. This could be to gain an advantage over other competing elements or to defraud a company and its users financially.
In the mobility industry, this type of fraud typically sees some form of collaboration between drivers and passengers directly or between multiple drivers to exploit the system.
What are the most common forms of collusion fraud on ride-hailing apps?
There are three core types of collusion fraud that take place on ride-hailing apps.
Drivers-passengers collusion
The most common type of collusion fraud on ride-hailing apps sees drivers and passengers conspire to conduct fraudulent activities. This can lead to a number of scenarios that harm both a business and its users.
Drivers-drivers collusion
Another type of collusion fraud involves a group of drivers who team up to take advantage of the ride-hailing platform. A business is likely to find a fraud ring that is related to this form of collusion as it often is one part of a wider fraud scheme.
Self collusion using multi-accounts
Unlike the types of collusion mentioned above, where two or more people work together to manipulate the platform, in this case a single person controls multiple fake accounts to deceive the system into believing that many users are interacting with the app.
See how we helped ride-hailing app BEAT (now part of FREENOW)
Examples of collusion fraud on ride-hailing apps
Collusion fraud and abuse can take place in a varied number of ways on these types of on-demand mobility platforms.
Fare surges
This is a sophisticated form of fraud in which drivers collaborate to artificially inflate prices. A group of drivers agrees to go offline or stop accepting rides simultaneously in a specific area. This sudden decrease in driver availability tricks the system into thinking there is high demand but low supply, triggering surge pricing. Once surge pricing is activated, the drivers come back online and start accepting rides at the inflated prices, earning more per ride than they would under normal conditions.
To carry out this type of fraud, drivers typically use location spoofing and fake GPS apps to create false signals about their whereabouts, deceiving the app into believing they are in a different location.
Fare surges can also occur when drivers and passengers collaborate to artificially inflate demand in a specific area. The passenger, using multiple fake accounts, repeatedly requests rides in a particular zone, causing the system to detect high demand and increase fares. The colluding driver then accepts these high-priced rides, profiting from the artificially inflated rates.
In both of these circumstances, it creates a false demand for dynamic pricing and incorrect fare charges.
Fake rides
In this instance, a driver and passenger would coordinate to create fake ride requests. The passenger typically uses a fake or multiple accounts to request rides, which the collaborating driver then accepts. They may also use GPS spoofing to simulate the trip.
By doing this, the driver can earn incentives or bonuses for completing rides that never actually took place, these earnings can then be split with the passenger.
Promotion Abuse
Fraudsters exploit ride-hailing platforms' promotions, like discounts or free rides meant for new users, by creating multiple fake accounts. They then book rides using these accounts to claim the promotions.
The driver, in collusion with the fraudster, accepts these rides, knowing they are part of a scheme, and splits the financial gains from the fraudulent promotion with the passenger.
Driver rating manipulation
A driver might create several fake accounts on a single device, using app cloners and emulators, to impersonate different passengers, request rides that never happen, and give perfect ratings.
This can help them maintain a high rating and secure better opportunities on the platform, even if their actual service quality doesn't deserve it.
The consequences of not spotting and preventing collusion fraud
Collusion fraud in ride-hailing platforms poses a significant threat to the fairness and trust of these services.
When drivers or users exploit the system for personal gain, it disrupts the natural functioning of the platform, leading to a range of negative consequences.
Unfair opportunities
When drivers manipulate ratings and ride requests through collusion, they gain unfair advantages over honest drivers. This skewed system can lead to better placement and more lucrative opportunities for those engaging in fraud, while legitimate drivers are left at a disadvantage, receiving fewer rides and earning less money.
Increased costs for passengers
Surge pricing leads to significantly higher fares for passengers, who may not realize they are paying more due to fraudulent manipulation. As a result, passengers may grow frustrated and leave for other platforms in search of fair pricing.
Distorted market dynamics
Continuous manipulation of surge pricing can distort the natural supply-demand balance that the platform relies on, ultimately affecting the overall efficiency of the service.
Loss of trust and reputation
Beyond the financial impact, collusion erodes the trust that is critical for ride-hailing platforms to thrive.
Passengers may lose confidence in the reliability of ratings and reviews, and drivers may question the fairness of the system. This loss of trust can lead to decreased user engagement and satisfaction, harming the platform's reputation and overall effectiveness.
When passengers and drivers begin to doubt the reliability of rides, ratings, and the fairness of the system, they may seek out other platforms. This can severely damage the platform's reputation and threaten its position within a country/market so uncovering potential collusion schemes is vital to long-term sustainability.
How to detect and prevent collusion fraud on ride-hailing apps
Efficiently detecting and preventing collusion fraud involves implementing a risk AI solution capable of monitoring and analyzing user behavior in real-time.
Device Identification
Device identification can detect anomalies such as creating multiple accounts or attempting to disguise the device by performing a factory reset to appear as a new user accessing the platform.
This is a powerful solution that persistently identifies devices and spot tactics commonly used in collusion fraud.
By using device fingerprinting technology, it’s possible to prevent fake accounts at scale and block fraudsters from accessing the platform.
Real-time monitoring
Implement a platform that manages risks in real-time, continuously profiling device sessions and returning real-time actionable risk signals. This helps platforms identify the exact moment a good user turns bad.
In the context of ride-hailing companies, it’s important to flag immediately when a user starts to use GPS spoofers, app cloners, autoclickers, or any other suspicious software.
Artificial intelligence and machine learning algorithms
AI can continuously analyze vast amounts of data in real-time, faster than any operative in your team. Risk managers can identify suspicious patterns and flag potential collusion fraud attempts as they happen, minimizing losses.
Machine learning algorithms can learn and adapt over time, becoming adept at recognizing complex patterns in fraudulent behavior. These patterns might be subtle, but AI can identify seemingly unconnected data points that indicate a high risk of fraud. Furthermore, you can enhance any algorithm by tapping into an intelligence network, which can provide you an insight to fraud rings and methods from around the world.
How SHIELD works to prevent collusion fraud
SHIELD can help you to stop fraudsters working together to collude to abuse the platform by continuously profiling each device session and providing real-time risk signals.
Our Device-First Fraud Intelligence acts as the first line of defense against fraud and does this by accurately identifying the physical devices behind attacks. This is down to our persistent device identification and real-time actionable risk intelligence, offering a comprehensive view of user activity on the platform throughout their session from start to finish.
In one case, we identified a single device linked to 310 accounts. 286 accessed the passenger app and 24 accessed the driver app. All from the same device, this clearly indicates that the same person created multiple fake accounts and accessed both apps, a strong sign of self-collusion fraud.
With SHIELD Device ID in place, it’s possible to identify fraudsters even if they try to spoof the ID or perform a factory reset, tactics that bad actors typically do to hide their tracks and conduct fraudulent activities.
SHIELD Fraud Intelligence also detects commonly used techniques for speeding up the fake account creation process or spoofing a device, such as emulators, app cloners, and GPS spoofers - tools frequently used in collusion fraud.
