Thirdwatch prevents fraud and reduces RTO in e-commerce transactions in real time using AI.
OverviewBack to top
Thirdwatch prevents fraud and reduces RTO in e-commerce transactions. Fraud is a big problem for e-commerce companies. Despite having multiple checks and balances in place, e-commerce companies continue to fall prey to the machinations of fraudsters. Most of the systems used by e-commerce companies are rule-based, rules are not adaptive and therefore fail to identify new fraud patterns and these broad rules are not capable of differentiating between genuine customers and fraudsters.
For instance, if someone returns more than three packages with an e-commerce company the user is not allowed to make further purchases. Rule based system block more genuine customers than fraudsters as they are very generic in nature like blocking entire zipcodes and even regions.
First Register here to get an API secret key. This registration is free of cost, and on successful verification of email address, your API secret key will be sent to your email address. This API secret key will be used in the Magento configuration page for Thirdwatch settings. No additional fee is required for the integration of Thirdwatch services for trial period. Post-trial, a nominal pricing model will be levied. To know more about our pricing, please click here.
How can Thirdwatch help?
Thirdwatch is an Artificial Intelligence based solution which catches possible fraudulent transactions before shipment. Once an e-commerce company integrates with Mitra, the AI engine from Thirdwatch, it generates a flag for every transaction, green for genuine transactions and red for risky transactions. While green flagged transactions directly go to shipping, red flagged transactions go to a review dashboard, to be reviewed and acted on by customer service team of the e-commerce company. The e-commerce company can also automate part of the review process by auto canceling transactions above a certain risk score, further reducing the cost of manual operations.
Mitra uses 200+ parameters to evaluate each transaction and evaluates each transaction in less than 200 milliseconds. It also uses micro models for address profiling, device signature, account profiling and velocity profiling. It captures the whole user behavior, including time taken to place an order, the device used to place the order, IP address of the user, location of order and many more.
When a transaction is flagged as red, most important reasons of why the transaction is risky are also shown on the dashboard along with transaction information and transaction risk score. Our AI-based engine is adaptive and by using an ensemble of AI models including classification models and anomaly detection models, Mitra can flag even novel fraud patterns.
Issues Thirdwatch intend to reduce are:
- Return to Origin: In this, the item is not received by the customer and company loses a significant amount of money in logistics, inventory, and operations.
- Promo code abuse
- Return Duplicate Item fraud: Duplicate item is being sent on returns.
- Payment and Chargeback fraud
- Network effects benefit all Thirdwatch clients. For instance, if a customer has done some fraud with e-commerce player ecom1.com, this information is being leveraged by the AI engine for all other clients without disclosing on the source of information and maintaining complete anonymity.
- Data captured from clients is secured in multiple ways. All our API’s are on https and any data stored on Thirdwatch’s servers are always saved in encrypted format. Each chunk of data is encrypted with a different key, which is saved on a different server making it extremely secure.
Read more about Return to Origin here
Release NotesBack to top
- Compatible with CE: 1.8 1.8.1 1.9 1.9.1 1.9.2 1.9.3
- Stability: Stable Build
Stable Release version 1.0.0