Thought Leaders in Cyber Security: Rich Kahn, CEO of eZanga (Part 2)

Sramana Mitra: What I’m trying to understand is what is the philosophy. What drives this cleansing? What are the things that you’re looking for? What kinds of issues are you trying to cleanse out in that process?

Rich Kahn: What you’re asking for is the secret sauce. I’ll be honest with you. If I said to you, “We are looking for the color purple floating through our click stream because we know that the color purple is a bad color on our network.” The people who are writing this fraudulent technology know to not make their traffic purple. They get around your system.

In the media recently, the biggest thing that people have been talking about is ad view-ability. Basically, they want to make sure that their ads are viewed by a certain percentage of people. Otherwise, they consider it fake traffic. Now, people who are writing this code have gotten around that particular rule. We have an extensive rule set that consists of somewhere between 30 to 40 different rules that are used to identify the traffic as real or not real. We look at a bunch of analytics. We look at a bunch of behaviors. There’s a lot that we go into.

Sramana Mitra: It’s not going to be interesting if you can’t provide any value in the interview. Just saying that we cleanse traffic is not interesting. I understand that you don’t want to give your secret sauce.

Rich Kahn: I understand. For example, when you talk to a security company, they don’t tell you exactly how they stop people from hacking systems.

Sramana Mitra: Everybody knows that there is click fraud out there. The only reason they would read a thought leader in cyber security story is to learn something. I imagine this is not your first media interview.

Rich Kahn: No. In fact, I just got back from talking with Red Herring about this technology.

Sramana Mitra: What did you tell them that’s interesting?

Rich Kahn: Ultimately, the things that we’ve done over the years. Most of our stuff is behavioral. We’re looking at the behavior of users. We have a number of different things to identify behavior by looking at how traffic behaves once it leaves our network. We look at how traffic makes it over the client side and what it does. Ultimately, that’s what counts. It doesn’t matter which IP address you’re coming from. There’s some value in some of that stuff but that’s so old school. Most of the stuff we do nowadays, at least on our network, is look at specific behaviors of the clicks.

Let’s say you have a site and you’re advertising divorce attorneys. Somebody comes to the divorce attorney. They read some information and click on some more details about divorces. That’s a normal flow that you would expect. We’ve seen traffic where it comes in to a site where they’re looking for a divorce attorney. The very first thing that they click within five seconds is on bankruptcy because it’s one of the first links on the page. If you’re getting divorce, you might eventually go through bankruptcy but that’s not the normal flow for that particular site. By looking at behavioral analysis, we can identify real traffic versus fake traffic. That’s ultimately what we do.

This segment is part 2 in the series : Thought Leaders in Cyber Security: Rich Kahn, CEO of eZanga
1 2 3 4 5