Signifyd raises $20M to stop false credit card declines

The company saves retailers money they lose by declining non-fraudulent charges

With credit card fraud in the news so much over the last few years, with so many big name stores, like Target, Neiman Maruc, Home Depot and K-Mart getting hit, it's understandable that credit card companies would be on the lookout for potential fraud. It's kind of their duty to do that.

Have they taken it too far, though, potentially costing retailers billions of dollars in legitimate transactions? That's the belief behind Signifyd, a fraud technology solution for e-commerce stores, which closed a $20 million Series B funding round on Thursday.

The round was led by Menlo Ventures, with additional participation from Allegis Capital, IA Ventures, QED Investors, Bill McKiernan and Tim Eades. This new round brings Signifyd’s total funding to $31 million to date.

Signifyd solves the challenges that e-commerce businesses persistently face, which is money lost in chargebacks, customer dissatisfaction from mistaken declines, and operational costs due to tedious, manual transaction investigation.

"I was recently traveling in Asia, and I went to buy a pair of shoes from a U.S. retailer website. I was declined because the retailer thought it was a fraudulent transaction. I went to another site, and I was declined again, for the same reason. I'm not alone in this, a lot of people have been in this position. 1 in 6 credit card holders are declined because they are suspected of being fraudulent," Rajesh Ramanand, Signifyd's co-founder and CEO, told me in an interview. 

"This is a fundamental problem.  The fear of losing $9 billion due to credit card fraud, costs e-commerce retailers $118 billion in good revenue, customer who were good but were not approved. That is a huge amount of money."

To solve this problem, Signifyd offers E-Commerce Assurance, a financial guarantee protecting online retailers in the case of chargebacks. It is supported by a full-service cloud platform that automates fraud prevention allowing businesses to increase sales and open new markets while reducing risk. It also overhauls how risk assessment is done.

"Traditionally what used to happen in risk assessment is that it generates a score based on patterns of past behavior. You've bought from this location 400 times in the last two years, for example. And that score gave merchants the ability to make an assessment, like a FICO Score, so they could approve or decline," Ramanand explained.

"We are different because we do not just look at traditional sources about past history, but the overall offline and online identity, including where the purchase is coming from, where it is being shipped to, and even their social activity. That information comes down to us so we can say, 'This person's offline identity matches to their online identity.' It's a decision that's backed up by more data, not just that score, and it's telling you go ahead and ship this because we've seen these patterns of behavior."

Signifyd takes less than one second for a decision to be made on whether a charge is fraudulent or not, and it has seen approval rates greater than 90%, whereas, with older scoring methods, had approval rates of 86 percent to 88 percent approval rate, with someone manually review the other rest. 

"What this means for merchant is that they don't have to hire a  team of people, spending hundreds of thousands to review transactions," Ramanand said. "Even if they made mistake, after reviewing they were liable to pay for the charge back so it made them risk averse. After one or two charge backs from Russia, Russia becomes bad."

Merchants see benefit in three areas. First, they have a charge back rate that is 0.5 percent to 1 percent, which Signifyd eliminates. It also sets rid of the 0.7 percent to 1 percent they spend in in people costs to review transactions. Most importantly, most merchants tend to decline 2 percent to 5 percent of charges that are not fraudulent. 

As a result, merchants using Signifyd typically see gross profits rise by 20 percent. 

Signifyd has been growing fast. In 2015, it increased to a run rate of $5.6 billion in transaction volume, with an 8X year-over-year revenue growth. It also Grew to more than 2,000 customers adding clients such as Lacoste, Peet's Coffee, Shane Co. and Jet.com; it is also in use by multiple companies on the Fortune 1000 and Internet Retailer Top 500 list.

It also tripled the number of employees, including bringing on executives from Axcient, Citrix and PayPal, and will use this funding to growing the team even more. It currently has 50 employees and plans to double that over the next 12 months.

"We are doing millions of transactions, from the smallest mom and pop store to large retailers. As we scale and add more retailers, we have to make sure we can continue to get the kind of engineering to support that, so we need to add a lot more engineering talent," said Ramanand.

The hiring will also include fraud experts.

In addition to the funding new, it was also revealed that Pravin Vazirani, a general partner at Menlo Ventures, will also join the Board of Directors at Signifyd, joining Bill McKiernan, founder of CyberSource Corporation, who joined the Board earlier this year

Ultimately, Ramanand sees Signifyd becoming larger than monitoring potential credit card fraud.

"The way we think about what we're building is like a  new kind of insurance that hasn't been done before. There's lots of different data points, and we will always protect you so you can scale the business. We call this 'micro insurance,'" he said. 

In the short term, Ramanand wants to scale this market, and become the go-to solution for small and large retailers  around the globe. In the long term, though, he does not want to be limited just to fraud and e-commerce.

"We want to break into other markets. Something as small as the way you buy car insurance, in 2005 it was same way that is in 2015. It's based on fact that you're 35 years old, have two kids, you're married, you've never had an accident. Health insurance is based on how old are you, and information about your past medical history," he explained. 

"Whats changing in today's world is that you don't drive five days a week, you take Uber. Why should I pay for insurance if I'm not driving? It can be entity based model, based on events."

So, if someone is driving to San Francisco, they can insure that one drive. If they are taking a physical, they can insure that one visit. 

"There are lots of interesting opportunities. We are building a platform and can use the same technology to almost disrupt insurance."