Expanding Access: How Alternative Data Helps Evaluate Thin Files and Drive Financial Inclusion

In this episode of Credit in Focus, we explore how alternative data is helping lenders expand access to credit for thin‑file and credit‑invisible consumers, without increasing risk. Drawing on research from over four million credit card applications, we break down why nearly a quarter of applicants lack a traditional credit footprint and how solutions like RiskView™ make 95% of them scorable.

Join Zach Tondre and Kristin Carlson-Vinjamuri as they unpack key findings, including how positive alternative data can double the likelihood of becoming scorable, improve future credit performance, and drive a 12% increase in approvals. They also explain why financial inclusion and risk management don’t have to be trade-offs—and how lenders can say “yes” with greater confidence.

Tune in to learn how smarter, more complete data is transforming credit decisioning and unlocking growth for lenders and consumers alike.

You can reference our data discussed in the episode here.

DISCLAIMER: The information provided in this podcast is for informational purposes only and is not intended to and shall not be used as legal advice.  The views and opinions expressed in this podcast are solely those of the speakers and do not necessarily reflect the views or positions of LexisNexis Risk Solutions. LexisNexis Risk Solutions does not warrant that the information provided in this podcast is accurate or error-free.

LexisNexis and the Knowledge Burst logo are registered trademarks of RELX Inc. Other products and services may be trademarks or registered trademarks of their respective companies. Copyright© 2026 LexisNexis Risk Solutions.

Transcript

Welcome And Why Credit Invisibles Matter

SPEAKER_01 0:01

Welcome to Credit in Focus, a podcast series by LexusNexus Risk Solutions that unpacks the global complexities of Credit Risk across the customer lifecycle, from marketing and origination to account management, collections, and recovery. Let's get into it. I'm a director of market planning here at Lexus Nexus Risk Solutions, and I focus on our Credit Risk Solutions. And along with joining me here today is Kristen. Kristen, why don't you introduce yourself?

SPEAKER_00 0:38

Yeah, hi everyone. Nice to be here with you, Zach. My name is Kristen Carlson Binjamuri, and I'm a senior data scientist here at Lexus Nexus Risk Solutions. And I work on our team called the Information Hub, which is all about doing in-depth research, looking at consumer trends, business trends, fraud trends, all of the above. And today I'll be sharing one of our most recent research works with Zach today.

SPEAKER_01 1:03

So just setting the stage for today's conversation, nearly a quarter of credit card applicants are credit invisible at the time they apply. For many consumers, it's not necessarily that they are more risky. They just really have no footprint inside the traditional credit reporting system. These FinFile and no-hit consumers tend to be disproportionately younger. They may be new to credit or new to banking, maybe because they're new to country or they just haven't engaged with traditional banking in the past as they've kind of been more of a part of the cash economy or using alternative payment methods. And when we think about traditional credit alone, kind of the Bureau trade line data, when used by itself, it can unintentionally exclude some of the borrowers inside this set that are actually credit worthy. And so in this episode today, we're just going to explore how alternative data, which is defined as data that sits outside of the traditional credit reporting ecosystem, how this alternative data helps lenders evaluate these thin files and expand access to credit without adding additional risk. So with that background, I will turn it over to Kristen to jump into the data.

Study Design And Scoring Definitions

SPEAKER_00 2:17

Thanks, Zach. Yeah, well, I'll go into describing the methodology of this study so that we all have the right information to understand the insights that I'll be sharing. And what we wanted to do was take a really large sample of credit card applicants, four million in fact, and track both their traditional uh score at time of application and follow that one to two years later. But with the key difference being that we're looking also at their alternative data at time of application, because that gives us a different read and a more full picture of that applicant. And from there, we get a really interesting picture of what alternative data can help us understand early on at time of application and where that applicant is headed in their traditional score as time goes on. So the methodology for anybody that is watching along with this podcast, you'll be able to see some of the content. And yeah, we have a research question here that is also really important, which is there is this entire credit spectrum and decisions on the far ends of that credit spectrum, the super primes applicants and deep subprime applicants, from a lender's perspective, when their traditional score indicates that they are on that sort of extreme end, it's going to be an easier decision. But it's really more of the middle ground where things can be not as, you know, more information is going to be very helpful in determining is this applicant truly at a prime, a subprime? How about credit invisibles? So that's also part of this study is to really look at the full credit spectrum and where alternative data at time of application helps predict where they are headed from a traditional standpoint, data standpoint as time goes on. And one of our measures of performance that we use to say, you know, uh, are these applicants, where are they headed with their delinquency rates as time goes on? We look at the 90 days past due rate. Um, and then we also I wanted to, you know, help us understand what is what are we defining as a traditional score? I also wanted to explain the traditional credit score that I've been referencing. What we're using here for research purposes is an internally developed score using trade line data from Innovus, and we build it so that it mimics sort of a uh FICO or vantage cross-industry score. And in this score, a 616 above is considered a traditionally prime applicant. And in contrast, our alternative risk view score is um a 600 or above is considered prime here, and that aligns the score to odds ratio. So just a little bit more background on what we're using for a traditionally uh prime applicant versus a risk view prime applicant. All right, I think I've set the stage for the entire methodology. We always got to get that out of the way. And now we can transition to talk more about the background of what we wanted to do with this study. Let's head to the next slide, which I think gives a good um illustration of that. And if anyone's uh if you're watching online some of this content, I think there's uh a nice pie chart and some other information and infographic here that helps illustrate some of the points that I'm going to make. Um so we hear the main business problem is that there are a lot of decisions on whether it's credit cards or other lending applications that need more insight beyond just traditional trade line information. When we look at a credit spectrum, there are gray zones within that credit spectrum. Certainly on the ends of the credit spectrum, say if you're in the super prime range or you're in the deep subprime range, those are easier decisions to make from a lender's perspective because the data, the traditional data, is clearly pointing where you're at from a credit risk perspective. But if you're an applicant that falls into, say, the subprime or prime credit tiers, um, those thresholds can, you know, move, and so can a consumer and their their um their information that conveys that credit risk. Similarly with the credit invisible population, these are all parts of the applicant pool where uh more information, more data, like alternative uh credit risk data is going to be very helpful in more definitively understanding where they those applicants.

What Alternative Data Signals Look Like

SPEAKER_00 6:57

Something I uh wanted to describe was when we took that population of 4 million credit card applicants, we looked at what were their alternative risk view uh data that really helped determine um their subprime versus prime status. And when we compared those groups, we found a few key groupings of attributes that were indicative of a prime applicant. And so what we found was that property ownership rates among prime applicants were four and a half times as likely compared to that subprime applicant. Similarly, if the applicant owned a home, um, those home values were actually 70% higher than a subprime applicant who owned a home. Um, and then college evidence. This was another big factor in um those that are prime, the the prime applicants. They were 40% more likely to have college evidence on file. And then finally, we look at derogatory records, including liens, judgments, and bankruptcies. And this is where we see um that risk view prime applicants were five times less likely to have derogatory public records compared to subprime applicants. That makes a lot of sense given, you know, subprime versus prime factors. So wanted to share some of the types of data, alternative data that go into a risk view alternative data score. And these were the four groups of factors that really stood out in our research.

Rethinking Subprime With Better Signals

SPEAKER_00 8:28

All right, let's look at the subprime applicant pool and look at how risk view alternative data can further segment the subprime group. Now, here we find that 11% of applicants were subprime from a traditional data standpoint, meaning they had a traditional score between 581 and 660. And at that time of application, they're subprime, but alternative data also at that point in time for a large portion actually points to a different story, that that um applicant from an alternative data standpoint is actually prime. And that's for some of the reasons that I just listed earlier about homeownership rates, um, college attendance, uh, property values, that sort of thing. What we find is that of the applicants where their alternative risk view score disagreed with the traditional data, it said, no, in fact, this applicant looks to be prime, not subprime. When we track their traditional scores a year later, two years later, we see that their traditional score increases, um, indicating, you know, a more trending towards more traditionally prime. And so alternative data early on at time of application helped um find that those applicants that actually show data that speaks to where they're headed from a traditional standpoint. And when we look at delinquency rates a year out from that time of application, we actually see that where alternative data disagreed with the traditional score and said, nope, these are actually prime applicants based on alternative data. Well, their delinquency rates are half that of where risk view indicated, yep, this is a subprime applicant. And so if you're able to see some of the content we have going along with this podcast, we've got great illustrations of these trends, but it's pretty easy to describe when delinquency rates are half that of the one group compared to the other. All right. One other thing I wanted to mention is that with that 11% of the applicants from our population that were subprime, it was 63% actually showed that their alternative data indicated that they were in fact prime, not subprime. And that's a very large share. And it just speaks to the potential for those in the subprime credit tier for that upward mobility, for them to get the credit offered in the first place when their alternative data speaks to their prime background or for more competitive credit terms. Zach, I'd like to pull you in here. I'm sure that you've got some additional thoughts.

SPEAKER_01 11:17

Yeah, I mean, I think when we look at this population, the traditional credit reports would call subprime. Um, I think that we're really seeing a couple categories. And so to kind of personify these two lines, if you're following along with the uh with the slides, we really have a segment of consumers who are maybe um habitual bad users of credit. And so those are kind of those those consumers that uh we show as subprime and really risk view agrees with the traditional scores that these folks are for subprime. And we can see that over time, one year later, two years later, their scores continue to go down, their traditional scores. Um and that's kind of compared to consumers who maybe are good managers of credit but suffered a credit damaging event. And so with some of these consumers where the traditional score disagrees with the risk view score, you know, we could um we could guess that maybe this consumer suffered some kind of event that damaged their credit, whether that was a medical event that that um drained cash or a car accident, um, something that that caused them or a loss of job, something that caused them to have a decline in their credit. But, you know, foundationally, these consumers are still solid. And risk view kind of shows um that difference there. Uh, and when we can see that that consumer kind of foundationally is still solid, has that ability to repay and willingness to repay, um, then over time we actually see their traditional score improve. And so I think those are really the two different segments that we're talking about. Um, and the industry is talking a lot about how to differentiate credit damage individuals from just kind of, you know, your subprime uh habitual bad users of credit. And I think this really illustrates that.

SPEAKER_00 13:05

Thanks, Zach. I knew you'd be able to take this out of the research uh mode and into really illustrate the point with some other um key pieces of information. So thanks for for that. Yeah.

Prime Applicants Who Quietly Deteriorate

SPEAKER_00 13:17

I wanted to also talk about the prime part of the credit spectrum because we did this analysis looking at three parts of the gray zones that we identified in the full credit spectrum. So here the prime uh group in within our applicant pool, we found 17% of our applicants were credit prime. They fell into a traditional score of 661 to 720. We did this exact same analysis of saying, well, what does their uh alternative data say at time of application? Does it contrast what traditional data says? Here we found 21% of the applicants that are considered prime actually had a risky alternative score that disagreed with the categorization of being a prime. And that when we follow, so we say, all right, well, here are the here's a group of applicants whose alternative data says that you know, or something that's not looking so good in their background. Um, maybe it's some of those attributes we talked about earlier, um, like property values or derogatory events, but something's indicating that in their future, their traditional uh score is going to fall. And what we see is that yes, those applicants where their risk view scores indicated, hey, this is a subprime applicant. They're not prime. Well, their traditional scores fell one year later by about 17 points. And then by two years later, their scores had fallen by 25 points. So their traditional score fell 25 points two years later, since time of application, when their risk view alternative data indicated early on, no, this is really a subprime applicant. And similarly, what are the aspects going into that traditional score? Well, it's going to be delinquency rates, and that's absolutely where we see a difference because that group shows more than double the delinquency rate just one year later compared to places where risk view data and alternative data said, no, we we agree. A traditionally prime applicant, uh, yes, they look prime from an alternative data standpoint, too. Um, and so those are the two groups that we compare throughout uh this study is just simply breaking them out by what does the alternative data say to add to the full picture of this applicant? Because sometimes that data agrees, sometimes it disagrees. And just the disagreement part is really where we hone in on this discussion. And um I think what this represents is that swap out potential. Um, these are applicants who their traditional data says they're prime, but they've got aspects that indicate, hey, maybe maybe the credit terms offered should be a little more stringent. Um here again, I'd like to pull you in, Zach, on any additional thoughts you've got from taking this out of the research and study mode, but um back into sort of the real real world examples that you've seen.

SPEAKER_01 16:22

Yeah, for sure. So I think when we were, when we were talking about subprime in kind of the previous section, we were really talking about like, hey, there's this set of consumers that is on the edge of the lender's policy. And so it's all about kind of adding that clarity to um have more approvals for that for that subprime segment that's actually going to perform a little bit more prime. Now, when we're talking about this more prime segment, it may be that that this entire population ends up getting approved. And so even though we've got some of these prime consumers that are on their way up in credit score when it comes to traditional, and then we've got another segment that's on their way down. This is all about refining pricing. And so, as Kristen mentioned, that the swap-ins, swap outs are happening here, but not necessarily whether you approve or don't approve the consumer. It's more about getting really refined with that pricing and making sure that that risk-based pricing is kind of appropriately applied. Um, this also leads to a more profitable portfolio, right? Those consumers who ultimately aren't going to score as good as their traditional credit score says they are at time of application. We maybe price them at a higher rate, maybe lower limit amounts, kind of balance out that risk. And then on those consumers that are going to score much higher in a year or two years from the time that they did at application when it comes to their traditional score, the alternative data is kind of telling us, hey, these consumers are actually going to perform better if you want to get this application, if you want to be competitive, you better offer them an even better term on this loan and kind of nab that you know cream of the crop here and get those prime borrowers that are actually going to perform really well. And so this is just about competitive advantage, accurate pricing, maximizing profits inside those portfolios.

SPEAKER_00 18:19

Thanks, Zach. Yeah, I love the context that you can add here and speak to sort of the different angles of why alternative data is important. Um it's different sort of in the subprime side of things, the the swap in grouping, and then from the prime perspective, how that's

Credit Invisibles Ranked By Risk

SPEAKER_00 18:36

different. And next I want to go to the credit invisible part of the sort of gray zones that we're talking about in the credit spectrum. We find in our study that 22% of applicants are credit invisible at time of application. But with risk view alternative data, we can score 95% of them. And with that score, we can then determine all right, is that a prime risk view prime score or a subprime score? If, again, if you're following along with some of our visuals, we have some bar charts that break out where a uh risk view prime scoring, traditionally credit invisible applicant, what happens to them a year later? Well, we find that 58% of those credit invisible applicants with a prime risk view score end up becoming visible a year later. That's important because it means that when that applicant went to apply, they didn't have any traditional data background. That's why they came as credit invisible. Clearly, they have the means. They're going to be approved at some point. They're going to become uh traditionally scorable. And that means that the when risk view alternative data can score them as prime early on. That's a really helpful signal to anyone using alternative data because it means being able to get in early on for this applicant to get credit earlier and to be sort of 100% of wallet share. So when we look at their traditional score a year later, we see that those who had a risk view prime score at time of application, well, their score a year later ends up being around 655. That's on average. Meanwhile, where risk view was able to score a traditionally invisible applicant, but they look to be subprime at time of application. Yeah, a year later, their score from their traditional score is around 562 on average. So clearly in the subprime side of the spectrum. And finally, we look at delinquency rates a year later. And this is the biggest difference that we see that's really sort of stark as a data scientist. You look at this and it's very impressive how risk view, the alternative score when it's a prime risk view applicant, we see that their performance, their delinquency rates, just 4% a year later. Meanwhile, those credit invisible applicants that show up as a subprime applicant from a risk view perspective, their delinquency rates are at 32%. So that's an 8x difference between those two groups. And so just big uh differences in the ability for risk view when we add additional data at time of application, it gives a much more full picture of an applicant. And that full picture helps to understand where they are trending towards, where are they headed in their entire financial background, because their traditional score sort of captures all of the elements of what's happening in their financial history and uh alternative data from the beginning has a way of segmenting out those that are on the credit sides of the credit spectrum that are more of the gray zone. We can more cleanly push those applications. To being solidly in the prime space or solidly subprime, or just being able to score more credit invisible applicants in the first place. So all of these are uh good moves. And I'd like to toss it back to you, Zach. I know um credit invisible side of things, um, you've got quite a lot of details on as well.

SPEAKER_01 22:23

Yeah, I mean, credit invisibles, the the idea of being able to score those that don't have any kind of presence at the traditional bureaus um is really kind of the the core use case that risk view or alternative data started off being used for is how do we get more insights on these consumers that we otherwise don't see anything. But I think that this study really kind of gives life to that. So it's not just about like, hey, um, is it just a uh red light, green light, can we approve this consumer or not? But you can actually see that, no, actually we can rank order risk across this entire population of consumers that is not traditionally scorable. And so interestingly enough, we can we can see that while the the traditional trade line data is basically a history report on what's happened in the past, how the consumer has performed in the past, we can see that the alternative data is kind of showing us the foundation from which the consumer is working and their ability to repay or their stability just financially that that gives them kind of that springboard to go into the lending space. And you think about kind of before some of these scores were available, um, lenders would do things like secured credit cards or other kind of low-risk lines where they help the consumer prove out their ability to repay before they launched into something like this. But when you think about something like a secured credit card, it's really just cyclical. I mean, it's something that it's not really showing the consumer's ability to repay. It's kind of just almost cheating the system a bit. And so I think that by using risk view, lenders are really able to not just get these consumers approved, but uh place them in the appropriate risk band, give them the correct amount of limit up front. And so it's not about just giving a consumer that we've never seen a $500 limit and just taking a chance on them. Some of these consumers may deserve much more than that. And um, we're able to kind of get them into the credit ecosystem faster and get them kind of ramped up to where they should be with their kind of foundational alternative credit data, where that data says they should be from the beginning rather than kind of having this slow grind to building credit history. And like Kristen mentioned, many of these consumers that were able to score with Risk View and that apply for loans, they become visible in the core system very quickly. And so then at that point, we flip all the way back to the other use cases, which are more about refining and adding additional insights on top of that, that performance data that we see from the bureaus. And so you can kind of see how how the whole how the whole picture uh comes together from the consumer having never had any credit to entering the system, becoming credit visible from the traditional data standpoint, and then lenders being able to kind of further refine that data even more and make sure that they're just aligning the rates and terms of their portfolio with the correct risk levels and the correct consumers.

Portfolio Lift And Closing Takeaways

SPEAKER_00 25:24

Oh, perfect, Zach. I also wanted to take a step back because we've we've zoomed in on three parts of the credit spectrum, but let's look at the portfolio level results, because that's what also really matters with this. In this study, we used using traditional data alone, lenders we found could approve about 55% of applicants. But when alternative data was added across the credit score spectrum, so everything from uh credit invisibles and then deep subprime to super prime, if we do this um analysis of adding alternative data in, the approvals increase to 67%. So that's a 12% lift in approvals without increasing risk, which I know you were just saying. Zach, I mean, how how should lenders interpret that result from a business and inclusion standpoint?

SPEAKER_01 26:13

Yeah, I mean, it's I think for for lenders, it's all just about maximizing profitability. I mean, through through financial inclusion efforts that get consumers started into the ecosystem to being able to further refine prices. Um, the idea here isn't to compromise on risk thresholds. The idea here is to maintain those risk thresholds while adding approvals, adding incremental revenue into that portfolio and really just maximizing profits by controlling risk and adding more consumers into that, into that pool, having kind of that marginal incremental approval and at the same time refining pricing. So I think it's just all kind of points back to that ability of, again, alternative data to be useful across the entire credit spectrum. And the end result is really just maximizing profits for the lender.

SPEAKER_00 27:07

Yeah. Yeah. I mean, it's important because when data complements each other, it means that each data set picks up that unique set of signals on the consumer that provide that full picture. Um, there's really like synergies between traditional and alternative data.

SPEAKER_01 27:22

Yeah, definitely. Well, thank you so much for joining us today. And Kristen, thank you so much for sharing your insights. Uh, listeners, if you'd like to learn more about this research or how LexisNexis Risk Solutions uses alternative data to support inclusive risk aware credit decisions. Visit risk.lexisnexus.com and the link in the show notes to see the data. This is the last episode of season three, so stay tuned for season four coming up in a few months. Thanks.