A proven approach to balance sheet and net interest margin planning

All right. Good afternoon, everybody, and welcome to what will be the first in a series of webinars that we cross-country will be co-hosting with OneStream. We're delighted to have you here today as we dive deep into the topic of how to use balance sheet and net interest margin planning using OneStream.

We've assembled a great team for both cross-country and OneStream who are deep in financial services and banking, and we're going to provide a couple of, we've got an action-packed agenda for you today.

So with that, let's go to the next page and do a little bit of round of intros, and we'll get into today's objective. So, Mac, if you don't mind flipping the screen. By way of introduction, Pat Belfolino, I'm a managing director here at cross-country.

A bit of my background, I've been in the industry for over 30 years, started my journey in finance and accounting, and spent over 25 years implementing CPM solutions at some of the world's largest financial institutions.

And with me is Jaime Ahn and Kyle and Brandon. We'll go through introductions. So, Jaime, over to you. Sure. Thanks, Pat. Everyone, thank you again for joining us. We appreciate the time. I'm Jaime Garza.

I'm a partner with our banking and financial services practice here. I spent about 10 years in industry in a CFO and treasury roles, and then the last 20 years in consulting around those areas. And a lot of my focus area has been around balance sheet planning and product profitability.

I'll turn it over to you. Thanks, Jaime. Thanks, everyone, for joining on Merchant here. Roughly around 20 years of experience between industry and consulting. My specific focus has been around treasury as well as balance sheet planning.

Thanks for joining again. I'm looking forward to show you what we've got in store. Kyle? Hey, everyone. Kyle Webb, a manager here at Cross-Country Consulting. I am a certified OneStream solution architect with about six years of OneStream and finance tech transformation consulting experience.

And then prior to consulting, I spent a few years in banking, specifically in the credit analysis and underwriting department. So, looking forward to showing you what we have. Hey, guys. My name is Brandon Beam, Senior Solution Consultant here at OneStream, supporting our financial services team specifically.

Just about 15 years of experience in the industry across multiple roles, helping support and demonstrate how OneStream can be the solution of the future for our financial services customers. Looking forward to today.

Excellent. Thank you, Brandon. Thank you, everybody, for jumping on. Let's move to the next page. So, I want to tell you a little bit about Cross-Country if you're not familiar with us. So, we are an alternative to the Big Four with a primary focused on the office of the CFO.

We're technology-enabled and we embed AI into everything we do. Over 1,200 practitioners located across the world. And our OneStream partnership, we've been recognized as one of the fastest -growing partners in the marketplace.

We have a dedicated team focusing on financial services and banking, and we deliver with 100% success. What we believe is that OneStream is an important part of your finance transformation and journey.

So, we like to think about that OneStream is end-to-end in solving your EPM, CPM needs, and then think bigger and broader, right? So, if I think about how data works and think about how my overall architecture works and where OneStream really fits into that, and then developing best-in-class solutions to meet our clients' needs.

Now, let's talk a little bit about the agenda for today. If we can go to the next page. So, three things we want to cover today. One, provide some perspectives on what we're seeing in the banking industry and talk about a use case around auto loan portfolios.

So, one of the challenges we hear from our clients is that it's really hard to model a lot of the data that we see. Everybody's relying on Excel spreadsheets and thinking about how do I quickly spin up what-if scenarios.

What we've done is built up a OneStream asset that really highlights all the capabilities of what the platform can do. So, thinking about everything from dimensionality to modeling capabilities that allows me to say, what happens when interest rates change? How do I model runoffs in my portfolios? And Jaime and Ahn are going to really walk us through the industry trends and how we've brought that use case to life.

Great. Let's go to the next page, please. So, Jaime, do you want to give an example, give an overview of the case we're going to cover today? Absolutely. And thanks, Matt. So, this example really follows probably, I would say, a set of about 10 to 20 clients that have all had similar states, right? I give the example of every RFP I've received in the last 20 years has had one common message.

We want to spend a lot less time gathering data and spreadsheets and a lot more time actually using the data to understand the business and run the business. And this example is that. We're going to start with this organization first wanted to fix its actuals and get actual product profitability and the dynamics, capturing the dynamics of margin, expenses, and capital, and then replicate that in plan.

So, you know, we're going to cover on what it kind of takes to migrate from a typical current state to a highly improved target state leveraging one stream. It's going to be focused on consumer lending, auto finance.

We're going to look at one product, 72-month indirect auto. And we're going to look at that in the context of inclusive allocations, capital and risk, FTP, and of course, the natural coupon. Then we're going to focus on the balance sheet and interest margin analytics and forecasting, because a lot of this is actually being able to separate the existing book assumptions and its relative contribution to the overall forecasted number along with layering in the new book and analyzing that both individually and together.

And honestly, without it, people are guessing. A lot of times people say, well, we went through all this work and we ended up with the same number. I would say you were lucky if that occurred. The ability to really analyze the interesting components, the coupon, the dealer reserve, losses, charged off interest, FTP, liquidity, capital, credit, et cetera.

It's the model goes pretty deep into those areas. Now, the last thing I'll talk about is volume-based forecasting, because a lot of what we're going to discuss today and showcase is based on the concept that instead of just planning a target ending balance of, let's say I want to go from $8 billion to $9 billion in a year, I just need to add a billion.

That's not necessarily the case, but instead we talk about, okay, what if I did X amount of loans at an average car price or loan balance of X to get to that so that we retain those volumes for allocations and other statistics.

Let's go to the next page. So here's kind of the representative current state once the client fixed its actuals. They said, okay, we got actuals for what we need. Now for plan, where are we? Well, this is very representative of the current state.

Cash flow generation. Highly manual, the ability to directly, distinctly analyze because it's the new ones. A lot of spreadsheets, a lot of files from treasury and custom templates that maybe FP&A built, et cetera.

Timeliness. We all know that you can come in on a Thursday and like, in fact, I think today the Fed is meeting and your boss says, look, I want to see what happens if rates fall. Well, that single what if can take a half day to a day.

And then the CFO decides, well, what if we change these assumptions, you're back to another half day to day. Reporting and analytics, minimal ability to really decipher drivers at the same levels of comparisons to actuals.

Now this is a big thing. While it may sound trivial, if you don't have that same or near comparison, it's hard to understand the drivers of why you were off or why you were on. Allocations. Very manual, simple factor base.

And I've always argued that revenue is not a good, it's not linear to expense. So it's probably the worst driver to use. There are cases where it can be used, but those are far and far. FTP, calculated offline.

Every time there's a new scenario, have to recalculate it and bring it into the platform. Capital and credit, I get very limited ability to integrate the assumptions. Chart of accounts, very GL based with difficulty in mapping between FP&A, ALM and accounting.

Right? So migrating to a platform, then you have to start thinking about, well, if I, let's say I implemented a new RP with a product chart field or work tag, how can I then leverage that within the model rather than having to reconcile my chart? User defined drivers, no ability to plan, volume times balance and retain those statistics.

That's kind of a representative current state. These eight items aren't necessarily all the pain points that we've seen. These just represent the medium. Let's go to the next slide. So one of the things that Ann and Kyle and Pat will be talking about is OneStream has some very unique capability about its use of dimensionality, right? When I used to run ALM systems, every time I wanted to get more detailed on a product, and let's say I added three attributes for each three attributes I added, I needed to create one new product for each.

So I'd have a chart of accounts of 800, 900 accounts to get what was needed. OneStream allows us to bypass that. So we start at the 72 month auto. So if you hit enter, that's where we start. If we go down to the next level, that's our product.

That's our chart of accounts. We go to the next level. I now want to look at super prime, prime and near prime credit grades, which are priced very differently. You know, the difference between a super prime and a near prime is anywhere from three to 400 basis.

I can do that without having to set up another auto 72 month new product. Super prime, new product, prime, new product. Then I go into another category. So, for example, auto prices are going to be dependent on the type of auto, right? The domestic and foreign, you're talking about 30 to $40,000 average, but foreign luxury is a lot more.

You're talking about 80 to 100, $125,000. So that makes a difference. And our portfolio of actuals captures that data. So we're able to use that data to plan and forecast. Let's go to the next level. The next thing I spoke about is bringing in a lot of information from multiple places.

What FP&A has, what treasury and capital have, or treasury has, and then what capital and risk have. So the user defined fields, in fact, Kyle and Ann will show UI that captures a lot of these items. I will say that the bottom in the expenses, we not only capture the expenses, but then we convert them into a process view, like servicing, origination, credit risk, et cetera.

We go to the next area. We get from treasury the existing book cash flows, FTP assumptions on pricing. That's something that the planner can change, like prepayment speeds, but they cannot actually change, like what FTP will be or what the credit spread to coupon.

That is something that's fixed and is controlled by treasury and economics department. The next part is incorporating capital. And from capital, we get the capital assignment methodology. And the way we're represented today is capital plus a triple L.

And we get figures around PD, LGD, recovery rates, all that stuff that we have put into the model to have the output of the forecast based on the expectations from our credit group. If we go to the next page.

Now we take all that, we take the user defined input, what we get from treasury, what we get from credit, we run it through one stream. We're able to produce output that shows that new book and existing book separately or aggregated.

We look at the dynamics of nonlinear pricing. We can change portfolio mix and credit quality. We see variable expenses driven by volume, such as the cost of repo. We can change full and partial prepayments.

Because in a portfolio like this, you tend to see many full prepayments as people a lot of times refi within the first year or people sell or trade in their cars sometime during the life of the loan. Increasing ROE with a target of 12.7 pretext.

And we're going to, you know, a lot of it driven by the fact that the super prime, which is the biggest part of the portfolio, is at a very low ROE. So we're dealing with, okay, what do we need? What are the what ifs to see where we would be at the end of the year if we, let's say, have flat product balances, right? So there we would be really showcasing the dynamics of the repricing risk to our 2025 interest rate forecast.

What if we kind of took that scenario and added 10 basis points to pricing across the board? So the spread that Treasury has given us or the coupon pricing add 10 basis points for that because that is something that the planner can control.

And then the last two is where we apply to growth, right? The number three really shows the growth in prime and near prime and keeps super prime flat because we expect that prime and near prime, those are the ones that could get us to our target quicker.

We have auto loan prices between $40,000 and $100,000 depending on the dealer type. And then the last, we have a little more aggressive pricing at, you know, prime and near prime and anywhere from 8% to 10% and super prime at about 4% and auto loan prices, you know, around $35,000 to $95,000.

So we took that and we run it. Now, you can see on the right hand side, the coupon rates and the net funding rates that we expect to go on each month are declining because the rate forecast is showing a decline in rates both from a prime rate perspective, treasury rate perspective, and the FTP note curve perspective.

Let's go to the next slide. Okay. Okay. One thing that people say, well, you know, why go through the trouble of trying to segregate and do all this cash flow stuff? Well, in a portfolio like this, if you don't take this approach and you say, I'm starting at 8.2 and I want to get to 9, hey, that's simple.

I add 800. And what people forget is, well, what about what's repricing and maturing? What about what's maturing on the existing book? And what about what's maturing on the new book? Because after the first month, you subsequently start having amortizations of that new book.

Well, our model says, well, for a flat growth, you need to put on about 4.3 billion. That's a lot. That's about half the portfolio, a little more. So if I then look at the bottom, I have all these volumes, and it's telling me that I need to, my average loan price at about $51,000 to get to that number.

And, you know, it can vary by the credit quality. But this showcases the importance of having a model that can do this, because that way you can look at the data and understand very specifically why you met a goal or did not meet a goal.

Let's go to the next slide. So now we say, okay, one of the other effects of auto loans and anything that amortizes is if you are using what I would call a true process-based allocation method that works on volumes, there's going to be a point where your revenue falls below the expenses, because the expenses are based on a volume, and those can maintain fairly flat or with a little bit of growth except for the variable ones, whereas the revenue stream is just declining.

So what we did is we ran the nine pools of 500 loans each in July to see at which point do we start not meeting our ROE goal or still producing positive but not meeting the goal. I won't go into all the detail, but you can see the near prime because it's a much higher coupon.

That is what is really carrying a lot of the book, because those are ROEs in the high 30s, you know, up to the 50s. In prime, it's kind of in the middle, and we start seeing that by year four, we are still positive, but we're not meeting our goal.

Super prime, which is the largest part of the portfolio, actually right off the bat starts not meeting our goals. It's profitable for a while, but it's under target. So this is the dynamics. Now what we put in the model is, okay, here are all our forecasts, and we're able to see this dynamic around, you know, our actual forecast.

Let's go to the next slide. So before I turn it over to Anand Kyle, I'm going to leave you with this, because one stream can be used. You know, I met with the CFO yesterday. He said, look, I need true transformation.

I got a lot of team members that say, well, let's buy a new tool. Let's replicate what we do. And he said, that's not, he asked me, do you think that's transformation? I said, no. I said, you want to take a tool like OneStream.

And not only do you want to produce a model like this, but you want it to guide your decisions. So I said, if we look at the super prime portfolio, and we say, okay, tell me what it would take to meet a 12.7% ROE in this forecast.

Well, this is telling me that including all the expenses, all the projected losses, et cetera. Remember, the existing book is already, it's there, right? It's not changing. I have to rely on the new book.

And this is telling me that on average, in order to meet that goal for this category, I need to put new loans on at about 6%, where I'm putting them on an average of super prime at 5%. I'm off about 100 basis points.

Now, what if I increase the loan car prices by 10,000 to about 65,000? Well, that helps me, but only 0.8 of the, you know, 800,000 of the 9 million. So the one thing the model does is a lot of times, and when I was in industry, I would tell the team, I don't, if it's not good news, I don't want you to fudge the forecast to give me the answer I want.

I want the tool to tell me what reality is, and then we can plan on action to try to mitigate and improve what we are within, let's say, the first year, second year, third year, et cetera, okay? So this can allow you to take an action, drive decisions, and you can quantify, you know, those things that are achievable, not achievable, or both.

So now I'm going to turn it over to the team who's going to spend some time actually showing you our kind of what we call version 1.0 of auto lending. So let me turn it over to Kyle and on. Hey, Jaime, just before we do that, I think we wanted to open up a polling question, so we want to make this as interactive.

So if you go to the polls tab, or polls should show up on your screen, we're going to go through the first question, which is, do you currently have these capabilities fully automated within your planning process or platform? So we take about one minute just to click and answer that question.

And then we'll get into the application itself. All right. If that poll is closed, Jaime, if you don't mind turning over screen share to Kyle, so we can jump into the application. I think that's Mac. Thanks, Mac.

Thanks, Jaime. And then I think someone currently has the screen for the poll. There we go. This will take a second to boot up with the VDI. Yeah. Thanks, Jaime. So as the tool actually boots up, one of the things that we want to talk about, we did a lot of discussion around context, thinking about what the different scenarios are, what are the sort of different dimensions and attributes we're thinking about.

As we're thinking about the planning process, we've got some things to think about when it comes to integration. You've got materials coming into the treasury side of the house. You've got some capital and credit related items that we need to think about as well from an integration standpoint.

So those are all going to come together. And what we're going to show you, we're going to go through historical data analyses. And we're going to walk through some metrics. And we'll look at some composition over time, yields over time, product balances over time.

But really, that's all to say, then we'll talk about how the dimension capability works within one stream. Sort of tying back, one of the issues we keep hearing about is, well, we plan at this level, but we need to be able to do reporting at four different levels.

So I'll give you an example. So we've got board level reporting. We've got C-suite reporting. We've got line of business reporting. Then product profitability. So we'll talk about the dimension capability in one stream, how that basically allows these things to stay in sync.

And so we'll get to that in just a second. And then in terms of the forecast itself, we'll talk about the forecast. And we'll talk a little bit about what the different scenarios are and how we set it out.

Now, what you're going to see is only a snippet of the information that's there that we talked about. So let's get into that column. Let us know when we're ready to go through the historical financials here.

Ready to go. And before we go, just a reminder for the folks on the webinar, if you have questions, please do enter your questions in the Q&A area. We'll be monitoring those and we can answer those questions as we go along.

All right. So at the starting point, really what you want to talk about, like we sort of think about this and the way we've designed this in a particular way. Right. We've got this executive view at the top of the house where we can look at historical information.

And then we've designed this particular way. It doesn't have to be designed this way. So if we had a discussion, we go through requirements design. We want to make sure we sit down with you and sort of understand how you like to adjust the data and how you'd like to visualize the data.

So we'll walk through some visuals. There's plenty of options of how we go about it. Right. Creating heat maps and so on and so forth. So, Kyle, before we go for the historical data, can we show the viewers how we look at loan balances over time by grade? If that's something we can bring up for the folks on the phone.

Yeah. So thanks, Kyle. So what you see here are loan balances by grade. And one of the things that you see here is a couple of things. You'll see a couple of drop down menus here. You'll see loan grade.

You'll see dealer type and then vintage. Right. So as we're exploring some of the historical data, these are all live things. And we'll talk about this, why these are why it's set up the way it is. You can kind of see what's happening, total loan grade.

If you can see based on historical data, one of the things that we found out in historical data is that there has been a market move towards super prime portfolio. So if we look at that, if you look at total grade and we look at super prime, you can see it going up as well on a relative basis to other loan grade.

And that then correlates to how much we see on past two loans. Right. So if you go to the past two side of things, you can see what's ended up happening is the organization has taken a market move to say we're going to get away from a little bit of credit risk.

We're going to de-risk our portfolio and move more into super prime. So if you get to that over total loan grade, you'll see the number of past two loans going down. Yeah. So in terms of and this is so then, you know, when you think about historical, one of the things that we want to also look at, we want to look at historical profitability.

As Jaime talked about the different attributes that we've got, we've got FTP baked in there. We've got allocations baked in there as well. So if we go to the metrics tab here, Kyle, if we can show what yields look like over the course of historical ratio here.

So we have our yield calculations. Yep. So you can see here what we've done here is you can see the net dealer reserve calculation baked in here. You can see what the base yield calculation over time is.

Again, these are all, this is all historical data and we can, you can slice and dice and view this information and explore the data on a historical basis. Any way you see fit, right? Whether that's by loan grade specific channel, we've got a dealer type here or vintage in a particular year.

We want to take a look at it. Right. So what is very common for auto loan portfolios is trying to take a look at historical from a vintage standpoint, whether that's month on a month basis or over a year basis.

Right. To see what that storyline looks like. And if we can show what total net interest income looks like by, by grade as well, we can show that as well. Yep. So we, again, this gives us an idea of total loan grade and what looks at by loan grade here.

Again. Now, the reason why we've set it up this way, and we'll get to this in just a second, is that this is all connected under one platform. As we go right under the hood here. So why don't we actually go there, but before we go there, can we just do a pause to see for another polling question here? Yeah.

And while we're going live with that, I think one key differentiator here that I want to expound on is this is the aggregation of, you know, each month, hundreds of thousands of lines potentially of instrument level detail.

So what you're seeing here isn't just calculations from a cell in Excel that you would, you would typically see it's, it is like an aggregation that's performing these calculations across all of the detailed data in the system.

So. We'll give maybe 10 more seconds for the polls to close here. Okay. So we talked about the historical data here. One of the things I want to do is before we go look at forecast data, I want to go right just a little bit under the hood to understand design and show folks on the webinar, how this is designed so that historicals and forecasts come together.

And when you're looking at one unified platform. So Kyle, do you want to walk the users here and what that dimension and what that design looks like? Yeah, absolutely. So if I can collapse these really quick, just to walk it down.

Essentially, you know, I wanted to take some time to go over this concept of extensibility with one stream, because I think it's, it's one of the core differentiators that separate separates it from other technologies like it.

So let's say that we have a product dimension within one stream and the FP&A team wants to budget products at a higher level of detail and accounting wants to report actuals at a more granular level of detail, a use case that we see across all industries very, very frequently.

So in other systems, managing these two requirements is fairly inefficient. So if you think Oracle EPM versus BBC SU, you'd often need to have a separate actuals and planning module, you'd have to maintain those hierarchies separately in those and then that, you know, that's going to be additional costs for licensing as well.

Or a tool more of like a planning focus tool like plan for a lot of times you need to create alternate structures in general for actual versus budget scenario modeling, create extra members for adjustments at the base level and so on.

But the thing is, with one stream, we have the ability to integrate granular levels of detail for actual reporting, and then use extensibility to aggregate this data, and then plan it at summarized levels.

So I'll just give you an example of structure here. So if I can zoom in. So if you see here under this UD6 member, this is a user defined member that we created, we have our product planning level detail.

So in this scenario, you see that we have, if we blow this out, we have different product lines, we have loans, wealth management, we can drill into it, go to consumer loans, auto loans, etc. So let's just say that, at the planning level, we just want to plan by indirect and direct auto loans, and fixed rate total ARM construction home mortgages, for example. So at this dimension member, this is the farthest level that we can go.

So at the aggregate amount, we can just plan drivers, growth rates, etc. at this auto indirect in an intersection with our chart of accounts. So this is just taking that sum, and then we can calculate on that data and report on it.

If I was to go a level deeper here, and then blow this open again, you can see that these dimensions, these members that you saw in the last hierarchy are now gray. And then these are now giving us additional levels of detail here for the requirement of actual reporting.

So let's say, you know, we need to actually like view and report on this data by term and by indirect and new for the auto loans. And the same thing could go for, let's say, like adjustable rate mortgages, fixed rate mortgages, and so on.

So basically, this allows all the stakeholders to have their requirements met through an out of the box, easy to configure capability while using the exact same hierarchy here. So there's no additional maintenance from the systems team.

So there's quicker time to value. So you can get these reports easier. If you think about it, you can scale this much quicker. So let's say that, you know, we want to blow out our wealth management division, or we acquire a bank that provides, you know, more of a service line.

So we need to expand our level of detail for planning quickly. We can just expand this down at this next level to meet the requirements of the business without having to go through a significant amount of maintenance and and efficiency to get to these points.

So I really want to own on this because, you know, I think this is a this is a functionality that and a flexibility that no other CPM tool on the market has. And if you think about the number of use cases, it's essentially endless.

It doesn't have to necessarily be, you know, actuals versus budget. It can be strategic plan at this level and then our long term plan at this level and so on. So the possibilities for it are pretty endless.

Thanks, Kyle. I think what this really kind of as a practitioner will make me think about right is really managing master data management when we think about master data management and how complex managing different hierarchies becomes different GL structure deal accounts.

You know, one of the things we constantly hear about is like, hey, our call report has this view, but we have for manager reporting internal reporting. We have this view and we're trying to reconcile back.

And when we think about our line of business profitability or product level profitability, we can't do it because we don't have the design capability or the construct available to it. So this is one of those areas when we talk about line of business profitability and product profitability.

What we talked about earlier, I want to be able to tie the instrument at the product level in terms of FTP. I need to account for capital there. I need to account for credit in there. I need to account for allocations.

This allows me to plan at that granular level or at least understand you wouldn't necessarily plan. You could plan at that level, but at least understand my profitability metrics at that product level, individual product level.

So let me pause here to see how to see if we have any other any questions at this point. Pat. No, I mean, on the we've highlighted a lot of the flexibility that one stream has. Right. If I think about the number of dimensions that are available to us, the extensibility that we're allowed to highlight and then showing the different reporting views as we get through, let's say, from gap to management to statutory type reporting.

And what we'll start to show is how we've built out some of the model overall. Yeah. Yeah. And I just do to expand on that, Pat. This is not specific to just obviously product. We can build this feature out within our chart of accounts within our entity structures and so on.

So. Well, why don't we pause here for our next polling question, please? Yeah. And as that polling question is being done, next place we're actually going to go is try to tie all this stuff that we've been talking about.

We talked about our store goals. We talked about the dimension capability within one stream. Now we're going to go to actually practical, practically show what you have, what if model actually looks like.

And we're going to talk to you some of the storyline to say what is actually happening. We talked about the four different scenarios and what is actually happened with the actual cows here. So we'll give it a 10 more seconds just to close out the poll here.

Fantastic. Okay. All right. So what you see here, what you see here is kind of an executive summary of the what if model. Like one of the things was we always think about is, well, what if my spreadsheet breaks, right? How do I do this very quickly? I have four or five key metrics or key attributes that I want to be able to adjust in order to do a model.

For instance, we talked about the Fed meeting today. Fed decides to cut. Well, I'll say, well, now that's going to change my origination plans. Okay, well, now what if I change my originations from 15% to 20%? And you know what? Because I think that rates going to flow, there's going to be more money flowing to the macro level.

I think the average car price is going to go up another $5,000. If you scroll down, Kyle, just a little bit. Pause one second. I just need to reshare. So as we look at the executive screen, one of the things is like key points that we want to be able to model, right? So loan price, you're going to have coupons spread.

You're going to have a dealer reserve calcs in there for an auto loan portfolio and average car sales, right? So we talked a little about foreign car. We haven't really gotten to the channel. That's baked in here.

You see across the top, it says foreign luxury dealers. So that is also can be changed. And one of the things about this, because going back to the accessibility we talked about was you can model this because we've set it up this way.

All right, so this is where the design part and the detail up front comes into play. And how do you want to design it so that this enables you to do the type of reporting and modeling, one-stop modeling you want to do? And, you know, the biggest benefit from a practitioner point of view is you want to avoid the offline spreadsheets.

We've all been there. I've got multiple offline spreadsheets, and I don't know which version is correct, and I want to make sure that I have it right. So this allows you to do something like this where you have, you can update car prices.

You can update, for example, new origination growth. I want to have instead of 10%. I want it to be 20%. And that instantaneously, you can see target balances. You're going to have target new balances.

All this is connected to that end. All this is live now. And you'll see that in just a moment when Kyle hits save and hits Cal. It'll take a couple seconds to run. But what's actually happening behind the scenes is it's actually running thousands of rows of instrument level detail based on the design and the structure that we just talked about.

And what the benefit of that is not only is it running this, just this screen, we're going to look through the visuals that we have in a bit. All of those are going to get updated as well. Right now that you have.

So now that we've done that, I think the one thing here to keep in mind is what you see, you know, you've got the top half is kind of your forecast for the new book. But we're also integrating the existing book that we have that I mean, talked about over earlier, right to kind of keep in mind when we think about forecasts, we're kind of blending those two things together in order to create our all up here.

So if we can go to the forecast visuals so we can show the users what that might look like in terms of storyline. So you'll recall we talked a little about the scenarios in terms of like sitting here for being the growth, the highest growth port growth scenario.

So top right, you can see ending balances growing goes from 8.3 billion, 8.8 billion. This is only a 12 month outlook. So there's obviously growth in there. But what's interesting is, as you minimize this, one of the things that you see is that you start to see is that the efficiency ratio for scenario four on the bottom right corner is actually lower.

Right. And the reason for that is we did as you dig through the data is that while our donations are going up, expenses are going at a faster clip. And because prime and subprime growth rates are higher than subprime growth rate or super prime growth rates, that expense is now flowing through in the new book.

And so automatically you can kind of see what's, you know, you've got the existing book, you had shifted to super prime, but now you've got more of a shift back to prime and subprime. You can see the efficiency ratio start to start to get some creative gap.

Now, if you were to go out 12 months, 24 months over a two year period, that gap is probably bigger now. So, and again, because we've been talking about historically designed, you can look at it by different loan grades here.

If you want to look at my super prime, that's certainly available. And you can see it's instantaneously near real time is actually changing. Right. We can start to look at profitability metrics by scenario as well as we want if we wanted to look at that.

So you've got different gear. So if we can go to the forecast by scenario here and just look at some of the metrics or profitability metrics. Yeah. So composition by band. Remember, we talked about scenario four being high grand.

So you can see composition band for this is, you know, you can see super brand. So if you now super prime, if you go to scenario four, you'll see that the super prime composition is going to be higher.

So this is all connected in terms of like kind of a forecast for outlook here. So all that said, we've talked a bit about, again, this is a design that we've talked about. Now, these are designs that can be changed.

Obviously, you may want to look at a different color scheming, all that stuff. Those are all options that are built in here that can be done depending on what you want to see. Now, in terms of metrics, are we baked in there as well? We've got those built out as well here.

Let me pause here just for a second. If there are any other comments here from the team. Okay. I think did. Go ahead, Iman. Did we have a question come in? We did. We wanted a question came in that said, what are the basis for the allocations here? Sure.

Let me take that one. I got one. So if we start with planning the expenses, the direct expenses and other indirect expenses, they are on GL for kind of the entire portfolio of auto products. That gives you a number.

And there's some variable expenses like the cost of mail statements, the cost of repo, etc. Those are then relatively assigned to this product. So as Ann was mentioning, the dynamics is if, for example, like my super prime, which is the higher balances, but, you know, and also higher volume to a degree, but not the same relativity as the others, you're going to see that dynamic.

Another way to say is, as the revenue grows, the expenses don't necessarily grow on a one-to-one. The more growth that does not eat into the capacity of the expense base is captured here. So you should always see efficiency ratio go up as you have not exceeded that excess capacity you have within the expense pools.

Yeah, I think one thing to note, right, as you think about historical, one of the reasons why we might explore historical data to understand the correlation between dependencies, right? So if we've got origination in a particular credit ban, what are the expenses related to that? That's something that can be modeled as part of this as well.

When you think about it, you don't need to manage those offline and offline spreadsheet. You build those historical analysis and have that baked into your forecast scheme as you think about like which particular assumptions are related to which other drivers.

And so you have that linkage, so you don't have to manage that separately. So that's a critical component. Typically, most of the time planning, they'll think about like what are the dependencies and this is where the platform becomes you want to keep it all in one platform.

So we just got, you know, about a few minutes left here. It's all I've got to cover in terms of the forecast planning. Hopefully this has been helpful in terms of viewing and I'll turn it over to Pap to wrap up here.

I'm last words. Sure thing. Thanks, all. Do we have another polling question we wanted to close out on? Yeah, they're right. Yeah, they do. Let's flash that up. So next question really is, are you expecting to evaluate a new planning platform in 2026? So we can take a minute to answer that.

I know some of you on the line are current OneStream customers, right? So think about how can you leverage this platform to do more with what you have today? We'll give it another minute or two and then close up the poll.

All right. All right. Let's go ahead and start our Q&A session. So again, if you have any questions on what we presented today, we'd love to hear from you. And we'd love to stay in touch as well. You know, again, this is going to be the first in a series of webcasts and webinars that we do with OneStream.

As we start to expand out on the portfolio and model more of the what if. So we want to build out a full bank balance sheet planning model within OneStream and walk you through that along the way. You know, feel free to reach out, connect with any one of us if you have questions or even if you're going through an implementation now.

Feel free to bounce ideas off of us. You know, we're here to help. Jaime on any other closing questions or comments that we should bring up. Yeah, one one thought, I think, as as we're looking to continue, as Pat mentioned, that we're going to build on what we've got here.

If there are specific things that you'd like to see, we'd love to hear from you and certainly consider those as well as we build this up. The only thing I would add is not every portfolio will be the same.

Right. So I would say is a type of modeling like this should be on the material balances. Right. Right. If you have a small product that relative to the entire plan is is it's not that much and it has some complexities, you might want to not start there.

You want to start with portfolios that are, you know, if you're more commercial focus, commercial loans are, you know, in many cases very unique. If you're a mortgage company, then, you know, start with, of course, the mortgages.

If you're diversified, then select when you might want to consider a proof of concept. I think the successes we've seen is to do a proof of concept and maybe a pilot get buy in. Another thing I would say is, you know, in today's day and age, the ability to grab cash flow data, et cetera, from ALM systems should be fairly simple.

You know, it may take a little bit of time, but the capabilities there. So you also have to balance is to what you build within one stream or what you integrate with your ALM tool. Or is there a hybrid in between? Because we have seen organizations take a hybrid approach and there's pros and cons to each.

Thanks, Tommy. We did get a question in the Q&A. So Sarah has asked if it's possible to see some of the FTP tabs on the dashboard or explain what that looks like. So, Kyle, if we want to pop back into the application, we can do a quick walkthrough of that.

Yeah, and apologies for the lag here. I think that's just the VDI working with the Goldcast right now. So let me. There we go. OK, well, yeah, here we have. Go ahead. So just to set this up for FTP here, there are three major components, right? The base rate, which is based on swap.

The term liquidity, which is based on the spread to whatever the institution can borrow. This is an A rated credit spread. And then the capital in triple acre and a triple L credit is essentially if I'm allocating capital, I need to give credit back like as I would a share deposit to that.

So that is actually reduces the overall funding costs. And then we get the total funding expense. And each of those are stored in here and calculated in here as well. But if you want to scroll, that's the base.

If you go to the next. So here we have the monthly. And essentially most of these in this metrics tab are just trending. So you can do anything with this. The dashboarding capabilities with once within one stream are incredibly flexible.

So most things that you can do within a tool like power BI, it's probably more difficult to create and to like one stream. But there's it's invaluable to be able to have it all in one application, one centralized location for security purposes and data governments and limiting the number of integrations and tools that you're utilizing.

So. All right. There's no questions related to this. I think I'll unshare. And let me let me just add, because I know we got about eight minutes, but to add to the question that had come from Sarah, each each product is going to have different FTP, right? Things that amortize like your auto loans and your mortgages are going to be done similar to this on a cash flow or average life or duration approach.

Your your shares and things that don't have a maturity will have a different methodology. So the question within one stream is if you want to replicate FTP. You really have to look at it from the basis of the product set, right, because each will be different.

So in some cases like this, we can build it in one stream to look at things like an average life or something that comes from Treasury and apply it that way or actually calculate the cash flows for things like a share money market share, etc.

Those methodologies can either come from the way ALM is treating it in your way FTP street in it and replicate that, which is different. So from an FTP perspective is a great question. A lot of it is going to be dependent on the different on the product itself and the repricing and the life characteristics of those products.

Yeah, and I think so just to add to that, if you recall the conversation we had around the design as well as the structure, the call walkthrough, right, that's where if you might plan at a particular product category level.

But when you think about profitability, right, you're now you're at a lower level dimension where you don't have to then say, now I'm going to I plan at this level, but my actions are this and now I have to try and merge them together and say, okay, how do I do a compare and contrast like after the fact, right, so that's where the design component comes in if you're doing it, if you have the information available at the product level.

Okay. Okay. I think we're just that time, Pat. So do you want to wrap up here for us? Yeah, sure thing. Thanks, everybody, for joining today's webinar and really hope you learn something new. Like I said, you know, our contact information is up on screen.

Feel free to reach out if any questions. We're here to help. We're here just to provide guidance and advice and look for us to continue this series over the course of 2020. So with that, we'll leave you and give you some time back.

Hope everybody has a happy holiday and a very happy new year. And we look forward to speaking with you sometime in Q1 next year. Thanks again, everybody. Thanks, everyone.