<img height="1" width="1" src="https://www.facebook.com/tr?id=316029722675592&amp;ev=PageView &amp;noscript=1">

With our New York headquarters packed for a data science team meeting this week, it’s even more obvious than ever we’re bursting at the seams. How many Socureans does it take to screw in a light bulb? I don’t know, but we’ve probably got more than enough. So our intrepid CEO Tom Thimot broke things up a bit by taking a whole bunch of us on a field trip to check out a new office space we’re considering. We’re going to have to branch out someplace. We’ve outgrown our footprint, as we’ve brought on more people to build the product, test the product, sell the product, and support our expanding customer base. We’re looking at a spot about a twelve-minute walk from the current digs, with a ton more area, beautiful views, larger and more plentiful meeting rooms, and hopefully a trampoline.

I’ve been through a number of successful startups over the years, which is how I’ve paid for all the crap my wife buys. But this is easily the most exciting, and it’s not just because we’re growing. It’s always difficult trying to differentiate one software package from another, especially when the competitors always go to battle with vague features and widgets. After a while, they all look alike. But Socure’s value prop is exceedingly clear.

Our customers need to identify who’s knocking at their door, so they can safely execute financial transactions. They want a greater number of good customers, fewer crooks, and as little friction as possible. To achieve this end, we deliver more auto-acceptance of applicants, we capture more fraud, we lessen the percentage of new accounts that need to be subjected to additional friction, and we reduce the Total Cost of Ownership by simplifying identity verification. The numbers don’t lie. We produce measurably superior results.

Our accuracy just keeps growing. This means we can tell you with an amazing degree of certainty that a person is who they say they are. That’s pretty important when said person wants a loan, a credit card, an account opening. We measure this accuracy by pitting our robotically-generated models against our training data and improving our AI-driven predictive power. We’re in the high 90’s, hurtling toward complete and utter accuracy.

So how does our accuracy keep growing? Because our training data set keeps growing. Socure trains our models against more applicant data, millions and millions of rows, more than anybody on the market. No other vendor throws as many known outcomes at their machine learning models as we do. The more you learn, the smarter you are. And that’s precisely how our platform works. We learn from this ever-broadening data store how to recognize good from bad, how to understand at a very granular level the characteristics that make up a good applicant, how to see the quality (or lack thereof) in the details, things you don’t catch when you’re only doing binary matching.



The more good applicants we accept, the more money our clients make. The more fraud we prevent, the more losses our clients avoid. The more friction we prevent, the more satisfaction the applicants feel in a streamlined process.

Our training data gets bigger. Therefore our accuracy gets bigger, so our coverage gets bigger, so our customers’ customers get bigger, and to keep it all on track, our headcount gets bigger, which is why our offices are getting bigger. Now all we need is that trampoline.

Topics: Machine Learning, identification verification, socure, office

Jeff Scheidel

Jeff Scheidel

Jeff Scheidel is a technologist with 34 years in software, including 26 years in security solution design. He is the author of numerous white papers on security and regulatory compliance, as well as a McGraw-Hill book on identity, access, database, and application protection. Jeff is an expert on compliance requirements across a number of industries, and has presented at a wide variety of security events.