Welcome to the Figure Technology Solutions third quarter 2026 earnings conference call. At this time, all participants have been placed in listen-only mode, and the floor will be open for your questions following the presentation. If you would like to ask a question at that time, please press star one on your telephone keypad. If at any point your question has been answered, you may remove yourself from the queue by pressing star two. To get as many questions as time permits, we kindly ask you limit yourself to one question and one follow-up question. So others can hear your questions clearly, we ask that you pick up your handset for best sound quality. Lastly, today's call is being recorded. I would now like to turn the call over to Bryan Michaleski, Head of Investor Relations.
Good morning, and welcome to Figure's third quarter 2026 earnings call. I'm Bryan Michaleski, Head of Investor Relations for Figure. Joining me today are Mike Cagney, our Executive Chairman and Co-Founder, Michael Tannenbaum, our Chief Executive Officer, and Macrina Kgil, our Chief Financial Officer. Before we begin, I'd like to remind you that today's discussion will include forward-looking statements within the meaning of federal securities laws. These statements are based on management current expectations and involve risks and uncertainties, many of which are beyond our control, that could cause actual results to differ materially from those expressed or implied. For discussion of these risks, please refer to the risk factors in our most recent 10-K, 10-Q, and other SEC filings. These statements reflect our views today, and we undertake no obligation to update them except as required by law.
During the call, we will also refer to certain non-GAAP financial measures. These measures should not be considered in isolation or as a substitute for GAAP results. Reconciliations to the most directly comparable GAAP measures are included in the earnings release we issued earlier today and an appendix to our supplemental slide presentation, both of which are available in the investor relations section of our website. Replay of the call will be available shortly after it ends, and following our prepared remarks, we'll open the line for questions. With that, I'll turn the call over to Michael.
Good morning, and thank you all for joining. Our third quarter was big for Figure, financially and strategically. Our top and bottom line results again outperformed, demonstrating the durability and resilience of our model. Our Figure Connect platform continues to gain momentum, and the acquisition of Kiavi takes us another big step closer to our vision of bringing the capital markets on chain. I'll cover each of these topics in my remarks, as well as provide more color on our margins and capital allocation as we build on the growth we've generated year-to-date. In Q3, we grew volume 107% year-over-year at 58% EBITDA margin, with over 70% Figure Connect mix, and maintained a 3.7% net standalone take rate, exceeding everything I said we'd do on our last call. First, on volume.
In another record for Figure, our consumer loan marketplace processed $5.1 billion on an organic basis. Kiavi closed on September 1st, so the reported $5.8 billion volume reflects that one-month contribution. Our ecosystem now produces approximately $30 billion annualized run rate volume, making us the largest credit tokenizer by over two times, achieving this scale in only eight years. As projected last quarter, we saw 72% of our standalone volume on Figure Connect, reflecting the increasingly balance sheet-light nature of our business.
This quarter saw another record number of partner additions, ending at 612 standalone and 894, including Kiavi's broker partners. In addition to all of the net new growth I just described, we are also seeing expansion from existing origination partners. Those that were with us a year ago in Q3 2025 are now doing over 40% more volume this Q3 2026, a testament to the value we drive for partners.
Growth vectors we have been highlighting in the past are strengthening. We now have over 20 active depository partners and over 150 partners in our fast-growing SMB business, which crossed $500 million in run rate volume in Q3. There is more growth to come. Our pipeline is robust, including some of the largest mortgage companies and banks in the country, larger than any we work with today. As we expand our marketplace, we are also making important improvements to the platform itself. Most notably, our new agentic loan officer assistant has shown a 29% increase in funded dollars per loan approval. Figure stands out here, having trained it on outcomes only possible via our standardized data and marketplace, with tangible immediate payoff. We jointly developed this with Sierra. In just the first few months since launch, we funded roughly $260 million in loans when Sierra is involved.
The agent is now live with our wholesale lending partners and will complete full partner rollout shortly. Prospects are noticing. It is among the highest priority buying criteria they raise in meetings with us. Now turning to the Kiavi acquisition. I will touch on three primary levers and what they mean for our business: new asset classes, the integration, and the impact on our product roadmap. With Kiavi, we have added residential transition loans, RTL, and debt service coverage ratio loans, DSCR, diversifying our platform beyond HELOC into a $200+ billion addressable market. RTL loans are made to professional investors who improve properties and sell them or rent them out. Kiavi's loan origination software estimates the home value post-renovation by underwriting the property, the contractor, and the scope of work.
They do this quickly and with strong accuracy, with only roughly 50 basis points of annual losses over their 13-year history. Kiavi's other large product, DSCR loans, are also investor loans made to rental property owners. Kiavi's technology and capital markets are the closest thing to standardization for these products. The Kiavi RTL business was roughly $600 million volume in September. In RTL, we leapfrogged into our marketplace structure through a capital-efficient transaction with Sixth Street. Sixth Street bought the balance sheet and is now an originator of the assets, and Figure provides the rails. Once we open up the RTL technology to Figure's existing partners and net new partners, Sixth Street will then be only one of many originators using the Kiavi technology. Sixth Street committed to a forward flow post-transaction and to using Figure's loan marketplace, securitization shelf, and Prime platform in connection with the loans.
With DSCR, which was a roughly $60 million volume in September business, we are marrying Kiavi's lending technology stack with our existing distribution rails to scale across our partners. Our existing loan investors were already buying this product, so we don't have the same Sixth Street involvement. Our DSCR plans are among some of the most meaningful for this quarter and beyond. In short, this quarter we will launch a DSCR product that was built entirely agentically. Arvind Mohan, Kiavi's CEO and now Figure's Chief Business Officer, brings a technical background and business domain expertise to orchestrating where AI can accelerate Figure's build. This is our first product built with AI at every level, from writing the code to reimagining the workflow itself, allowing us to move faster and scale throughput dramatically.
We are closing loans within two to three weeks, and when we add Figure's approach to property verification, we think this can be reduced further. Later this quarter, we'll offer DSCR on a brokered basis to our existing customers. Next quarter, we will offer the full white label Figure Connect option for our partners. Kiavi closed ahead of schedule. The loans have been tokenized, and we've actioned over $30 million of initial cost synergies. We are now focused on upselling existing partners, of which we've already onboarded 12 and have over 30 in the pipeline. Kiavi also brought on a number of new full loan and securitized investors. Macrina will cover more here, but moving forward, all of our metrics and our IR will reflect Kiavi volume. In addition to the technology and revenue synergies, Kiavi also adds purchase market expertise to grow our product roadmap.
In fact, Kiavi does primarily purchase mortgage volume, and its platform represents about 50 basis points of the national purchase mortgage market. In summary, we expect Kiavi to be materially accretive to Figure. We added around $8 billion of volume to our marketplace, important new asset classes, and acquired market-leading technology we can offer to our partners, originators and investors alike. All this for under four-year unlevered payback period. Transitioning from Kiavi to discuss our growth resilience, I'll start with the interest rate environment. Being the most efficient provider in the marketplace has benefits when the broader mortgage industry is navigating higher interest rates. For many of our partners, rate increases harm their funnel, as marginal consumers are less likely to borrow at higher rates. We certainly feel some of that.
However, what's more notable for us is that since we're 80%+ B2B, the environment is actually improving our sales motion. Prospects want to add HELOCs, which are popular, as borrowers don't want to refinance their low existing rate loans. Right now, we're aggressively getting in front of all of our prospective B2B partners with what I refer to as polite persistence. This polite persistence is paying off. This recent quarter, we saw record volumes, continued 100% volume growth, and record partner additions, all as rates rose rapidly. The fact that we are now 70% Figure Connect further insulates Figure's P&L from volatility tied to rates. Our extremely high product velocity, a median of 10 days, including the government-mandated three-day loan rescission period, allows our partners to adjust pricing rapidly as interest rates change.
This quarter's results capture very clearly why we have been so focused on a fee-driven marketplace model. As rates rose, the investor side of the marketplace also showed strong resilience. in Q3, we processed $2.7 billion of loan sales and executed eight securitizations. We added marquee AAA buyers, including New York Life and Nationwide Insurance. In rockier markets, our loan investors value the differentiated aspects of our marketplace, the liquidity, the credit ratings, the years of history, and the transparency of the servicing data we put on chain. As we grow more, we attract larger loan buyers who bring lower cost of capital, leading to more growth. Lastly, in terms of the capital markets milestones we saw this quarter, our stablecoin YLDS is reducing the cost and increasing the speed of settlement, and we now have 28 participants in our marketplace doing so.
Our capital market strength gives our partners the confidence to grow with us in all market environments. As shared similarly last quarter, we have seen greater than 40% of our growth this year from existing customers. If 2026 was the year of the first lien, which was significantly bolstered with Kiavi, 2027 will be the year of the upsell. Two primary reasons here. We have expanded product significantly with Kiavi, and we have organic product expansion as well. Per Home Mortgage Disclosure Act data, our partner base today had greater than $500 billion in originations in 2025. Moving on from growth, as we talked about last quarter, we plan to provide updated disclosures on our margin trends as Figure Connect rapidly expands and as Kiavi comes into the fold. First, our Q3 3.7% net take rate, excluding Kiavi, exceeded the guidance we gave last quarter.
This was despite the aggressive increase in interest rates during the latter part of the quarter, again, a reflection of our increasingly balance sheet-light business model. As we have shared, the decline in take rate over the past year has been the output of the faster-than-expected success of Figure Connect, which is lower take rate, but capital-light, and drives strong margins. The Figure Connect-only take rate has been steady in recent quarters. As we have stated before, Connect take rates are an output of relatively consistent volume-based tier ecosystem fees. As partners scale their volume, ecosystem fees, and therefore take rates, drop as a percentage of volume. But take rates are not meaningfully influenced by negotiated partner rates.
The take rate forecast range we are now sharing accounts for both the introduction of new large-whale partners likely to hit lower-tier pricing, and also for the increased role of banks, who may not use Figure's capital market for all of their originations, as they keep some production on balance sheet. As we said before, growing the platform as fast as possible is how we win. It extends our first-mover advantage and makes our marketplace more robust. And as we add asset classes like Kiavi's loans, we further widen our moat. It is a trade we would make any day of the week. And that trade gets even easier when we consider contribution margin, which was reflected in our march higher toward 60% EBITDA margins. As I have shared before, we price our partnerships to contribution margin.
As we grow the marketplace to new products and become the place that lenders, and increasingly agents, search for new yield, take rate will move. Judge us by our volume growth, our platform traction, our ecosystem, and our bottom line. We earn our business on the value we deliver to customers, originators, and loan investors alike, and our products and marketplace therefore command strong margins. Before I close it out, I want to briefly cover capital allocation. Figure is in a great position with access to the equity, debt, and securitization markets, a robust balance sheet, strong liquidity, and meaningful cash flow generation. Our stock is attractive at these current levels, and we will be actively pursuing buybacks. We have been trading below our IPO price recently, despite over 100% annual volume growth since IPO, and that's before the full benefit of the highly accretive Kiavi acquisition.
While we had approved a $200 million share buyback plan earlier, we hadn't implemented that plan in the open market. Today, we're announcing an expanded $400 million share buyback plan that will start tomorrow. Given our profitability, cash levels, balance sheet liquidity, and access to the capital markets, we will be opportunistic about share repurchases on an ongoing basis, including beyond this initial $400 million program. That said, we see huge organic growth and look at inorganic opportunities as well, especially those that can accelerate our marketplace reach. We will remain disciplined, and we will allocate capital accordingly. Our business continues to fire on all cylinders, and the durability of our model gives me the confidence to share that we're going to continue to deliver rapid growth in Q4 at attractive economics. As the buyback indicates, we're in the early stages of value creation.
I'm now going to turn it over to Macrina, who will review financials. After her section, as Bryan noted at the start, our Executive Chairman, Mike Cagney, will speak. He'll share more about why the company he co-founded, Navra, matters for Figure, but doesn't change what we're building today. Knowing Mike as I do, he's thinking expansively about the future while continuing to work as hard as ever to grow Figure. We get the best of both worlds. He'll also stay for the Q&A period. But first things first, Macrina, the floor is yours.
Thank you, Michael, and good morning, everyone. This was a great quarter. Growth continued across every part of the business. Loan marketplace volume was $5.8 billion, including Kiavi, up 135% year-over-year as reported, and 107% on a Figure standalone basis. On last quarter's call, we guided Q3 standalone volume to a range of $4.8 billion-$5.2 billion. We came in at $5.1 billion, toward the upper half of that range. The Kiavi acquisition, which closed September 1st, contributed approximately $670 million of volume for the one month in our results this quarter. Figure Connect volume, including Kiavi, grew to 74% of overall volume, up from 65% last quarter and from 46% a year ago. Ecosystem and technology fees are now roughly 40% of our revenue.
These more than tripled since a year ago this quarter, after becoming the single largest contributor to our P&L last quarter, and we expect that share to continue climbing as the business shifts further towards fee-based, capital-light volume. Prime, which was previously known as Democratized Prime, ended the quarter with lender supply of $825 million and matched offers balance at $588 million, including third-party borrow of approximately $200 million. As with Figure Connect, every dollar of third-party demand on Prime is a dollar choosing our infrastructure over a traditional warehouse line, and we expect that to keep building. Net income was $89 million, flat year-over-year. As a reminder, last year, primarily from a recent merger with Figure Markets, we recognized a one-time $31 million tax benefit, which increased net income to $90 million.
Adjusted EBITDA was $164 million, up 91% year-over-year, with margin expanding to 58% from 55% a year ago, continuing progress toward our medium-term 60% target. With acquisition of Kiavi in September, we expect near-term EBITDA margin to decline below Figure standalone year-over-year margin growth, as Kiavi pre-Figure was running at or below 40% adjusted EBITDA margin. As we head into 2027, we expect revenue synergies and operating leverage to kick in and normalize back up to overall medium-term targets. Take rate is an output of our channel and product mix, not a target we manage to, as Michael has discussed. We assess our marketplace partnerships and product diversification to contribution margin and to the benefit each brings to adjusted EBITDA.
This quarter, I want to provide additional color on gross take rate, especially for Connect, as we have shorter duration products such as residential transition loans starting in September. As a reminder, Connect volume carries no customer acquisition cost. That is why we are shifting our focus to gross take rate going forward rather than net take rate. We expect gross take rate from shorter duration RTLs to be in the 1.5%-2% range, whereas for longer duration Connect HELOC, we expect to be in the 2.9%-3.4% range over the next six to 12 months. Figure Connect may carry a lower gross take rate than our other channels, but has strong contribution margins and the lowest capital intensity. With higher volume year-over-year, adjusted EBITDA margin continues to expand even as take rate decreased on Connect.
Contribution margin is profit that each dollar of incremental revenue generates, taking into account operating leverage and efficiency. In future earnings calls, we plan to discuss contribution margin upfront with color on gross take rate. To share with you net take rate, what we had been reporting previously for one final time, net take rate was 3.5%, including one month of Kiavi results, and 3.7% standalone Figure, within the range we have discussed previously. In terms of additional expense from the Kiavi acquisition, one-time M&A and restructuring costs were $22 million in total that were incurred in Q2 and Q3. Of the $35 million run rate cost synergy target we set at announcement, we have actioned $30 million at close, with the remainder to be actioned over the next 12 months. I will keep you posted in future quarters with progress on cost synergy.
Our cash position remains strong post-acquisition of Kiavi, with $1.59 billion in cash and cash equivalents. We have been deliberate in building our capital toolkit, starting with our IPO over a year ago, and in Q3, we raised rated unsecured debt of $600 million. We are also closing on a $465 million revolving credit facility this month. This is in addition to our ability to generate cash flow each quarter from strong adjusted EBITDA. We approximate 50% or so of adjusted EBITDA converting to cash inflow, taking into account income tax payments and interest expense on unsecured debt. Turning to other parts of our balance sheet, our loan receivables of $644 million include approximately $395 million of loans Figure has placed as collateral on Prime from the beginning of the year, as well as approximately $16 million of DSCR loans that we acquired from Kiavi.
Typically, loans held by Figure stay on the balance sheet for a few weeks before being sold or securitized, and fluctuates as a percentage of volume during that month. On share repurchase and returning capital to our shareholders, we have continued our routine practice of retiring shares withheld for employee tax obligations on vested equity. $35 million this year, which is approximately 1.1 million shares. That is ordinary course and not part of the announced share repurchase program. As Michael outlined, we are beginning to execute on share repurchases, including our entry into a 10b5-1 plan to begin executing the up to $400 million authorization with cash flow we have generated this year and expect to in the future. Credit quality has remained stable across the first three quarters of the year. Underwriting discipline and portfolio performance held consistent quarter-over-quarter, even as volume scaled at triple-digit rates.
Our underwriting and pricing platform that we offer to our partners is nimble. With HELOCs, our largest product, funding in a median of 10 days, we can reprice the loans as overall macro markets move faster than other lenders that fund on average in 45 days. We also have been hedging rates for the loans that stay on our balance sheet for a few weeks before they are sold as whole loan sales into Connect or securitized. These actions have allowed us to continue to execute consistently through rate volatility for loan sales directly on Connect or via our securitization shelf. Looking ahead for the fourth quarter, we are establishing our Q4 loan marketplace volume guidance in the range of $6.2 billion-$6.7 billion, and combined gross take rate guidance of 3.1%-3.4%.
I want to briefly touch on the overall Q4 outlook in terms of seasonality and ongoing expenses coming from the Kiavi acquisition. As a reminder, our HELOC, RTL, and DSCR products experience seasonality, similar to the broader mortgage industry, and we have historically seen the winter months from November through February be slower than Q2 and Q3 volume levels, as much as 10%-15% from Q3 to Q4 sequentially. Last Q4 was an exception as Figure experienced an increase in volume sequentially from Q3. This was due to IPO tailwinds that helped us originate more Figure direct-to-consumer volume, as well as the addition of more Connect partners, including a few whales earlier in the year that helped offset the natural seasonal decline. Returning to seasonal norms in this year's Q4 suggests lower sequential volumes from Figure and Kiavi, which would also result in lower adjusted EBITDA margin.
Q4 will also be the first full quarter post-acquisition as one company. We expect to see incremental $18 million-$23 million of adjusted expenses in Q4 from the acquisition, as compared to Q3 with one month of expenses from legacy Kiavi overhead. As a reminder, Kiavi prior to Figure acquisition was at or below 40% adjusted EBITDA margin. Moving forward, as seasonal pressures abate and as we execute on planned revenue synergies in 2027, we expect adjusted EBITDA margin to improve meaningfully, reflecting the full benefits of the Kiavi acquisition alongside continued rapid growth in our HELOC business. Before I go, I want to close with the same note I opened on. This was a genuinely strong quarter across volume, revenue, and margin, and consistent credit quality.
Kiavi is integrating ahead of plan, Prime continues to build real third-party demand, and our balance sheet is stronger and more diversified than it has ever been. With that, I will hand it over to our Executive Chairman, Mike Cagney, who will share more on the longer-term vision before we open the line for questions.
Thanks, Macrina. Last month, most of you likely saw that I launched a new company called Navra. I know we periodically get questions about how I prioritize my exec chair role at Figure and my entrepreneurial initiatives. As Michael said earlier, I am full-time at Figure and leading our zero-to-one efforts across blockchain. They are not mutually exclusive roles. Figure is on the leading edge of bringing the capital markets on-chain, and the time I spend adjacent to Figure is focused on that objective over the long term. With that, I want to spend a couple of minutes on Navra, and specifically on why it exists and what it does for Figure. Figure uses blockchain in three ways.
We use the immutability of the ledger to take out transactional expenses, we use bilateral marketplaces that self-settle and self-clear as the foundation of our capital markets, and we use DeFi to reduce financing expenses. These three things underpin our ability to rapidly grow our lending volumes in our Figure Connect marketplace. They enable our flywheel and our competitive advantage. The DeFi piece has always had a simple thesis. We take out the middleman, and both sides do better. Take our warehouses as an example. We pay SOFR plus 250 basis points to the street. The street finances those warehouses in the asset-backed commercial paper market at roughly SOFR plus 15 basis points. That 235 basis point spread pays the warehouse provider to manage risk and liquidity. On blockchain, the protocol does both. You cannot double pledge an asset on-chain, and the collateral has a referenceable liquid market.
The spread gets split between borrower and lender instead. This has become more than theory. We have roughly $600 million financing HELOCs on-chain today at SOFR plus 120 basis points, and this month, we plan to launch a AAA ABS issuance on-chain where the buyers can finance the bond at better terms than through traditional prime brokerage. The model works. The constraint is the size of the market, not the economics. We estimate from public data roughly $6.5 billion of stablecoin available to borrow across the top five DeFi protocols, against upwards of $2.5 billion a month of assets Figure is putting on-chain. If we financed our flow through DeFi, we would consume available capital in months. To be clear, we do not rely on DeFi for growth or margins today. It is an interest expense advantage. But over time, it bears on something bigger.
Cheaper bond leverage means better securitization execution, and that shows up as both margin and defensive moat. The question is how to make the market bigger given the obvious institutional investor aversion to the wallet choices today. [Wallet UX] is unfamiliar at best. Nobody has solved self-custody with qualified custody, and wallet lack enterprise controls, who can do what, with an audit trail that connects to core systems. These are product problems. Michael and I discussed this, and we agreed on two things. The problem is worth solving because solving it widens the capital market for everything Figure puts on-chain, and the solution has to live outside of Figure. A wallet owned by one issuer caps out that issuer's assets because no other issuer will adopt it. That is Navra. The Figure ABS deals are the first proof point.
Real yield on real assets is what pulls TradFi dollars on-chain, and every dollar that comes over is a dollar that can finance Figure collateral. Figure is aligned with Navra in two ways, directly via a 10% ownership stake, and more importantly, indirectly as Navra's success brings on-chain capital for Figure loans. A second company was not on my bingo card after the IPO, but Navra widens the market for what Figure builds, and that is why I am spending the extra time to put the solution into the marketplace. With that, we are now ready for questions. Operator, please open the queue.
Thank you. The floor is now open for questions. At this time, if you have a question or comment, please press star one on your telephone keypad. If at any point your question has been answered, you may remove yourself from the queue by pressing star two. Again, we kindly ask you limit yourself to one question and one follow-up, and pick up your handset when posing your question. We will take our first question from Ryan Tomasello with KBW. Please go ahead. Your line is open.
Hey, everyone. This is Huan Tang on for Ryan. Congrats on the stellar results. Just on supporting the third-party assets that are not originated through Figure's LOS, particularly for those depositories you called out, remind us again what sort of economics you are expecting from this, and how do you expect to build out the investor demand for these third-party assets?
Thanks for the question, and thank you for the compliment. With third-party assets on Figure, what we are doing is starting with Prime. So you have seen us take originators on HELOC and home equity, as well as in the SMB space and as well as in auto. Those are all the categories that we announced. We have shared already that the economics of Prime are roughly around 50 basis points. We also expect to do securitizations in those asset classes, which will bring additional ecosystem fees. Then we have banks and other non-bank originators that will use Figure Connect as a way to do whole loan sales and will extract ecosystem fees from those transactions as well. So that is the rough economic model. The reason why we started with those asset classes is we have some lighthouse originators that are participating.
We bring on the investors, we bring on the securitization, and then we start to flow lots of volume through. I think in the coming quarters, you will see us talk more about that as we have more proof points and share more about what it means for economics long term. But as Macrina shared in her prepared remarks, most of that benefit is going to flow through that ecosystem and technology fee line item, which has been the priority for Figure, and again, reflects that balance sheet-light business model.
Okay, that is helpful. Just on this HELOC Connect and near-term gross take rate outlook guiding to roughly 2.9%-3.4%, that seems to be a decline from recent trends. I know you called out that that is being driven by these large whales. Could you give us some context around what large kind of whale pricing looks like and what that pipeline in the near term looks like as well?
Yeah, as we shared in the prepared remarks, our pricing doesn't reflect negotiated partner rates. When we talk about whales and pricing, what we're talking about is an expectation that we hit high volume tiers or lower pricing quickly. You've seen some of that variability in our take rates over the past couple quarters, and we give ourselves that flexibility, one, for those large whales that come on, right? We don't expect those to come on at worse economics than other large whales. But two, we also give, which relates to your first question, we also give us some room for banks which may want to use our origination technology but keep some of the loans on balance sheet.
When they do that, they're not going to pay the same pricing, they're not necessarily going to securitize the loans, and they may in fact sometimes service their own loans. Because of that, we wanted to give that flexibility. But at the end of the day, as you've seen in recent quarters, Figure Connect take rate has been relatively steady, and we're feeling really good about where we stand economically in the coming months.
Okay, thank you, and thanks for taking the questions.
Thank you. We will take our next question from Patrick Moley with Piper Sandler. Please go ahead.
Yes, good morning, and thanks for taking the question. I just had one on the 4Q volume guide. The range looks consistent with a normal 10%-15% seasonal decline that you described. Michael, I think you also said that rates moved up sharply, and you are feeling some of that in the consumer funnel. Is there any cushion in the 4Q guide for rates beyond seasonality? If you could give us any update on what you are seeing beyond that in the fourth quarter, it would be helpful as well. Thanks.
Thank you. I will turn it over to Macrina, but just to quickly address that comment, what we do see is when there are rate movements, and there has been continued rate movements through this quarter, certainly at a greater velocity than has been in prior quarters, we are repricing loans all the time. That is a benefit of our platform. Most of our partners look to us to help on pricing, and that is a service we provide. When there is volatility in pricing, there is always going to be some near-term impact in the funnel, and that is what I meant in the prepared remarks. In terms of volume guidance and seasonality, I will turn it over to you, Macrina.
Good morning, Patrick. I did mention in my prepared remarks that we do expect to see seasonality of 10%-15%, as you mentioned. This quarter is the first time we have Kiavi volume coming through, and we wanted to be able to give a range that we felt comfortable with. I expect us to be somewhere down the middle in terms of guidance. As always, that is how we are thinking about it. This is a new quarter with Kiavi coming through with their RTL and DSCR loans, but we are very confident with the pipeline that we are seeing.
Okay, that is helpful. Just as a follow-up, Macrina, you mentioned that you are shifting the disclosure from net take rate to contribution margin going forward. Since that is kind of how you price these partnerships, could you give us a sense today for what the contribution margin looks like on Connect HELOC versus non-Connect versus RTL? I guess I will just leave it there. I think that would help us get a little bit more comfortable with a 3% Connect dollar looking as good as a 6% balance sheet dollar.
Sure. Two parts, really. Going forward, we want to talk about gross take rate instead of net take rate, and that is because our Connect volume is the majority of what we do, and that's what you're seeing in our loan marketplace volume. In terms of contribution margin, we will talk more about the details of how contribution margin comes together in our next earnings call. We want to be able to see how Kiavi continues to season with those loans coming through and then be coming out with more information in the next quarter earnings call.
Just to give you a little bit of a flavor, as Michael mentioned, contribution margin is something that we look at when we price for our partners and when we look at each of the different products. At the end of the day, our contribution margin does need to be higher than our adjusted EBITDA margin in the medium term. It's something that we look at as part of our platform and discussions with our partners.
Okay, great. Thank you, Mike and Macrina.
Thank you. We will take our next question from Craig Siegenthaler with Bank of America. Please go ahead.
Hey, good morning, everyone. Hope you're doing well. I wanted to start with the long-term growth one. How are you sizing your unique TAM for Figure Connect, given that most of your partners are not in the bank channel, and are not traditional HELOC originators?
Hi, Craig. Thank you. For us, we've often shared that we have 10 of the top 20 independent mortgage banks. One of the reasons we share that stat candidly is because we don't have a large number of the top 10. We do see, as I shared in my prepared remarks, meaningful traction with some of the largest non-bank originators, as well as some of the largest depositories in the country. For us, we see meaningful growth ahead there. I'll also point to a few other levers of growth that we've been sharing. One has been our SMB business, where we see huge amount of white space. We actually have not yet brought an SMB originator onto Figure Connect. What we've shown over the past two years is that when Figure Connect's involved, there's significant incentives for partners to do more volume with us.
We think there's significant upside there. We also have been building a lot in the home improvement space, and with the addition of Kiavi and the purchase technology I mentioned, we see a big amount of opportunity to combine our product set there. Lastly, I mentioned a few numbers in the call that I'd point you to. One is the $200 billion addressable market just for DSCR and RTL. We think Kiavi is the market-leading technology in that business, in both of those businesses, so we expect significant upsell opportunity.
Again, we've proven our ability to do that. Lastly, I shared $500 billion in origination per HMDA data for our existing partners. That would include a significant amount of business that's addressable today, in particular first lien, and in particular purchase, which we've shared already a few calls ago, now bolstered with Kiavi technology, is a product we're launching in the near term.
Thanks. Just for my follow-up, I am curious if you could provide an update on your ability to expand into other consumer loan segments post-Kiavi. I am specifically thinking about auto after the Agora partnership and non-prime auto. I am not that familiar with auto, but I think a lot of the relationships are with big banks, top of the house, across auto brands. How do you sort of break into auto and gain meaningful share, if you do not sign up the big banks?
Yep. In two ways. One, we have a partnership, I called it a lighthouse partnership, in a previous answer, with Agora. Their role is to actually interface with a number of dealers. They have significant amount of volume growing through their platform, and what we are doing in the coming months is a securitization with them. We also have significant amount of our existing whole loan investors that buy mortgages, but also buy auto.
So we already have the pieces of that marketplace, and what Agora will do is start to originate more volume, again, because they are capped on their capital markets and their ability to grow until— that is why they came to Figure, that is why they see the opportunity in our capital market. Again, very similar to the people that came to us for HELOC, because there was no, and is no Fannie Mae for home equity.
People want liquidity, and as we bring those investors, both in securitized and whole loan form, Agora can grow its business, working with getting more volume from their existing dealers, and that is a big factor in auto. We are going to bring that standardization to the space, because ultimately, it is very difficult, especially outside of the banks, to your point, to build a non-bank capital markets business, but a huge amount of auto origination is non-bank, and we feel really good about our position there. Similarly, in SMB, I know you did not ask about that, but it is a very similar market, where we now have 150 different originators.
Yes, they are doing our HELOC today, but they all have other whole loan products that they are looking for capital markets takeout for, and we can again provide that standardization layer through Figure Connect, whole loan sale securitization, and we have a lot of momentum there. Look for more from us in the coming quarters.
Great. Thank you.
Thank you. Once again, if you would like to ask a question, please press star one on your telephone keypad now. We will take our next question from Dan Dolev with Mizuho. Please go ahead.
Hey, Michael, Macrina. Really, really strong results here. Congrats. Well-deserved stock reaction. Wanted to ask about the partners and the year of the upsell. Maybe can you help us understand what you need to unlock within your partners to be successful in the year of the upsell, and how did you size the opportunity in your partner set? And congrats again.
Thank you, Dan. For upselling the existing partner base, one, I will just remind you, it is a motion that we are very familiar with. It has been a huge part of our success. I think last quarter, we talked about a few originators that had been with us for years, like Mutual of Omaha. NAF, New American Funding, those are originators that have been with us for many months, and we unlocked them via Figure Connect and other upsell efforts in Q2. So we have proven our ability to take what could look like, I know our data is very transparent, relatively small originators and find bigger shares of wallet. So we have done that in HELOC. There is a lot more room in HELOC, and I shared that in my prepared remarks.
But we also now have the two very high-quality products from Kiavi that are the closest thing to standardization and liquidity in both RTL and DSCR. Our partners do a lot of those. We have a waiting list. There is a lot of demand for those products. At the same time, just as we have unlocked smaller balance refinance and cash-out refinance with our home equity alternative, we are going to do something similar in the purchase space. Purchase volume has been resilient in this rate environment on a relative basis, so we see significant opportunity there, and that is why we are comfortable talking about 2027 as the year of the upsell.
Great. Thanks, and congrats again.
Thank you. We will take our next question from James Yaro with Goldman Sachs. Please go ahead.
Good morning, and thanks for taking the question. I was hoping you might be able to talk to the buyer base growth trends more recently, as well as the reactions from that cohort to the Kiavi product. Do they exhibit the same demand for the Kiavi loans as for the core Figure HELOCs?
Yes, happy to. Well, as you know, we worked on the transaction with Sixth Street, which had been a key player in our guarantor vehicle and was also a key buyer for Kiavi. There is some overlap, and there is overlap among other existing investors. I also shared in my prepared remarks that with Kiavi, we brought on new investors into our marketplace, and one of the benefits of the Figure Connect flywheel is that investors want to buy more when they work with us. We are confident that some of, I think there has been at least a couple hedge funds that are now on Connect that were not previously that are now excited to buy more loans and securitized products from us. What we have seen in the past quarter is, and also all year, has been really strong securitizations.
I think we shared a little bit more in our new supplemental financial materials out this morning, but we have had really both robust securitization activity and whole loan activity. We feel strong about that activity continuing through the end of the year. We have also seen even more foreign demand recently. Continue to look for us to have more capital markets milestones in the coming months, and we are feeling really good about our position, the liquidity that we bring to the marketplace. That stands to be true even more so in more rockier environments.
That is great. I wanted to touch maybe a little bit on Sierra, which I thought was an interesting product rollout. Could you just maybe expand a little bit on exactly how this works across the platform and what allowed it to drive the 29% increase in the funded dollars per loan approval?
Thank you for giving me that opportunity, James. We are really excited about this. It is only rolled out in a small portion of our business, and partners have been clamoring for this. It is very special for a couple reasons. One, as I shared, ultimately, and you saw this in the SaaS era, right, there is lots of mortgage SaaS that has not lowered the cost to originate. There is tons of SaaS out there, and at the same time, cost to originate a mortgage has gone to $12,000. We have obviously taken a very different approach with a very different cost profile, and the agentic era is going to bring the same problems. Just because you throw agents at the problem, that is not going to lower the cost. What we have done is something a bit different, right?
We have trained this Sierra agent on our specific business model, our marketplace, which is much simpler, and the outcomes, therefore, are much more dramatic. Another thing that we did is we specifically trained it on conversion, right? Lots of people will train things on volume. They want the bot to make as many calls as possible. We trained it on conversion to optimize for that 29% that I shared, and as a result, we've done something really meaningfully different, and we are now hearing in many of our sales conversations, as I shared, that people want this. This is part of the buying criteria, and it's a real differentiated moat for Figure, so we're really excited about it.
Thanks a lot. Very helpful.
Thank you. We will take our next question from Kyle Peterson with Needham. Please go ahead.
Great. Good morning, and nice results. Thanks for taking the questions. Wanted to start out a little bit and ask on the growth, particularly with the different mix of assets now with Kiavi in the fold here. I guess, is there anything we should be mindful of on either the RTL or DSCR growth rates or interest rate sensitivity? Obviously, they're smaller businesses, so kind of law of numbers makes it so it should be easier to grow, but the momentum you guys have had in HELOC is so strong. So I guess just how should we think about the different kind of push-pull dynamics in the mix of the volume, at least in the near term?
I'll start, and Macrina, feel free to add. When we bought Kiavi, we assumed that we were in a very almost record-low housing inventory environment. That was baked into the assumptions and that four-year payback. Any increase or improvement in the market will just be upside to that four-year payback that we've shared, that unlevered payback. This is definitely a dynamic that we understood. At the same time, Kiavi has had, at least prior to Figure, its business model and approach was a direct-to-investor originating model.
They were really focused on taking their secret sauce, which is meaningful, their technology, and using it only for their own distribution and brand. We think the growth is going to come by taking that secret sauce and unlocking it for our very large and increasingly growing partner base that's eager to originate those assets. This is all part of our expectation, and also is reflected in the guidance that we've shared. But, Macrina, I'll let you add anything.
Just one item to add. Historically, and we've mentioned this before, Kiavi had been growing at about 25%-35% in terms of volume. That does not include the revenue synergies that Michael had mentioned. We're optimistic about what's to come in 2027.
Great. That's really helpful. Thank you. Then, I guess for my follow-up, I want to switch over to funding. There's been an area you guys really shined, but I guess, how are you guys thinking about whether it's whole loan sales or securitizations, especially given there has been a little more volatility in the debt capital markets in the last few weeks? So I guess, does this make whole loan sales potentially a little more attractive, or how are you guys thinking about kind of matching execution with investor demand for these assets?
Yeah, liquidity and flexibility for both Figure as well as for the Figure Connect marketplace is our superpower. What we're ultimately offering is permanent takeout to our partners, and they're looking for us to standardize a capital market that has previously not been so. As a result, we're always going to look at, and our partners are going to look at, best execution across securitization and loan sales. That's a very natural thing. I think what you see both in the additional disclosure that we shared, what you saw in particular in Q3, and what you're hearing from both the tone of our voice as well as the information we've shared in the prepared remarks, is we feel that this is a huge competitive advantage for Figure. There is a lot of investor demand for what we do.
We've been doing it for years with great ratings, strong credit performance, and a lot of work by our capital markets team to engage with investors and ensure that they're excited about the high-quality collateral that our marketplace partners produce. We're always going to have that flexibility, and it's a huge advantage of the marketplace that someone can look to the securitization market, they can look to the whole loan market. We have increasing conversations, as I've shared, with depositories coming onto the marketplace that are liking the volume that we see. One of the things that we even show is how many of our customers are their customers, right? A lot of the non-banks are originating customer transactions from consumers that are ultimately depository customers.
When people see that, when depositories see that, it makes them interested in the assets, and it makes them interested in engaging with us. That's been a big shift we've seen as we've started to scale, and it's driven depository interest, which is just another lever that we have in the coming months, as well as insurance, right? We shared two insurance names in our prepared remarks. We see more opportunity ahead, pension funds as well. Thank you.
I would just add that the securitization or whole loans or pre-funded deals, it really depends on where the demand is and where the buyer market is. The other part I would also add is that the players that are contributing to the securitization are either our origination partners who can make that decision themselves depending on what is more attractive to them, in addition to the whole loan buyers who bought in the past and then can contribute to these securitizations.
Excellent. Thank you very much.
Thank you for your question. We will now take our next question from Randy Binner with Texas Capital.
Hey, good morning. That's actually well-timed. I'd like to follow up on the capital markets question. I think it looks like you're back in the market with a securitization in the last few days. The question, and kind of building off your comments on confidence in whole loan sales and securitization, and there's a good slide that's been included on that. When credit markets got wider over the past month or so, were you able to pivot to whole loan sales? If so, can you speak a little bit to kind of that process and how, like what the frictions are or how easy it is to pivot, if that's the way you can handle markets when they get a little bit wider?
Yeah. To start, we are in the market, and we do see some recent strong AAA prints. Expect that despite the interest rate volatility and just the overall amount of noise in the capital market, Figure has done about 20 securitizations to date, and we expect many more in the coming months. We also do not see, just to address your question, any friction between moving from a whole loan sale to a securitization. At the end of the day, Figure Connect is a marketplace, and what you see is auctions that people are running, and they're always getting feedback on where the market is. At the same time, there's going to be times where whole loan sales and investor demand there is going to be stronger. It's a dynamic market.
Especially as new investors come on the platform, that will change those dynamics at any one time. I think it's really important to share two things. One is, Macrina just talked about those pre-funded securitizations, and what that shows you is there is so much investor demand for this standardized product, and we're really one of the, I think we're the only non-government agency that's been able to pull something like that off, and we have more capital markets milestones coming. We also ultimately have really significant permanent capital demand, right?
At the end of the day, whether the loans are going whole or whether they're going to securitize form, at the end of the day, there's been this huge rise of insurance companies and asset management, non-bank financial institutions seeking yields. There's retirement, all sort of the tailwinds that have been growing that business. We see none of that changing. If anything, and this is something we shared previously, the diversification that we bring with high-quality consumers at high FICO with lots of home equity behind them is increasingly attractive in any market.
All right. Great. Thanks for the response. If I could just squeeze one other one in. The G&A expense line just continues to outperform, at least my model. I know you have a very scalable platform there that's been built over time. I guess the question is, as we think about G&A expenses, your largest expense line in the income statement, relative to revenue, does it get a lot better than this? Just how do you think about that kind of expense control margin? It's been really good, kind of going into 2027. I'd just be kind of interested in thoughts on that.
Sure. Good morning, Randy. Just a few items on the G&A expense line item. We do have the Kiavi-related M&A costs flowing through there for the most part. There's a lot of legal and accounting fees that went through. I mentioned on the call, it was about $22 million in Q2 and Q3. So, you're seeing a large number there coming across. The other part is that there's some stock compensation expenses for a lot of the executives are going through the G&A line item as well.
So, you're seeing that pretty much stable across 2026. We expect that to continue as a percentage of revenue, stock compensation expense is going down because the absolute dollar value is pretty stable. I would highlight those two items. At the end of the day, this is our second year as our public company. We will continue to look at optimization, but this year, I would say we also have SOX controls going in, so we have some additional professional fees coming through.
Okay, that's helpful. Thank you.
Thank you for your question. Our next question comes from Dan Fannon with Jefferies.
Thanks. Good morning. Michael, I wanted to follow up on your comments around the pipeline for new origination partners and how that maybe compares to a year ago in terms of the type of partners that you're talking to now versus previous periods.
As I shared in the prepared remarks, there are conversations today at a level of partner that we have not yet been able to work with. Not due to lack of trying, but as we have scaled, as we have gone public, and as I shared, I think, a few quarters ago, as frankly the growth of the non-QM and home equity space has materialized, you just have more people that are looking for standardized capital markets. So, what we are offering is increasingly attractive to larger partners. So, you absolutely have a level of partnership and size of partner that we have not previously engaged with. But at the same time, I am very close to the sales process, and we feel significant runway across smaller partners. We feel that in the SMB space. We feel that in kind of the broad non-bank space as well.
We are also starting to see, as I mentioned, significantly more activity with depositories, so expect more to come there. Flagstar was very much a light I keep using this word lighthouse. I am always doing ocean analogies, the New England heritage. Anyway, it is a strong That was an example for many other depositories that someone that is publicly traded and well-known is working with us, and that has unlocked a lot. Sales cycles are slower there, but pipeline is very robust.
We had, prior to our IPO, started to do some in the fintech space, which has just been crowded out with some of the other success we have had. But there are absolutely a large number of housing and home improvement-related fintechs that are looking to grow their businesses, and we can provide really interesting point-of-sale financing. Again, there is $35 trillion of home equity out there, and in any interest rate environment, our product is going to be the lowest cost and also the longest. So, the combination of those two things means it is going to be the lowest payment, and there is always interest in low payments.
Great. That is helpful. Thank you. Then as a follow-up on Kiavi, could you give us the latest run rate EBITDA? I think the last one we have is $100 million, maybe for 2025, and then also the $30 million of expense synergies. What is a reasonable time period you think you will execute upon that?
I will start, and Michael, feel free if you want to add anything. As you mentioned, we have said Kiavi had $100 million of adjusted EBITDA before joining Figure. We have not given guidance or indication on what Kiavi's EBITDA will be in the future. This is the first month that we have done the integration. We will come back to you with more information in the next earnings call as we move forward is really how we think about it. Oh, and in terms of the $30 million synergy, we have actioned the $30 million at close in September.
The majority of that has gone through when we onboarded the company. Some costs have gone to the JV, some have come to us, Figure, and then we are continuing to work through the rest of the synergies. The other part I would mention is this is expense synergies, and we have talked about revenue synergies and being able to onboard partners to use these products. That will continue to improve the operating efficiency and the leverage that we expect to see in 2027.
Thank you.
Thank you. This concludes today's Figure Technology Solutions third quarter 2026 earnings conference call. Please disconnect your line at this time, and have a wonderful day.