Years ago, I saw a coffee commercial in which a guy sits on the balcony of his lake house, looking out on the water, drinking his Instant Garbage, and hovering over his typewriter (like I said, years ago). In said typewriter is a sheet of paper on which he’s begun with the words “Chapter One.”
Okay, so only a moron writes a book this way. You need an outline. Some notes. Novels don’t spontaneously combust from your fingers. And of course, with a typewriter, there’s no real-time editing. You can only copy and paste using actual paste. So how should this process really work?
About the time I started using computers, I read a book called “Zen and the Art of Motorcycle Maintenance.” It’s a bit of a slog, but very rewarding. Besides being a great story, it offers an incredible creative lesson: don’t try to be perfect from the start. Writing, drawing, whatever. Make a mess, then clean it up. This is how I’ve written four books, lots of white papers, user documentation, and piles of blogs and letters to the editor. Spill the source materials on the floor, then reassemble them, polish them, make them fulfill your purpose.
This is how we’ve built the best identity verification platform on the market.
First of all, how does somebody digitally apply for a loan, an account, a credit card? They open an app and feed it their name, email, phone, and typically several other pieces of PII. Not all FI’s collect the same data, but the basics are always there. When you put your name in the hat for approval, you might think you’re a person. But to whoever is deciding on that approval or denial, you are the sum of your parts, a bunch of bits spilled on the floor.
And therein lies the problem with most identity verification products out there. They look at the parts and often deny applicants based on parts alone. Some of what you enter looks good, but just one thing out of whack might lead to rejection. You’ve only lived at that address for a couple of months. Your condo building is dual-use because there are retail shops or the homeowners association office on the ground floor. Your phone isn’t old enough. So even though everything else checks out, the app becomes a false positive. Bye!
Banks, lenders, remittance companies, fintechs, they stack up piles of point solutions to validate the individual pieces of PII, then have to write logic that ties together. If one solution likes your phone, it passes off to the solution that checks your email. Then your address. But what if one of them fails the test? How good is that logic? Can it pass only the goods? How many bads slip through? How many goods get rejected for the wrong reasons? It’s a mess.
So how does Socure decide if somebody’s good or bad? Again, start with that mess, a pile of PII that may or may not check out. First, we do have to look at those parts. But we go way deeper than the others, who only perform the basics. Is your phone or email on a list of bads? How old are they? Blah blah blah. And sure, that stuff still has to get done. But then Socure looks at literally thousands of other possible indicators of good and bad, data science-y stuff that I can’t talk about during parties without getting kneecapped by our geniuses.
After we look at the parts, and determine if your phone, email, address, etc. are legit, we put them together and look at the whole. Do all those good parts belong together? They might be individually okay, but they also might be stolen. I’m pretending to be you (third party fraud), or I’m pretending to be an invented identity (synthetic fraud). The sum of the parts must also check out. We are taking the mess and cleaning it up, by clustering those pieces into a holistic identity. This gives us a picture clear enough to say, this applicant is truly who they say they are, and therefore trustworthy, or not.
“Wow, Jeff, tell me how you do this.” Well, I’m glad you asked. When we build our engine, we do the exact same thing: make a mess and clean it up.
We leverage literally billions of rows of data, some raw, and some of it feedback data, i.e. the outcomes of the massive bank of transactions we process on a daily basis. Our automated platform ingests and cleanses that data, makes sense of it, discerns the patterns of goods and bads. This automation allows us to get more accurate more quickly than anybody.
We don’t make assumptions and decide how to decide. We let the data tell us how to decide. We take the lessons of that mess-cleaning and make order out of chaos, in order to provide our customers the information they need to make the best decisions on their own customers. Keep more good applicants, dump more bad ones.
At Socure, we train our market-leading verification engine on how to process these consumers as a whole. By replacing all those point solutions, we clean up the mess and provide fast, accurate decisions.
Give us a shout. Let Socure help you find your own zen for id verification.
Jeff Scheidel is a technologist with 38 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.
Escalating Fraud Solutions at Money20/20
Money20/20 this year is pretty huge–with nearly twice the attendance as...
The Business Impact of Identity Verification and Fraud
Economic, business, and social systems operate according to the trust that...