Episode 190: Orchestrating the Future of Credit: How Alternative Data Pipelines are Changing the Lending Landscape

Speaker A: Foreign. I'm Dan Correll with DCA Card Links and this is Commerce Code, a bi weekly digital commerce podcast for leaders in card linking, loyalty and digital marketing, mobile wallets and payments, and financial data. Thanks for joining this running conversation with leaders in the industry and if this podcast is helpful to you, come join us at a DCA Card Links event. You can learn more at www.digco. This week I'm talking with Arvind Ranta from Pentadata and ricard Bandebo from VantageScore. A little about the companies VantageScore is a leading credit scoring company whose model is especially predictive because it uses new data sources that make it possible to provide credit scores to people even if they have limited credit history. Pentadata is a big reason VantageScore is

Speaker B: able to do this.

Speaker A: Pentadata is a financial data orchestration platform, which means its customers can access many different kinds of financial data through a single single point of contact or in software terms, a single API. Arvind and Ricard have joined me today to talk about what's making credit scores more accurate. Things like rent and utility payments that are big predictors of consumer payment behavior but which have been omitted from credit scoring until recently. What it takes to get new kinds of data like that into the credit scoring system. It's not as easy as it looks and what all of this means for the marketplace for from consumers to businesses. So stay tuned for orchestrating the future of credit. How alternative data pipelines are changing the lending landscape. A conversation with Arvind Ranta of Pentadata and ricard Bandebo of VantageScore. Commerce Code is brought to you in part by VantageScore. Eight of the top 10 banks and over 3,400 leading banks and fintechs use VantageScore to predict and manage repayment risk. Learn more about the latest advances in credit scoring and how to grow your lending business by leveraging financial inclusion@vantagescore.com. Arvind and Ricard, thank you for being on Commerce Code and welcome. It's great to have you on the show.

Speaker C: Thanks Dan.

Speaker D: Excited to be here.

Speaker C: Glad to be here.

Speaker B: Great. Well look, we are here to talk about how access to better data is kind of helping consumers and businesses by providing more accurate credit scores. And I'm excited to have this particular combination of companies here, Pentadata and VantageScore, because your couple of companies are sort of at the forefront of this move. And I'd like to just start with, ah, maybe a quick overview of what each of your companies does And Ricard, we can start with VantageScore, if that's okay, because then I think Arvind can kind of explain where metadata fits in as the service provider to Vantage Core, like in a way that allows you guys to do what you do. So, Ricard, let's start with Vantage Core.

Speaker D: Vantage Score is pretty simple. Uh, we create consumer credit scores, and we are, in fact, the most used credit score right now in the United States. And to give you a little bit of context, we're an independent, uh, joint venture of the three national credit reporting agencies, which are Experian, Equifax, and TransUnion.

Speaker B: Okay, great.

Speaker A: Cool.

Speaker B: So, Aravind, you just tell us where

Speaker A: Pentadata, like, fits in there.

Speaker C: Yeah, absolutely. So Pentadata, think, uh, of it as a financial data orchestration platform. We sit above the aggregators. So think of Finic City, Akoya Plaid, and we give companies a single API to access financial data from all of these aggregators. So the key word I would say for Pentadata is orchestration. So we are just not connecting. We are routing, normalizing the data and optimizing it across sources in real time for data consumers so our, uh, clients get consistent data delivery without managing multiple vendor relationships. Then, uh, specifically for Vantage Score, the value is that we can reliably deliver alternative data signals like rental payments, cash flow data, and so on. That Vantage Score has done an amazing job and continue to do genuinely important work. And I think this, uh, infrastructure story behind how Pentadata is helping is one, uh, market doesn't fully understand. So looking forward to getting into it.

Speaker B: Great.

Speaker A: Okay.

Speaker B: All right, well, look, we'll, we'll stick with you, I think, which is just

Speaker A: thinking about the landscape overall in the, in the sector.

Speaker B: And it has been shifting. So I'll just kind of throw a couple of things out there and get your thoughts. So, like, you know, large banks like JP Morgan charging aggregators for data access, uh, the fate of the CFPB's 1033 rule is, you know, unclear, let's just say. And the bureaus themselves are partnering with aggregators to embed cash flow data. So, like, from your perspective, has the

Speaker A: ground shifted underneath the open banking model or how do you see things right now?

Speaker C: Yeah, so open banking has been around for some time, but I think it's significantly, the ground is shifting in three major ways. First, as you said, large banks are charging aggregators for data access is no more an exception. It's actually increasingly become, uh, becoming a trend. And more banks will follow this era of free access to data is over and that's forcing aggregators to renegotiate their economics which flows downstream to end clients. So that's very important to consider that the cost of data would go up if nothing changes, uh, for the Data consumers. Number two, I would say, uh, is the CFPB 1033. It's in a bit of a limbo. Companies that were building on the assumption that open banking mandates were coming, they had to recalibrate. It remains to be seen how powerful that policy tailwind would be. But unlike uk, US is a pretty much free market. And um, the access to open data is really dependent on commercials that these entities would have with banks. Number three, I would say is really interesting one that bureaus themselves are moving into data aggregation. So if you think of large bureaus who've announced their plans to embed cash flow data directly and of course with help of Vantage Core, they're helping build out more inclusive credit models which incorporates signals which are beyond traditional data sets. Think of rental data, think of BNPL data and so on. Uh, so that's really interesting and I think what it means for Pentadata here is this case for an orchestration layer has never been stronger because think of the cost and complexity of direct aggregator relationships has just compounded, they've increased and multifolds. So if you're a large marketplace or a large lender, you want the optionality to work across multiple aggregators at the right cost and the right data availability and with the most modern data sets that are available right now.

Speaker B: Okay, that's a great picture of how things are kind of changing where we stand. Arvind, so thanks for that. And Ricard, to kind of get your angle too, you've watched kind of credit scoring evolve from the inside. And I'm just wondering how the ecosystem of open data and credit scoring has that sort of changed in the US over the last few years. And then maybe how are some of the new types of data, I mean, does that make new types of data available? What does it do for you? So let me get your, your response to that.

Speaker D: Uh, it's evolved quite considerably, um, over the last few years. I would say there's been a lot of change. It's driven predominantly be by a need for better scores, you know, and as a result we've seen also competition has, has ramped up. The, the reason that competition's ramped up is because better scores have come around. Right. So we've been able to develop much more accurate scores and also scores that can score a lot more people. We wouldn't be able to do that if we weren't able to find data that would help accomplish both of those tasks. In other words, alternative data has enabled Vonage Score to create models that both are much more accurate than our competitor scores, but also they can score around 33 million more people. That's a very substantial difference. Um, in the marketplace, alternative data has been critical to that. There's different types of alternative data, and we'll talk more about that as we go through this podcast. But I'll start first with alternative data that is being collected, um, by the credit bureaus and incorporated into the credit file. And so what we've seen here is, you know, Vonishcore was the first to start using, for instance, rental data. But, um, they also added utilities data as well. And rental data in particular is really, really powerful because it can have a really big impact on a consumer's score, which shouldn't be that surprising given that people who rent their, their monthly rental is probably the biggest single outlay they have. So if they're able to consistently demonstrate good performance on that rental data, it's a really strong indicator of how they'll not only perform on other loan types, but also very importantly on how they're likely to perform were, um, they to get a mortgage. Okay, so we actually see that, and we did a study that we published, um, recently and we did another study a couple of years ago where you'll see that consumers, um, can get a lift of over 100 points, um, based on getting rental data added to their credit file. So that's great. Um, and, you know, I will say, though it's not ubiquitous, and this is one of those things that not a lot of consumers, um, understand is that, you know, a lot of people assume that, oh, wow, my rent, of course it's going to be added to the credit file because it's such a big payment. It's so important. Uh, but no, it's not. It's not one of the mandatory types of, um, reporting because it's not really a form of credit. However, what we are seeing is that the amount of rental data on the credit files has been increasing. We've seen the bureaus, we've seen different technology vendors out there really helping to escalate to get more rental data on the credit file. And that's a great win because then more and more people will be able to take advantage of this as well. Again, we also see utilities data being added. Utility data is also useful. It doesn't lift by as much as, um, rental data. But also though it is a very valuable input, um, particularly for people who don't have traditional forms of credit. It may be because they're new to credit or maybe because they paid off their mortgage and they're older, but they've still been obviously paying their utility bills and other types of bills. Right. So for those types of consumers, it's really important for that to be able to be captured. So those are two examples that have been really, um, well received. Uh, there's one type of data that isn't yet really being captured and that's within the, what's called the Buy now, pay later ecosystem. And so Pay in four, which is a specific type of, of buy now pay lender type of loan where you probably see it all the time when you go into a shop or um, you're online shopping and they say, hey, you can pay this in four installments. That type of data is yet to be reported in enough volume for it to be really used yet at the credit file and within scores. But we are working closely with both the national credit reporting agencies as well as the BNPL providers to really try to capture the value of that because we believe that could be really helpful, ah, to consumers by again, enabling more visibility into how they're doing on a product that could be very predictive as to how they'll perform on other products. But also for lenders who are concerned that they're not seeing the full picture of a consumer's lending profile when BNPL is not being reported. And then I'd say the final area, which is really interesting and we've had a lot of success in, and we've partnered with Pentadata closely on, is around what's called consumer permission data. And in particular the data there that's really valuable is your banking data, whether it's your checking account, your savings account, that type of data, that's really valuable. And I think that's where we see some of the greatest promise going forward.

Speaker B: Yeah, that's a big area of change. And Arvind, I want to get to you on that and other things in a second. But just as a comment, I mean it is. While you were talking Ricard, and I was paying total attention, uh, I looked up because I thought, well, how many. And it's consistent with what you said, 45 million households, ish rent. And the idea that that is an alternative form of data gives you a feel for kind of like how set in its ways or whatever the industry was in the year of our Lord 2026 that this is like new that you would have rental data in there. It's uh, something like 100 million Americans live in a rented home. Um, so uh, that's really interesting. And then you mentioned the permission to data which obviously gets to bank accounts, open banking, all of that. Arvind, I'd be interested to kind of get you know, your, maybe your angle or what you'd add. Like are there particular kinds of alternative data that you're excited about that you think will be, have a big impact? Where do you think you're going to see positive change coming from?

Speaker C: Uh, I think my list is going to sound very similar to what Ricard said. Uh, I would say the rental payment data without question is the most important data set. To Ricard's point. I think there are roughly around 45 million Americans who are credit invisible or thin file as you would call. Right. So as you just mentioned Dan. Right. Like there are uh, there's significant portion of them by the way who pay rent reliably every month. That's uh, a pretty strong signal but it wasn't being captured systematically before. And from an infrastructure perspective what Pentadata is doing working with scoring uh organizations like Vantage but also with the data providers, like rental data providers. We just signed up a uh, huge partnership with Isusu, one of the players in this ecosystem to source rental data for creating signals like this which could be then modeled by Vantage score into the full fledged credit file and gets reported as a proper score. And people can benefit from paying rent every month and that shows up in their credit history. Number two I would say is cash flow. It's interesting bank transaction history tells you a lot more than a um, point in time credit pull. The cash flow data uh, in your bank statement actually shows your stability and income spending patterns and um, sort of different sorts of behaviors that could be so meaningful uh for anyone from lending to loyalty providers to BFM apps. The challenge historically has been delivery, right. Getting clean normalized cash flow signals from so many financial institutions. It could be a challenge to have that consistently rendered as a data set. And that's an infrastructure problem as much as it is a data problem. Better has been spending a lot of time on that to curate those data sets. And then finally in bnpl I would agree with Ricardo, think that's a really uh, emerging, still developing as a signal but it's a very important uh, space. Consumers love bnpl and there's still a lot of ways in which BNPL could be more mainstreamed, uh I spent four years at Visa heading BNPL globally and I can attest that there's so much appetite with fintechs as much as they do the banks to put more guardrails on BNPL to make it more mainstream, to give people just like rental payment data, to give people who are making their BNPL repayments on time, give them credit where it's due for those good behaviors. Right. So and that um, as long as we can systematically capture those data sets and I know Vantage is doing a great job in working with the ecosystem of partners and we are partnering with that, but I think this is going to take a couple of years to get there as we see this space mature. Uh, but this is another interesting space to watch out in kind of plain

Speaker B: English I guess and this is a big question, but with all the changes that have been taking place, what does this make possible that wasn't possible five years ago? So you've talked about orchestration and these different things, alternative data pipelines, you know, in practice what are businesses and consumers going to be able to do do you think or do you hope you're

Speaker C: in the next few years from an infrastructure perspective? I can share with you that five years ago if you wanted to build a data pipeline that pull these rental payment signals from these multiple sources, you were negotiating with five different vendors, you're building up five different integration pipes, you are managing five different formats of data sets, right? So because these are different vendors, different sources of data, so on and so forth, that's an 18 month project before you've scored a single consumer. That's the infrastructure and data challenge. What's changed, uh, uh, is this abstraction layer now we can now sit above this whole complexity and deliver a much more normalized data stream through a single connection. That's where metadata spends a lot of time and think of this as multi source routing where you're not dependent on any one aggregators coverage. So if one of the aggregator doesn't have a relationship with a regional credit union, we could route to another aggregator automatically and client would never see a gap. There's a full on redundancy built in, full on coverage built in. The other shift I think is a data quality infrastructure, right. So normalization used to be very manual, very time intensive. Over the years we've automated the reconciliation layer so that the signals that arrive from say Vantage score, they're clean and consistent regardless of whichever source they come from. Whichever data source they come from, they're highly structured and consistent data sets. And then uh, finally I would Say in credit scoring is like garbage in, garbage out. So it's high emphasis that what gets fed into scoring models, it's highly structured, normalized, curated. And I think that's where infrastructure for alternative data comes in, that it's more clean and systematized. As we built out the scoring based on that.

Speaker B: Great. I mean, it follows sort of some of the usual arcs, which is that things get more complex and then a lot of work comes in to make it easier and more possible. Right. Like, it feels like a lot of what you're describing is lowering the barriers to being able to just do things for organizations and to build, you know, to kind of realize the potential. And so turning to Ricard, you know, we've talked about sort of new data streams, deliver capabilities, making things more accessible. What does that allow you to do advantage score that you couldn't do before?

Speaker D: So before working with consumer permission data, we were always, uh, just working directly with the data that's on the credit file. Um, and again, the different type of data on the credit file. Clearly, however, we'd always realized that there's data that's not on the credit file that could be really valuable in trying to better evaluate people's risk and again, be able to evaluate more people in a better way to get to the point of before. Right. You want to score more accurately and you want to be able to score more people. And so we always had a hypothesis that if you can incorporate cash flow data, that would really help get an additional signal or several signals on a person's creditworthiness. And just to put that in context of why would that be the case? Um, I'll just give a very simple example, but it's just one of many. The credit file, the key signals that you'll see there are things like whether somebody goes delinquent, that's the most consequential one, they miss their payments, or they get into a stage of, you know, foreclosure or delinquencies, etc. Right. So that's reported that that's obviously a very, very strong signal. You have things like ratios in there. So is a person using up all of their credit on their credit cards and their loans, or do they have a lot of buffer space? Right. So those things are very useful, but they aren't necessarily giving you a sense of how is the consumer doing financially.

Speaker C: Right.

Speaker D: When you get an opportunity to look into, um, the different bank accounts that they give you permission to look at, then what you see is, for instance, how is their flows of their Cash flow evolving. Right. So let's look at the checking account. You know, are they consistently you being able to increase their checking account, is it decreasing? Is there, are there any trends there that could either tell us that this person is, is doing, is doing well, this person has a long track record of having a very stable income and very stable outflows, or is there a signal that this person might be getting into some trouble because there's a persistent decrease in their checking account that might be driven by the fact that they're having higher expenses and they're having income. Right. So a lot of different signals that actually are often things that you'll see before you see those other effects that I mentioned before on the credit file. So in a way, the consumer information bank data can be a leading indicator and they can be very strong from that perspective. Additionally. So we did a lot of research and we found absolutely it did have lift. We also spoke to a number of banks who were already using this data on their own customers when they're trying to evaluate them for other products. And they would often see lifts of around 20% in predictive power. So that again, just validates that it's very powerful. So being able to harness this data for credit decisioning is really valuable. And, um, it definitely improves the predictive power, um, double digits, as we, as I, as I mentioned, in some cases. Right. And also it helps us be able to score more people accurately. So we see it as a really big win. And the various different lenders that have been doing pilots with us have been seeing strong results from incorporating this data into what we call our Vonage score four plus score, which is the score that we've created that incorporates this consumer permission bank data.

Speaker B: Great. Okay. Folks who listen to commerce code are sophisticated sort of industry people, which is great because it allows us to just dive in. Right. And so, um, I want to go one click further, I guess, and talk about what orchestration and Arvind, you talked about orchestration. Talk about what orchestration actually means in this context. So as I've heard it described, Pentadata sits above aggregators, not alongside them. And so talk a little to us about what orchestration means here. And why would a company sort of route across multiple data sources instead of connecting to just one? So, uh, explain. And maybe I've not described that quite right, but explain that to us.

Speaker C: Sure, Dan. So just to recap, right, so Pentadata sits between the suppliers of data. So think of aggregators who have pipes into individual banks and the consumers of data. Think of lending, credit scoring companies, loyalty companies, PFM apps and so on. The two sides of supply and demand. Orchestration is really core part of Pentadata's architecture. And essentially why orchestration is important is orchestration means managing complexity on behalf of our clients. Really that's what we are doing. So I'll give you some examples, not aggregating data. So think of routing, normalizing and optimizing it across sources in real time. We are not only able to aggregate data, but do all of the normalizing and routing. And it's important think of it as a company which connects directly to one aggregator. Today they will get one sort of coverage map of different bank accounts, right? It'll be in one data format and it'll be a single point of failure. Now when that aggregator loses a bank relationship, what happens? The client on the other side could be a lender, could be a loyalty company, their pipeline breaks. That's the risk of single source relationships and that's a real operational risk and credit decisioning. Right? We sit above all of that. So Pentadata's routing logic selects the best available source for each query based on coverage, data freshness and cost. So the client sends one API call, we figure out where to get the data, normalize it and return a very consistent response back to the client. So Ricard mentioned it, right? Like Vantage score, examples of how they're looking at. When you're thinking of scoring, you need both data quality and coverage depth, right? So you can't have either or uh, both data quality and coverage depth matters and it affects score accuracy and that reliability matters enormously. Right. You can't score someone you can't reach and you can't really trust a score built on signals that are inconsistent. Orchestration is designed to solve for this complexity. So you have bigger coverage, you have more standardization, and frankly you have an optionality as a client to decide how you want to route that data request based on the data availability and cost consideration as well as coverage.

Speaker B: That's great. I think I understand better than I did before sort of machinery of this stuff. So that's on the work that you do at Pentadata. And then Ricardo would be interested to kind of get. What does that enable for VantageScore to be able to do, to have that kind of orchestration layer in place.

Speaker D: It's incredibly important. Some of the reasons, I'll repeat that Arvind mentioned because they're important to us. But there are also some additional um, aspects to this. So when we embarked on uh, creating VanishCore4 we spent a lot of time trying to understand what had been some of the pitfalls that others who tried to use consumer permission data had fallen into. We thought it was odd that why isn't everybody doing this? Right. Um, and there definitely had been some attempts, some at a cash flow only score. We don't think that's great because we see that the highest predicted value comes from the credit file. But again, so I think the best value comes from a combination of both credit file data.

Speaker E: Ah.

Speaker D: And cash flow data. But there had been a couple of attempts even on that side to build models and they hadn't been very successful. And so we were trying to understand why. And what we realized was there are a few reasons, but one of the main reasons was that they had, when they built the sol, they built them with only one of the aggregators often. And that's something that is challenging for a number of reasons. One, as Arvind pointed out before, one aggregator may not cover all the banks or may have a particular issue with the bank that day, which creates a really bad consumer experience and a bad experience for the lender as well.

Speaker C: Right.

Speaker D: So that's not very good. Secondly, one of the things that we heard loud and clear from lenders were that many of them have already engaged an aggregator. And so when they're going through a process of setting up a loan or something, or a credit card or something along those lines, they'll often use an aggregator, for instance, so that they can set up a direct deposit.

Speaker C: Right.

Speaker D: Or so they can perform some other form of action. And so what they didn't want is for the consumer to have to use multiple different aggregators in the same process because a, that created a pretty poor experience for the consumer because they're basically having to repeat steps. But secondly, it also just, you know, incurred a lot of additional cost to that lender. Uh, and so we then realized that when we were going to build this solution, we needed a solution where it was agnostic to aggregators so that it could indeed work with whichever aggregator, uh, that a lender was, had a particular preference for, but then also that it could have this kind of waterfall. At least that's how I think about what you were describing before Arvind, where, you know, it will choose the best aggregator for the best bank and certain conditions. Right. So some lenders actually prefer that over using their, their expected one. So that was really, really important to us. And the other thing to Arvind's point too was that we, we realized also we didn't want to use specific attributes that have been created by any one given aggregator because then we would be beholden and stuck to using that aggregator. Also we don't know if they built them as well as we do like that. That is our thing. We build these things. Attributes are one of the most important things that we spend enormous amount of time and we've had a lot of intellectual property around that. So for us it's always better to work with the rawest data. But uh, what we also need is that we need that data in a consistent way so that irrespective of which aggregator or bank is being linked to that, the, the algorithm um, can feed that in easily. Right. Without having to do a lot of additional transformation on our side. And so for both of those reasons we again realized that we needed to set up a system here that can work with multiple different aggregators and can even prioritize aggregators based on performance. But that also could provide us with uniformed formats so that the process can work very quickly and smoothly. And we found that Pentadata was a great partner to help us accomplish that.

Speaker B: That's great. And we've gone deep into the sea on sort of how this stuff works which is great. And I want to come back up to finish, you know, kind of just to outcomes or results. Right. And just get a perspective. Maybe we start with you Arvind on okay so we've talked about uh, the results of some of this or some of what we can do, you know, opening up credit access through more accurate credit scores. Talked about rental data and these other things. So where do you think that businesses will see the benefits of this in the next few years? And Arvind, we'd start with you maybe and we go to Ricard after that.

Speaker C: I think the near term win is going to be around risk decisioning. So all the lenders who are incorporating cash flow now also have rent payment signals into their underwriting will just get better credit decisions. And I think we got pointed out to, I think it was just noting down 100 points improvement like with addition of rental data in the mix. Right. So importantly there'll be fewer defaults. They should have more approvals for credit worthy consumers that have been turned out turned on before. Uh, I think that's a real P and L impact that we would see with for lenders to have through all these alternative data sets. Uh, in the medium term I think there's going to be a lot of improvements in embedded finance experience. I'm a big proponent of embedded finance that was a big part of my life in my past building embedded finance products. So that's important because think of like as more financial products get embedded into these non financial experiences with think of payroll platforms or property management software or gig economy apps. This demand for real time permission financial data that will grow significant. All of us have used when we are creating accounts or when we are getting approval for a loan offer or a quick credit or buy now, pay later we've experienced embedded finance experiences. All of that is riding on a good orchestration infrastructure that's needed to really make that possible at scale. So really excited about that. As the industry grows more mature I think we're going to see more of that and uh, frankly if I'm honest about where I think this will all end up, the companies that own the data delivery layer will have a significant leverage. Right? Scoring models we've talked about lending products, decisioning tools, they all depend on clean reliable data inputs. And what we love to believe at Pentadata is we are focused on building that foundation on which all of this commerce and all the open data or open finance use cases will thrive in the future.

Speaker B: Great Ricard thoughts from you.

Speaker D: I've obviously been speaking specifically about credit scoring. I think the amount of valuable solutions and services that can be provided to consumers, consumers uh, through open banking is enormous and even growing potential. I'm not just a proponent of this and a fan of it but I do also believe that it is something that's going to grow. We'll see where the CPB and 1033 land. I'll be hopefully optimistic um, that a good version of that gets through that will encourage greater adoption um, of this data. I think we already seeing today that this makes a difference and I think what's even more important is that we've got a lot of uncertainty going on. Um, one of the things and coming back to credit risk in particular is that uh, the more uncertainty you have or particularly the more you head into a challenging economic environment, the bigger the difference a credit score makes. When the world is everything is going happy and fine, very low. There isn't a huge difference because there aren't that many delinquencies frequencies. But as soon as you start seeing economic challenges like we saw a blip in 2021 with the COVID we obviously had a very big one in 2008. But these things happen, they are cycles, they will come back when you start seeing those types of behaviors. That is when the difference between the credit scoring model that a lender uses makes an enormous difference on the losses that that institution will face. Uh, and the reason I'm mentioning this is because using consumer commission data, banking data, savings account data, et cetera. Right. Will add a valuable signal. And as we head into an environment potentially in the future where there will be a more challenging economic environment, then lenders are going to be even more driven to invest in better models. And I think that that may ultimately lead also to a greater demand for models that include consumer permission data.

Speaker B: Those are all terrific points that I hadn't specifically thought about before, Ricard. So thank you for prophesying our future, our near future, our distant future. We'll see. But yeah, I mean the environment is constantly, ah, changing. We can all remember times when perfectly credit worthy people after the crash wanted to, I don't know, redo their kitchen or whatever, and they couldn't get along. Right? These behaviors, the market changes and it matters. So. Great. Well, look, gentlemen, it has been, uh, a pleasure to have you on Commerce Code. I want to say, uh, thanks again for your insights and your thoughts and uh, we look forward certainly to having you both on again in the future. Thanks.

Speaker C: Thanks Dan. It was a pleasure.

Speaker D: Thank you.

Speaker E: Commerce Code is sponsored by Pentadata, the all in one financial data API. Whether it is bank account data, credit card transaction data, or credit reports and credit scores, Pentadata has it all in one, simple and easy to use API. With coverage of over 6,000 banks, over 200 million credit files and 60 million merchants. You can get all the data you need for your apps@, uh, pentadatainc.com commerce code is a bi weekly podcast bringing you conversations with executives who are leading the way in digital commerce. If you like Commerce Code, your company should join the Digital Commerce alliance and become part of our mission of advancing trade for good through standard setting, industry networking conferences and best practice sharing. Check out our website at, uh, www.digcoll.org. on behalf of DCA, have a great week.