Fraud protection solution Simility raises $3.45 million

The company combines machine learning with human analysis to detect hackers evolving tactics

As we have seen so many times in recent years, with hacks on Target, Sony and the U.S. government, and now most recently on Ashley Madison, Internet security is more important than ever before.

As hackers get more sophisticated, it become imperative that a system combine human analysis with machine learning, so that it can catch the signals of a breach before it ever happens.

Simility is a system that does just that, and now it has revealed that it has raised $3.45 million in its first round of funding, The seed round was led by Accel Partners.

Simility provides an adaptive fraud prevention solution, one that is highly customizable and that combines machine learning with human analysis in order to detect a wide range of fraudsters’ evolving tactics.

"Fraudsters have become very sophisticated with easy access to cloud computing and open source technologies. We see automated fraud attacks at scale previously unseen that have been crafted with the intent to go undetected by typical fraud rules or static models,"  Rahul Pangam, CEO of Simility, told me in an interview.

"The best way to stop these attacks is to have a system that achieves harmony between machine learning models and human analysis. This system provides simple yet powerful User Interface for Fraud Analysts to turn their insights into signals, test their signals, and scale those signals via machine learning. This system is Simility."

The typical customer for the service is a company in the sharing economy, in e-commerce and fintech companies.

An e-commerce site, for example, would use Simility to detect when a customer is purchasing goods using a stolen credit card. A fintech company, meanwhile, would use Simility to detect compromised user accounts. And a marketplace would use Simility to detect if a seller in engaged in scams.

The return on investment for customers comes in the form of smaller financial losses that come with uncaught fraud. In addition, inefficient and outdated fraud detection techniques cause friction for users. And low confidence in automated decisions made by fraud detection systems leads to need for manual fraud reviews and hence increased staffing cost.

There are plenty of fraud detection systems out there, but what sets Simility apart are three things, Pangam said.

"Simility's tools offer optimal balance between machine learning models and rule-based systems, which makes our solution more agile, adaptive and versatile. Second, Simility empowers fraud analysts with the most sophisticated tools to analyze fraud patterns," he told me.

"Third, in building our solution, we've leveraged our founding team's combined 27 years of cross-functional experience successfully defeating fraudsters at Google. We are in a uniquely qualified class of fraud detection and prevention experts."

Founded in 2014, Simility is currently in private beta. The company will use the funding to expand its data science, engineering, marketing and sales teams as it prepares to launch its public product early next year.

With 13 current employees the company hopes to add between 10  and 15 employees by the end of 2016.

"Simility comes from the word similis, which is the Latin word for 'similar' and that reflects our philosophy to fraud prevention. Our combined 27 years of fighting fraud at Google taught us two lessons. The first was to combine the power of machines to recognize similar and dissimilar signals with the ability of humans to create meaning out of them. The second was to give front-line fraud fighters tools that would empower them to put their domain expertise and knowledge to use without limitation. With Simility, we’ve achieved both and more," he said

"There is an old saying: 'You can’t really understand another person’s experience until you’ve walked a mile in their shoes.' We spent combined 27 years at Google walking in the shoes of our customers. Our fraud prevention solution is built by 'fraud fighters, for fraud fighters.' We are continually working with our customers to prevent the world's most vexing fraud and abuse attacks."