Signifyd, a company co-founded by PayPal and FedEx veteran Rajesh Ramanand, is launching an e-commerce fraud platform that attempts to combine shoppers' online and offline identities to better evaluate a transaction's legitimacy.
"There's often a mismatch in the different sources used in the data that's being used for ID fraud purposes," says Ramanand, Signifyd's chief executive. "A lot of merchants are declining good transactions."
Signifyd is formally launching its technology this week after about a year of development and testing.
Before founding Signifyd, Ramanand handled emerging markets risk at PayPal, where he was responsible for managing fraud and credit risk in Latin America, the Middle East and Africa. In an earlier role at FedEx Corp., Ramanand introduced a package intercept program when he led the company's risk management division for payments and shipping.
Ramanand's experience is a good asset for Signifyd, says Julie Conroy, a research director at Aite Group.
"PayPal is on the front lines of e-commerce fraud attacks, and as a result, we've seen a number of its alumni come out with very innovative fraud mitigation techniques born of necessity during their PayPal days," she says.
"Silver Tail is certainly the poster child for this innovation, and I think Signifyd is on to something very interesting in using social media data as a risk assessment tool," Conroy says. "This is certainly coming at an opportune timee-commerce companies are seeing attacks continue to increase and with little in the way of adverse incentive, there is no indication that this trajectory will change."
Signifyd's ID technology leverages algorithms to evaluate the risk of an incoming order to an e-commerce site. The company uses 120 risk indicators, and searches for consumer data trailsor footprintsin online and offline data sources.
These sources include a records of a person's online relationships (such as social networks), device fingerprints, IP geolocation (and whether the device uses a proxy to hide its location), customer history, bank data, cross-merchant blacklists, transaction velocity, search engines and public records.
"As we collect more data, we can add value of the identity vetting. The combination of public records, IP searches, social, etc., can connect the dots between a consumer's various profiles to see a match," Ramanand says.
For example, if a consumer makes a purchase from a computer located in San Francisco, but the customer's billing address is in France, that often prompts the transaction to be manually reviewed, Ramanand says. Sygnifyd attempts to find out which country the card was issued in, and the strength of the consumer's social profile. If that user's Twitter account reveals the consumer's card was actually issued in France, that information could justify approving the transaction.
"If there's a connection between the different data sources, we approve the transaction. If not we put them in a triage or a manual review," Ramanand says.
Signifyd also assigns all transactions a score to notify retailers of fraud risk, and guarantees payments in the case of chargebacks resulting from approved transactions, Ramanand says.
Signifyd charges $79 a month to vet 200 transactions, $299 a month to vet 1,500 transactions, or $999 a month to vet 6,500 transactions.