Transcripts For KNTV Press Here 20161225 : vimarsana.com

KNTV Press Here December 25, 2016

Scott mcgrew, this week i learned two things, one, fico, the Company Famous for the fico score, does not decide your Credit Rating. Who knew . And also fico is headquartered in Silicon Valley, about two blocks from this studio. Again, who knew . Will lansing is ceo of fico, i assume he walked over here from his office, hes past ceo of fingerhut, info space, joined by Mark Sullivan and laura mondaro of u. S. A. Today. Lets start with this idea that fico doesnt my fico score is so important. But somehow youre not the Credit Rating . The way it works is were an Analytics Company and we produce the algorithm and the score. But we dont have any data. So the data is held by a credit bureau, by someone else. And they use our algorithm to produce the score, which they then share with banks, thats how the credit determination is made. I have a macys bill, i dont pay the macys bill. Its expeerian or others who know i depart pay the bill and they tell you . Exactly. Were, a bank is trying to figure out whether youre likely to repay a loan if they give you a loan. So the best predictor for that turns out that if you have to pick a single data set, the best you could look at is whether youre good at paying things in the past. Sure. So a data set that provides that information is what we call tradeline data. Whether you paid your credit card bill on time. That turns out to be highly predictive of your propensity to pay debt and the fico score predicts your propensity to pay off debt. Its on top of this line data. Why dot Credit Bureaus tell you . In a sense the Credit Bureaus know they dont tell us, we take their data. We figure out a score for how likely it is that a consumer will repay debt in the future. And then they take their data and our score that theyve computed on top of their data and they share that with the bank and then the bank uses it to make a credit determination for you in the future. This calculation, i imagine its gotten more intense and more complicated over the years. I mean when you look up a fico score, its like whats your income, whats your payment history. What your debt or your credit ceiling. Are there more pieces of it now . Is it a more complicated algorithm . There are a lot of pieces. The most important thing in predicting Consumer Payment behavior is your past payment behavior. If youve been pretty good in the past, youre more likely to be good in the future. Its actually not really incomedriven. It turns out that theres wealthy people who are bad credits and theres poor people who are very good credits, it has everything to do with responsibility and payment behavior and not a lot to do with your wealth. So that sort of seems to have been the what youve done for years. Decades, right . The whole field, we hear a lot about this over the past couple of months seems to be anticipating what youre going to do. On your smartphone, what youre going to ask google and then after that, what google is going to tell you is predictive and Artificial Intelligence plays a big part of that. Are you, do you have a lot of a. I. Engineers trying to figure out what youre, what borrower is going to do in a year . Our greatest skill is predicting consumer behavior. We are all about understanding what a consumer is likely to do in the future. Historically, the people who are most interested in that were bigsix decisions like money decisions, should we give you a mortgage, a credit card . Increasingly, we made up lower costs to figure out how a consumer is going to behave. Now its interesting for marketing. Its interesting, what should we offer . How likely is a consumer to click on this link . For upsell, for crosssell. Youre not using my fico score, whether ill click on that link. Youre saying youre good at gathering data. Good point we have our scores business, where we predict consumer prepennsylvaniaty to repay demt and then our Analytic Software business, all about predicting all different kinds of things about consumer behavior. We look at past behavior and predict future behavior. Were with the analytics. Youre good at that. Were good at that, thats our main thing. Is it a lot more crowded market. It used to be most of this information was from payments, but now google or facebook or anybody who has the consumer basing websites have so much information and thats how theyre predicting everything. Where do you stack in. We dont have any of our own data, were just the math company that sits on top. What were really good at is taking whatever the data is, your customer data, thirdparty data, and figuring out what kind of decision you want to make with that data. We think of ourselves as a decisioning company. And its all about what decision you trying to make, whats the best data to predict that, the decision. And where the analytics that help you get from the data to the decision. And then we incorporate the decisioning into the work flows of the company. Scores are one way of doing that. Thats one small way of doing it. But were in all kinds of places. We have a big fraud detection business. So when your credit card gets evaluated for is it really you or not, thats typically us behind the scenes. We have a 90 market share in credit card fraud transaction, fraud detection. So in your, in your analytics business, can you name a reallife say a fortune 500 company that you work with to help them make some decision with your data science . Sure, i would say ten of the top ten banks in north america use our predictive Analytic Software extensively to make those decisions. Lending decisions, you mean . To make lending decisions. And some we have big retailers that use us for making marketing decisions. I see. If you are that good at, i mean if past behavior predicts future action, can you get any better at predicting whether or not ill pay that loan . I mean youve got more powerful computers, youve got a. I. , but it sounds like your system works pretty well now. Great point. Our system does work really well now, were proud of the fico score, which is the industry standa standard. Its all about fairness and transparency and access to credit and all of those great things have happened. Yet theres still people on the margin who are underbanked, who arent being recognized. Maybe you dont have your credit card. You pay your rent on time, but you dont have a credit card. Were developing new scores to solve for that. The other thing is you know i said we focus on the tradeline data. This credit card data. But if you mix in other data sets, youre going to get more information. Maybe a data set thats not as predictive as the credit card payment data. But still has some kind of caloric value. And a frequent guest on this show is doing something with a firm. Where he says im taking although he uses your fico data, too, doesnt he . Yes. But kind of what youre talking about this revolutionizing finding other ways of predicting how somebody who maybe doesnt have a great score, but there may be evidence that they would pay back that loan . Yes, very much so. In the old days you would look for a single data set that had the most predictive data. And it worked well. And we still use it. Today its much easier to mix and match Additional Data in. Rental data, utility payment data. Telephone payment data. Laura i interrupted you, lets take you with the last question there. This is all personal data. How do you incorporate sort of external events which can disrupt everything like the housing crisis . Theres a ton of data that allowed for the rise of option a. R. M. S and the explanation was we have all the technology to do that. And then there were many, many other factors which made all of those go under water or we do, we do revise and improve the score. And were, were on you know fico 8, fico score 8 is kind of the Industry Standard today. We have new fico score xt and 9, so were constantly improving and theres new data that comes into the picture. But theres like, is there like a black swan ability to you know you can predict say marks ability to pay. But if this external event, like the asteroid hits, he may not be able to pay that. How do you factor that in . That you got to get another way. Fair enough, will lansing is the ceo of fico, who knew, is a Silicon Valley company. Will, thanks for joining us. Thank you. Up next, inside one of Silicon Valleys most promising dna labs when press here continues. Every barracuda appliance, every barracuda piece of software is constantly being update. As new threats occur, were going to update our customers and allow them to be protected against those new threats. To learn more about barracuda networks, visit them at barracuda. Com. Welcome back to press here. There are Amazing Things happening in Genetic Research right here in Silicon Valley. Researchers and scientists working nearly around the clock. Beaten in their industriousness only by the technicians in the lab. Because the lab workers can work all day every day around the clock counsyls workforce works around the lab. Eric evans is one of the founders of. Robots lend themselves to this thing, because your requirements, youve got to be exact. Youve got to be, youve got to repeat the thing the same way every time, et cetera. So robots in medicine make a heck of a lot of sense to me. Thats absolutely right. So at counsyl were focused on providing dna information that can be used to make Important Health choices. And the advantage of having a robotic process is that you can have very high precision, and high reliability in the process. But it really goes far beyond that and we could get into that discussion. Id like to get into that discussion. You have x numb of robots now, where could robots work that you actually still have people working currently . I think its good to understand sort of how the lab is composed and what it can do. We have 36 people who work in the lab. Its not absent of people. But what robots help to do is to take a problem that requires a lot of road work and a lot of precision and then free people to supervise, to see what has been identified by say the robot in the Software Stack as an area needing troubleshooting and the humans can come in and look at those areas. Basically it frees scientists to focus on the area where theyre problemsolving and troubleshooting abilities can be best deployed. So the scientists are developing the tasks and the robots are going and doing them . At counsyl we have a number of different groups that are contributing to making this testing possible. One group is focused on really building the products, so we have software engineers, we have mechanical engineers, scientists. They build the products. And then in our laboratory, the Clinical Laboratory that serves patients and physicians across the country, they have specialized staff that have licenses that are required by the state of california to perform laboratory testing. Laboratory directors who are physicians who are experts in how to interpret the results, they work with the robots to perform the tests. What that permits is the division of labor between the robots performing tasks that require or are only possible with patients. So the ability to monitor every action thats performed to log it, to track it, to see if theres drift in the process. And then that leads the scientists to be able to lock at the data, to analyze the test results and deliver those test results to physicians. I have a question about just the genesis. I know several years ago we were excited about the human genome project and all the dna testing and some consumer face iing dna testing, 23 and me and some others. Weve seen the regulatory hurdles just be so steep. When you know, look back, was selling direct to consumers the right way . And what makes you confident that youre not going to have some of these same setbacks. One thing i want to correct, we dont sell any of our tests directly to consumers. Our customers are physicians who are treating their patients and want to have dna testing information to help them make choices that are going to direct their care. So for example, a woman who may have a positive Family History for cancer, could get a test to tell her whether shes a carrier for a gene like brca1. So were embedded in the traditional medical infrastructure. Very different from say 23 and me or other companies that are direct to consumer. Our physicians are sort of directly interfacing with us and ordering tests for their patients, we interface with the health plans that health insurers, so we have Extensive Health Insurance Coverage and we are doing all of these things to really supplement and expand the capability of physicians to offer these tests. Were there other more expensive, like more bureaucratic ways that they would have gotten these same tests . So whats changed is the number of things in the technology. Ten years ago we didnt have the ability to do nextgeneration sequencing and these other things that permit you to generate huge amounts of data from dna and now that these new tests are available, what has to change is the ability for physicians to offer these tests in the context of a 15minute appointment with the patient and for the patient to then be able to understand the test results. So we do a number of things, we provide Genetic Counseling Services to every patient so after a test is done they can contact one of our genetic counselors at no cost and talk about the test result and understand what their choices are. At the end of the day the test is all about making a choice available to a patient. Are there any medical ethicists involved when it gets time for the counseling discussion . So genetic counseling is a relatively wellestablished area in medicine. So they have their own guidelines about how to practice this area. And its really just a routine part of clinical medicine that folks only experience at particular times in their lives. So if you have a particular indication, that means you need testing, if youre planning a pregnancy, if youre already pregnant, then youre going to interact with this area of medicine. Eric, i want to ask you a question a little offsubject. But you got your p. H. D. At stanford, you founded this genetics company. You got funding. Elizabeth holmes, of theranos. I know its not exactly the same sort of thing, but its laboratories and funding. Got scads more funding than you did. When you looked at that did you think, wait a minute, ive got a ph. D. In genetics from stanford. Did you look at that and say how did this go her way and not my way . I think that were very different from theranos. You are. I admit from a laymans point of view. But you both have lab company. I think what it highlights is the importance of having sound Science Behind health tech, biotech, all of these areas. Need to be predicated on wellestablished science. You need transparency. Thats something that myself and my team is focused on. How we publish, putting it in venues where theres open access to the data. Being very transparent about what things we do and how we do them. I think at the end of the day, thats the right way for medicine to be practiced, the right way for science to be practiced. Has there been any aftershock from whats happened with theranos . I mean it looked like Venture Capital had a record year for biotech. But just the stigma perhaps attached to the general feel for startups . I think that if anything, its a lesson to Everyone Needs to Pay Attention to the details. And that its been an opportunity to contrast some of the hype with whats really very concrete, sound clinical medicine. And thats been sort of an opportunity. Eric, im going to jump in, ive got a couple of seconds left. You have a website in which you show some of the things you can do. But i dont get that stuff from you. I get it from my doctor. But is it as simple as talking to my physician about hey i saw something on counsyl. Com and im interested in getting this test. If you talk to your doctor, what theyre going to want to understand is do you have a Family History . Are you at the right time of your life where this is the right test for you . And doctors are going to talk to you about it. And then she can go from there and take it from there. Eric evans with counsyl, chief science officer, thank you. Good news in the battle to get women on corporate boards when press here continues. 19. 9 . Maybe the number went up because Theresa Briggs was apointed to the board of directors at deloitte. Shes made a point of helping other women find board seats in companies worldwide. Thanks for being with us. Though your research, i looked at that, too, says women on boards in tech is more like 18 . So even lower. Right. A little bit lower. What do you do to help people in general, and women specifically, get on boards . How, if i say to myself, you know id like to be on a board of directors, how do i go about doing that . You know so the first thing i would tell you is, you need to clearly define why you want to be on a board. What are the skill sets . What are the experiences you have that are going to make you valuable to a board. The second thing im going to tell you is, network like crazy. So some people think that board placements are similar to executive placements where executive recruiters play a big role. But in this field. Only about 25 of board placements come through executive recruiters. And 75 through the networks, of ceos and Board Members themselves. So networking becomes particularly important if youre trying to line yourself up for a board opportunity. Im curious, i talked to one of the founders of the board list, which is basically like a linkedin for just this purpose and interesting feature was that, men and women would endorse. So put forward somebody. And that whole interface was zised to service more women. Can tech sort of solve the problem . I think its a really interesting model. What im seeing be more effective today is more persontoperson relationships. I had a

© 2025 Vimarsana