Julie May, VP and GM for B2B Scores at FICO, and Ethan Dornhelm, VP of scores and predictive analytics at FICO, join theCUBE’s Rob Strechay at FICO World to explore how FICO’s scoring models are evolving in today’s lending environment. Their conversation focuses on post-pandemic credit trends and the technologies reshaping lender decision-making.
May and Dornhelm examine how consumer behavior has shifted and why national FICO scores remain remarkably stable during economic disruption. They also break down how trended data models such as FICO 10 T are changing the way lenders assess and manage risk.
Their discussion also highlights the growing role of consumer education and new data sources such as UltraFICO and “buy now and pay later” activity. These innovations aim to increase financial inclusion and sharpen the predictive power of credit assessments.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
FICO World 2025. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Register For FICO World 2025
Please fill out the information below. You will recieve an email with a verification link confirming your registration. Click the link to automatically sign into the site.
You’re almost there!
We just sent you a verification email. Please click the verification button in the email. Once your email address is verified, you will have full access to all event content for FICO World 2025.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
OR
Connect via Twitter
Connect via Linkedin
EDIT PASSWORD
Share
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
FICO World 2025. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Sign in to gain access to FICO World 2025
Please sign in with LinkedIn to continue to FICO World 2025. Signing in with LinkedIn ensures a professional environment.
Are you sure you want to remove access rights for this user?
Details
Manage Access
email address
Community Invitation
Julie May & Ethan Dornhelm, FICO
Julie May, VP and GM for B2B Scores at FICO, and Ethan Dornhelm, VP of scores and predictive analytics at FICO, join theCUBE’s Rob Strechay at FICO World to explore how FICO’s scoring models are evolving in today’s lending environment. Their conversation focuses on post-pandemic credit trends and the technologies reshaping lender decision-making.
May and Dornhelm examine how consumer behavior has shifted and why national FICO scores remain remarkably stable during economic disruption. They also break down how trended data models such as FICO 10 T are changing the way lenders assess and manage risk.
Their discussion also highlights the growing role of consumer education and new data sources such as UltraFICO and “buy now and pay later” activity. These innovations aim to increase financial inclusion and sharpen the predictive power of credit assessments.
Julie May, VP and GM for B2B Scores at FICO, and Ethan Dornhelm, VP of scores and predictive analytics at FICO, join theCUBE’s Rob Strechay at FICO World to explore how FICO’s scoring models are evolving in today’s lending environment. Their conversation focuses on post-pandemic credit trends and the technologies reshaping lender decision-making.
May and Dornhelm examine how consumer behavior has shifted and why national FICO scores remain remarkably stable during economic disruption. They also break down how trended data models such as FICO 10 T are...Read more
exploreKeep Exploring
What are the current concerns and potential risks surrounding the national FICO score and student loan delinquencies, especially in light of recent changes in consumer support programs?add
What does the T in FICO 10 T stand for and how can it help lenders increase origination volume while holding credit risk steady?add
What recent developments has FICO announced regarding UltraFICO and buy now, pay later loans?add
>> Hello and welcome back to FICO World 25. We're here in Hollywood, Florida really digging into the intelligence revolution and really understanding everything about FICO and their partners and their customers, and really how it's being utilized out there in the wild. I think one of the things you probably can't come to FICO World and not talk about is FICO scores and nope, I don't know that I could have two better people... There probably aren't two better people, I'm just going to say it like that, to help me really break this down and kind of understand this and really go a little bit deeper. We have Julie May, who is the VP and GM of B2B Scores for FICO. Welcome on board.
Julie May
>> Thanks for having me.
Rob Strechay
>> And we have Ethan Dornhelm, who's the VP of Scores Analytics for FICO. Welcome on board.
Ethan Dornhelm
>> Thank you.
Rob Strechay
>> So Ethan, I asked this kind of in a different way to one of the other contestants who was up here earlier in the day, and I think that everybody kind of draws a line at the pandemic and says, "Okay, things rapidly changed after the pandemic, especially in the world of data and information." What has really been the trajectory of risk since the pandemic?
Ethan Dornhelm
>> Yeah, it's a great question. Thanks, Rob. I would say one of the ways we like to look at consumer credit risk trajectory is with the average national FICO score, it makes sense to use that as a metric because it is the common language of credit risk in the US, 90% of top lenders use it. And what we saw through the pandemic, the first year of the pandemic, we actually saw a significant uptick in the average score, significant. And that was driven, we think, by lender payment accommodations that were being offered at the time, government stimulus payments. This all allowed consumers to save at unprecedented rates, pay down debt, really help their balance sheets and in turn help their FICO score. What we've seen since is really what I would call a treading water. We've really been stable with the aggregate FICO score in the four years since that real uptick in the first year of the pandemic.
Rob Strechay
>> Wow, that's good to hear. I think, again, everybody is... I think that is a good way to look at it as well. And I think that when you're looking across all of the credit landscape, what are some of the trends that you're watching for from that perspective?
Ethan Dornhelm
>> Well, I think, as I said, we've been stable for the last four years in terms of the national FICO score. The real question that's out there right now is, are we at some sort of inflection point? I don't think you would find anyone who would argue that the risk factors are out there higher than ever for a recession. There's this persistent inflation that consumers are facing. Some of the last supports that were put in place around the pandemic for consumers are being rolled back. I think all of that leads to this big question of, what's next. And in that vein, one of the topics that's really hot right now is student loan delinquencies. So you're probably aware, during the pandemic, there was a pause placed on all consumers who had federally backed student loans to have to pay those loans back. And only recently those consumers who aren't paying, it's being reported in their credit file and starting to impact the score. In fact, early this year, we saw the score drop in February, the average national FICO score, and that was the month that four million consumers had their student loan delinquencies reported for the first time.
Rob Strechay
>> I was going to say, I know I'm not delinquent on mine, on my kids that is. But I got the letter and all of that, and you had to log in and do all of the things and make sure everything was set up. But I think you brought up a good point, and I want to bring Julie in on this because I think to me, the lenders were doing stuff during the pandemic that they've since ended and stuff like that. But to me, the FICO scores give them insights into what's going on, what's really top of mind with those lenders and in the area in particular of credit risk.
Julie May
>> Yeah, I mean, I think at the end of the day, lenders are always looking for more information to make better decisions so that they can provide credit to their customers and that they can expand their business while actually helping people. And so what we're finding top of mind is what data can they use and how can they use our scores or what scores can they use to actually help them expand credit access. And so particularly we're finding that our customers are interested in hearing about our latest and greatest score, FICO 10 T, the T in FICO 10 T stands for trended data. And so what this enables our customers, the lenders to do is to actually understand across time how people's balances are changing if they're consolidating debt. And the type of impact that can have to them is it can actually increase origination volume for lenders while holding credit risk steady. So we're getting a lot of questions about this. There's a lot of interest in using this additional trended data in credit risk assessment.
Rob Strechay
>> So that would seem like, again, to your point about having more information to go and act upon and come up with, that can help them maybe go a little bit deeper or have, and I love this segmentation of one that was part of the keynote, is that really where this 10 T score goes and helps people with that?
Julie May
>> Right, I mean, exactly. The FICO score is unique to every individual in terms of the algorithm is an algorithm, but it looks at every individual separately associated with making assessment of their ability and willingness to pay back debt and makes that assessment based on data that we receive to calculate the score from the credit bureaus. And so the more data you have in that process, the more precise that you can get in the assessment of risk. And so we see lenders are incredibly excited about this, particularly we're seeing significant uptake in interest, and I can't walk through the hallway here at FICO World without having someone stop me and ask me questions about it, but particularly we're seeing tremendous interest in the mortgage lending space. And so we've had incredible adoption in mortgage by the non-conforming mortgage lenders of 10 T. We have approximately 30 lenders who are already using it, over 300 billion in originations are being scored with 10 T, and 1.5 trillion of servicing volume is being scored and reviewed by 10 T. And so when people out there hear about others using it, they want to get on board and they want to hear about, how can they use it, what can they do, how can they implement this in their process? So there's quite a buzz out there about it.
Rob Strechay
>> Yeah, I mean I think it makes sense. I think, again, this is one of the wonderful things about AI is it allows you to manage all the data and those models are behind the scores really, again, that makes it easy for these lenders to have that information and things. But what are some of the approaches that you've seen that are helping lenders make smarter, more nuanced decisions like 10 T and where do you think that's really shaping or how customer behavior is really shifting, is really making them reach out and look for these types of things? I'll start with Ethan.
Ethan Dornhelm
>> Sure. I would start by saying when it comes to credit risk, the rule of thumb I've always followed is the more things change, the more they stay the same. We built our first FICO score over 35 years ago, and when we built that, we found consumers who had higher debt levels, consumers who were showing signs of mispayments, consumers who were applying for a large amount of new credit, those were the folks who may be headed for trouble. Guess what, 35 years later, it's still by and large those same key drivers. So I take some comfort from the fact that while the products that are out there are certainly changing a great deal when it comes to predicting credit risk behaviors off of credit bureau data, those overarching patterns that held 35 years ago still hold today.
Julie May
>> What I would say is, I think the things that maybe have changed a little bit in the last 35 years, I don't know if I was around 35 years ago, but I kid. I think the things that have changed are consumers are more aware now. We have a lot of tools and capabilities available for the consumer to understand how important their credit score is to understand what their credit score is. One example of that would be anyone can go out to MyFICO.com and get their credit score, their FICO credit score for free. And the awareness and understanding and knowing what it is and why it is that, and the building blocks of credit really help enable consumers to better manage their credit. And so I think that's changed in 35 years because those tools were not available previously. And so we see tremendous adoption of those tools. We see consumers coming in, really being educated on what their FICO score is, and that only helps the consumer, but also the lender as well.
Rob Strechay
>> Yeah... Oh, sorry.
Ethan Dornhelm
>> Oh, I was just going to build... Julie's got a great point. There are definitely things that have changed, maybe less so what drives credit risk behavior, but the availability of data has certainly changed. We've seen the rise of alternative data, something that at FICO we've been innovating around for well over a decade now. We've got the UltraFICO score, which leverages consumer-contributed cashflow data. 35 years ago, there was no such concept of the internet, let alone the idea that a consumer could give their bank account credentials, offer up their DDA cashflow data, and in so doing improve their FICO score, which 7 out of 10 consumers we found do with UltraFICO.
Rob Strechay
>> Yeah, I've actually seen that and contributed to that. So glad to hear that product is doing well.
Julie May
>> 35 years ago you'd been mailing your paper bank statement into FICO, I don't think that would've worked.
Rob Strechay
>> Yeah, no, I mean the whole API economy and how things can... The data, and I think it seems like that consumers want to know as well more so. Are you seeing that, that the lenders, they're going and giving that data because I know my different financial institutions provide me my FICO score once a month and things like that. Are you seeing that that's also helping that consumer behavior as well where there's more awareness of it from that perspective?
Julie May
>> I think the tagline is the more you know. My mother always used to say, "When you know better, you do better." And I think that's actually true in credit too, when you know better and you understand what contributes to your score, you can do better. You can build a good credit profile, you can understand what impacts it. And so I think her advice is good for all the people who are wanting to access and build their credit profile today. So yes, I think that's true. We have lots of different programs through lenders to make the scores available through their online channels and other different ways that they make it available. Then we also make it available through myfico.com. There are so many different channels now that consumers can get access to their FICO score and educational information on how to drive, improve, maintain their FICO score, that I think that's really evolved over the last couple of decades, and it's really exciting. We have a much more informed consumer than we used to.
Ethan Dornhelm
>> Yes, and Julie quoted her mom, I'll quote my grandma, who said, "Knowledge is power." And I think we really have found that an educated consumer is an empowered consumer when it comes to knowing their FICO score and how they can improve it.
Rob Strechay
>> I would agree. I think to me, it's definitely one of those things in information and the information sharing. But one of the things, and this is right up your alley Ethan, is that the fact that the increasing role of AI in data analytics and the amount of data and the difference of the data, and like you were saying, it's not all the same data and more and broader, how is all of that used in the development of the FICO scores, the different scores that you have?
Ethan Dornhelm
>> I'm just surprised it took us this long into the interview for you to mention AI, Rob.
Rob Strechay
>> Well, I mean, I think you guys have used traditional AI for a long time, and I use the air quotes around that because old enough to know and have a patent in one of those for quite a while now, and it's not all Gen AI. So yeah, help us understand how that's used, AI really contributes to that, the scores.
Ethan Dornhelm
>> [How we use AI in the FICO Score development, I would say very carefully. And the reason that's my answer, I've gone on record saying this before, I think the FICO score is the most scrutinized predictive analytics in the world. We already talked about how millions of consumers are monitoring their score every month, seeing how their behaviors translate to changes in the score. We know that lenders are certainly scrutinizing the score, they're using it in billions of lending decisions every year. So the importance of a transparent model, the importance of an explainable model is at the very top of our priorities when we build the FICO score. And what we found is that some of these state-of-the-art AI, I think when a lot of people talk about AI, they're thinking about neural nets, they're thinking about stochastic gradient boosting, those sorts of techniques. We found that we're still not quite there as far as cracking open the black box to the extent that we feel it delivers on that transparency that's necessary for the FICO score. So the way we use AI is as a really helpful, fast, powerful assistant, let's say, providing benchmarks to us of, what's the ceiling of predictive lift and how close to that ceiling are we able to get? And the answer is very close, with some of our very complex but very transparent and explainable techniques that we use to build the FICO score.
Rob Strechay
>> Yeah, I think to me, I think explainability and responsible AI has been definitely a theme that I've heard here at FICO World this week and I would say that that makes a lot of sense given what you're dealing with and how emotional people can get things like their FICO score and FICO scores for that matter.
Ethan Dornhelm
>> And again, there are great strides being made in explainable AI. The standard is so very high for the FICO score in particular for explainability.
Julie May
>> At the end of the day, lenders trust the FICO score. They trust that it's going to be robust over time, that it's going to be consistent over time. So we take the responsibility of that very seriously because we know what it impacts in the ecosystem. And so it's something that Ethan and his team do an amazing job of managing and making sure that we can continue to be that standard of trust in the industry.
Rob Strechay
>> Yeah, I've even heard blockchain used about being able to stamp the models and things of that nature and understand the lineage and all of that, which to me goes back to, hey, how do I go to a lender and be able to say to them, "Hey, you know that this is where it is and here's the different revisions and things of that nature," and really making them mutable.
Ethan Dornhelm
>> But to your example, a powerful assistant.
Rob Strechay
>> Correct.
Ethan Dornhelm
>> Never in the driver's seat. I think that's a key philosophy we have with our scores is there will always be a human not only in the loop, but with their hands firmly on the steering wheel.
Rob Strechay
>> Yeah. No, I think that's a great segue into, Julie, let me ask you, the FICO Score team seems to be always innovating. I think, again, you're giving me the feeling that you're trying these things and making sure, which I love that, hey, when it's ready, it's ready when it's fully baked. What are some of the things that you see coming on the horizon for FICO scores?
Julie May
>> Yeah, so I think Ethan talked a little bit about UltraFICO, and that's using consumer permission and DDA data to actually provide a different lens of borrower credit readiness, and so we're excited about that. We have new developments coming out associated with UltraFICO. But also we recently published a study associated with buy now, pay later loans, so BNPL data with a partner, a firm, which is a very large BNPL lender, and we were really excited about the results of the study. With that result of the study, we also announced that we will have a product announcement coming very soon. I can't give you the scoop on it today, but it is coming very soon. And we'll be making and adding a proprietary treatment to FICO models to address by now, pay later information that will be furnished through the credit bureaus. Why this is important is we've seen incredible growth of utilization of buy now, pay later as a lending vehicle, we see consumers using it. We know that we need to be able to understand and use that data appropriately in the assessment of the FICO score. And this study allowed us to take over a year to take a significant volume of data from our partner firm, and really assess the impact of buy now, pay later information when furnished to the credit bureaus on the FICO score. And Ethan and his team took the challenge of basically analyzing that data and then understanding how that should be incorporated into the FICO score. So I'll let Ethan tell us a little bit more about the stats and the details on that because he's more appropriate to do that, but we're excited about the coming soon product announcement.
Ethan Dornhelm
>> Yeah, thanks, Julie. I mean, what I'd say, first and foremost, what we realized was simply feeding this new type of data into the existing algorithms, which were calibrated on traditional creative products is not the answer. So there is, as Julie said, a nuanced treatment that's required if you're going to feed this data in. And at the same time, again, through our innovation, through our creativity, we found a way where we can capture some of the signal that exists in this BNPL data to avoid lenders having a blind spot around the fact that consumer has this type of debt without having it inappropriately and adversely impact consumer's scores. So we found, especially for those consumers who are opening five or more BNPL, which is not uncommon, whereas it's unheard of for a traditional credit product to open five new personal loans in a month, we found that actually the majority of them with this enhanced treatment either had no score change at all, or had a positive score increase as a result of their positive behaviors on those BNPL loans.
Rob Strechay
>> Well, that's awesome. And I think that'll be something people will love to hear, and I think that's a great place to end this because I think, again, understanding that that can be a positive note to folks that are out there using those and that they, again, you're helping illuminate that so they get the good credit for that as well. So thank you both for coming on board, and this has been... No wonder you've been getting stopped. I mean, it's not shocking to me, the FICO scores are what everybody knows, so I'm not shocked that you are the two rock stars walking around here this week, or two of them.
Julie May
>> I'll take it. Yeah.
Rob Strechay
>> Well, thank you for coming on board.
Julie May
>> Thank you.
Ethan Dornhelm
>> Thank you, Rob.
Rob Strechay
>> And thank you for watching this episode of FICO World 2025. We're here in Florida, and we'll be back shortly. Stay tuned.
Rob Strechay
