Better Home & Finance Holding Company (NASDAQ:BETR) Q3 2025 Earnings Call Transcript

Better Home & Finance Holding Company (NASDAQ:BETR) Q3 2025 Earnings Call Transcript November 13, 2025

Operator: Ladies and gentlemen, thank you for joining us and welcome to the Better Home & Finance Holding Company’s Third Quarter 2025 Results Call. After today’s prepared remarks, we will host a question and answer session. If you would like to ask a question, please raise your hand. If you have dialed into today’s call, please press 9 to raise your hand and 6 to unmute. I will now hand the conference over to Tarek Afifi, Corporate Finance at Better. Tarek? Please go ahead.

Tarek Afifi: Hello, everyone. And welcome to Better Home & Finance Holding Company’s third quarter earnings conference call. My name is Tarek Afifi, on Better’s Corporate Finance team. Joining me today is Vishal Garg, Founder and Chief Executive Officer of Better. In addition to this conference call, please direct your attention to our third quarter earnings release, which is available on our Investor Relations website. Also available on our website is an investor presentation. Certain statements we make today may constitute forward-looking statements within the meaning of federal securities laws that are based on current expectations and assumptions. These expectations and assumptions are subject to risks, uncertainties, and other factors as discussed further in our SEC filing that could cause our actual results to differ materially from our historical.

We assume no responsibility to update forward-looking statements other than as required by law. During today’s discussion, management will discuss certain non-GAAP financial measures, which we believe are relevant in assessing the company’s financial performance. These non-GAAP financial measures should not be considered replacements for and should be read together with our GAAP results. These non-GAAP financial measures are reconciled to GAAP financial measures in today’s earnings release and investor presentation, and when filed in our quarterly report on Form 10-Q filed with the SEC. Amounts described as of and for the quarter ended 09/30/2025, represent a preliminary estimate as of the date of this earnings release and may be revised upon our quarterly report on Form 10-Q with the SEC.

More information as of and for the end of the quarter ended 09/30/2025, will be provided upon filing our quarterly report on Form 10-Q with the SEC. I will now turn the call over to Vishal.

Vishal Garg: Thank you, Tarek, and welcome to our third quarter 2025 earnings call. This has been a pivotal quarter with significant developments for Better as the leading AI home finance company. We have rapidly evolved from a dominant direct-to-consumer business into a platform powering the entire home finance ecosystem, both for consumers directly and increasingly through our growing list of institutional partners. Partners include both local mortgage lenders and financial institutions, and we empower them with our Tinman AI platform to serve their customer needs better. In summary, over the last couple of months, we announced three new partnerships, which we see as deeply validating and believe will meaningfully expand our market reach across the home finance landscape and drive profitability as we track to breakeven adjusted EBITDA by Q3 2026.

We are already pacing to fund $500 million in monthly volume as a result of the growth through these partnerships, and that momentum is accelerating rapidly. In the next six months, we are comfortable that this will double to at least a billion a month in funded loan volume. Our progress comes mostly from our soft launch during which we have marketed the power by Better solution to only a small fraction of our partner’s customer bases and seen great success. This partnership represents the most significant opportunity in Better’s history. Excitingly, thanks to our strong unit economics and best-in-class experience, powered by Betsy and Tinman, our pipeline of additional partners continues to expand rapidly. We expect to share further updates on these partnerships and additional ones in Q4.

Our pipeline of Tinman AI platform clients keeps expanding as the industry is seeing what our platform could deliver. We are in late-stage conversations to land partners in some of the biggest, most strategic verticals in consumer finance. Examples include one of the top home improvement lenders, two of the top servicers in the country, one of the top personal lenders, and an additional mid-sized bank. Additional partnerships will add an additional 10 million American homeowners to whom we can algorithmically qualify market mortgage and home equity products. All of these events validate our strategy of diversifying our distribution channels, as our AI-driven platforms, Betsy and Tinman, deliver the lowest unit costs while providing the best experience for both customers and partners.

This gives us strong conviction that our peak volumes in this rate cycle should comfortably exceed those achieved in the last rate cycle when we originated approximately $60 billion in one year, almost $5 billion a month. We have built a platform that is AI-first. We are one of the few players, if not the only one in the US, with a single full-scale tech stack. All in one place, all in one flow, and entirely API-able via our proprietary MCP server, the only one in the mortgage industry, to AgenTeq AI, which allows us to deliver a better experience at lower cost, scale faster than anyone else, and really continue to define the future of this $15 trillion industry. Better is the network for the largest tangible asset class in the US: residential real estate.

On one side of this network are the end consumers directly, and on the other side are consumers using the Tinman AI platform similar to that of merchants on platform networks like Stripe, Visa, and Mastercard. On the other side, our investors seeking to buy cash flow-producing assets secured by US residential real estate. We are the matching, processing, and fulfillment engine in between the two sides of this network. Our engine is called Tinman, which uses machine learning to triangulate consumer attributes, property attributes, and the unique criteria of over 40 institutional investors on the platform, including the GSEs, the FHA, and the VA. We have built a multisided matching engine, something that simply cannot exist outside of what we have built inside Tinman.

To contrast, most fintechs operate on a single path and distribute the product through securitization. With Better, the result for the consumer is a significantly higher approval rate and generally lower interest rates, which because Tinman matches consumer and property-specific attributes across a broad cross-section of the investors on our platform, on a single loan-by-loan basis. Further, despite Better being balance sheet light and not taking any credit or prepayment risk, the default rate of our mortgages is one-third that of the industry average on over $100 billion of originated volume over the past nine years. So the proof is in the pudding. Our deep proprietary data moat has been instrumental in training our AI models and powering our platform.

Betsy, our generative AI home finance agent, built on top of Tinman, has learned from over 12 million recorded phone calls, 6 million approved customers, 600,000 funded loan documents, and almost 5 billion pages of property and consumer data information. All in one place, all in one end-to-end platform, with all of the things that were done by humans on those data all in one place. And recorded through the platform. We believe that this is something that does not exist anywhere else in mortgage lending or even broadly in consumer finance. Today, we are at feature parity between Betsy and the bottom 80% of human loan officers. Betsy communicates across voice chat, text, and email, with consumers nearly instantly to compute various scenarios and learns how to better understand consumers’ needs every day through every interaction.

What’s more is Betsy can handle millions of consumer conversations at the same time, enabling infinite scalability without adding additional headcount. As consumers learn to adopt and integrate their consumer finances and transact with an AgenTeq AI, Betsy is not just a voice agent or chatbot. Betsy can perform the functions of a human loan officer, processor, underwriter, and closer. Betsy is the user interface, helping consumers step by step through their homeownership journey, performing hundreds of thousands of consumer interactions per month, and remarkably good at detecting fraud throughout the entire platform. Additionally, Betsy has mastered finding ways to get an approval with the lowest possible interest rate across our network of investors with the lowest post-closing defect rate in manufacturing a mortgage, approximately 19 times lower than the industry average.

In fact, as of September, no human underwriter is allowed to decline a loan in our system without checking with Betsy first as to the alternatives that are available to restructure the loan so that the consumer can be approved and move forward in their homeownership journey. We believe this is a first across lending in the United States. Since we launched Betsy, our lead-to-lock conversion rate has increased by approximately 84% from 3.3% to 6.1%. This has been transformative to our platform in driving incremental volume and revenue through our platform, and it’s still very early days. As we scale Betsy at near zero marginal cost, we expect to further improve our unit economics through cost efficiencies on a per-loan basis. During the quarter, Betsy performed approximately 700,000 customer interactions and our AI underwriting approved over 61% of locked loans, with a clear path to 75% in the near future and 90% after that.

Our loan officer productivity in terms of funds per month increased to over three times the mortgage industry median. We have been heads down over the past few years honing our technology and optimizing the business for efficiency. With Tinman and Betsy, we removed the traditional constraints to growth in the mortgage industry, which is typically throttled by a lack of specialized licensed labor, whether it’s loan officers, processors, appraisers, or underwriters. We can now grow infinitely with AI and with a single unified tech stack at the core. There’s almost no better use case for AI to disrupt the market than the massive and antiquated mortgage market. The majority of the mortgage market still operates on what was built in the 1990s, where eight different separate systems were integrated through dated middleware, old-school FTP servers, and disparate databases.

What’s more, this dominant platform, which has over 80% market share, only allows one person to work in the loan file at any time, a file that costs the mortgage industry more than twice as much as Better to make. AI was designed to disrupt industries like this and yet fails in most cases due to the lack of a singular database architecture, causing huge latencies for any LLM to intermediate data and capture context quickly between disparate systems. Further, the lack of a unified interface prevents LLMs from being able to handle every single task required to fulfill a mortgage. Those limitations do not exist in Tinman. Tinman shines as a brand new modern tech stack with AI in action delivering real tangible measurable results in a multitrillion-dollar industry at a fraction of the cost.

I often think back to when we were building our AI platform. One of the point solutions CEOs said to the then CEO of Fannie Mae that he thought Better was trying to boil the ocean. And here we are. We have gotten the ocean hot. And it’s starting to drive tangible results in a way that is groundbreaking for the industry. With some macro green shoots in our favor and momentum in winning new partnerships, we believe we are in a position to scale rapidly, profitably, and with AI infinitely. When you look back at the last time rates declined, Better grew its volume by over 100 times over a five-year period and over 10 times over a two-year period in 2020 and 2021. We are positioned to do it again. This time, more efficiently and much more profitably.

And we believe we can achieve significant market share as this next cycle unfolds. Betsy and Tinman is our flywheel. That flywheel is turning. The opportunity is massive, and we are ready to monetize. I’ll now turn to our third quarter results. Starting with growth, we continue to propel opportunities independent of broader economic and mortgage market conditions. In 2025, on a year-over-year basis, we grew funded loan volume by 17% to approximately $1.2 billion, and revenue by 51% to approximately $44 million, driven by funding more loans, both through our D2C channel and our Tinman AI platform. By product, year-on-year funded loan volume growth during the quarter was driven by home equity volume increasing by 52% year-on-year, refinance loan volume increasing by 41%, and purchase loan volume increasing by 5%.

We have been rapidly growing our home equity business, taking share in a market that is coming back quickly as Americans are sitting on $35 trillion of home equity, the largest untapped asset class in the country. We’ve grown to an approximately billion-dollar-plus quarterly run rate origination in Q3 2025, compared to approximately $100 million in Q3 2023 just two years ago when we launched. Our model does not require us, unlike many others, to take any credit, prepayment, or liquidity risk. Because we can sell HELOCs onto the investor marketplace we have built. We do not rely on securitization. And we are able to mimic what we have done in the mortgage space in HELOCs, allowing investors to buy and bid on loans at a loan-by-loan level, which is unique in the industry.

There are incumbents in the home equity space who have started to create their own version of our investor marketplace. But today, that marketplace only comprises a very small portion of their volume and revenue. Whereas for us, the marketplace is 100% of volume, and 100% of revenue, in the HELOC space. During the quarter, we broadened our already high approval rates for HELOC products by launching AI-driven HELOC underwriting for small business and self-employed borrowers, making approvals possible using bank statements only. This product opens the door for 36 million self-employed and small business owners who have traditionally been underserved by traditional underwriting methods in the mortgage and HELOC space. Another example of how we are using AI to widen use cases and enable home finance for more American families to help them save more money.

Turning to cost efficiency. Total net revenue in Q3 grew 51% year over year, while expenses remained flat. Demonstrating our ability to scale revenue at lower marginal costs. We continue to adjust our cost structure to be leaner in overhead, building adequate resources to support the ramp of our new partnerships, which we expect to drive transformative growth in 2026 and beyond. With the goal of reaching adjusted EBITDA profitability by the end of Q3 2026. While our initial goal was to achieve further expense reductions this quarter, the team was focused on launching our three new transformational partnerships and engaging with additional partners in our pipeline. As a result, the intensity of our cost-cutting was somewhat muted compared to the vigor we’ve had in prior quarters.

Looking ahead, as we get these partnerships up and running, and to scale, we expect these anticipated cost savings to materialize in 2026. With Tinman AI technology, we automate time and labor-intensive components of the mortgage process, consistently reducing our cost to originate to approximately half of the industry average. I’ll now turn to quarterly business developments. Unit economics in our direct-to-consumer channel continue to improve with revenue per fund increasing to $8,300. The labor cost to fund continued to decrease to $2,500 and CAC per fund to $3,200, by the implementation of AI in every aspect of the sales and operations workflow. Resulting in a net contribution margin of $1,772 per fund compared to $1,064 per fund last quarter, an approximately 64% increase quarter on quarter.

We have not seen these types of contribution margins since, like, 2021. We expect to continue to lower the cost to originate as we increase conversion, lower CAC, and improve labor costs. And while our DTC business has always been at the forefront of pushing the envelope of what technology can do in the mortgage industry at its core, we are making great advancements in broadening the use of Tinman through our partnerships. We are very excited to have recently announced three new partnerships that we see as deeply validating the Tinman AI platform and believe will meaningfully expand our revenue and drive to profitability in the year ahead. First, we partnered with a top five US personal financial services platform, which currently serves over 50 million customers.

Under this agreement, our partner will offer home financing products to its end customers using the Tinman platform on a fully white-labeled solution, and we will earn revenue on a per-funded unit basis. Essentially, this is mortgage broker in a box. For financial institutions across the American landscape. We are focused on financial institutions that have large banks of customers. 10 million, 20 million, 50 million customers, and we believe that these financial institutions who have traditionally been limited, especially post the global financial crisis, in being in the mortgage business or offering mortgage business mortgages to their customer base, will dive right in with our mortgage broker in a box Tinman AI platform. We brought this fit partner from being just a fintech to a fintech plus mortgage broker.

There will be no upfront tax spend required by Better as our partner will programmatically feed customer data into Tinman. From there, Tinman will manifest offers delivered through our partner’s app, which has tens of millions of monthly active users all nearly instantly and updated daily. We expect transformative volume potential from this partnership as we scale into their vast customer base. Second, we entered into an agreement with a top five US nonbank mortgage loan originator. By migrating from the incumbent solutions that they’ve traditionally had for years, if not decades, onto Tinman, our partner’s loan will dramatically scale ability to surface eligible customers for HELOC and HELOANS within their customer base. They’ll also be able to mine their MSR book of over $300 billion to offer HELOCs and HELOANS to those customers on a programmatic basis in a way they’ve never been able to do with the incumbent HELOC solutions that are available to them today.

The initial focus will be on home equity products, and we believe there’s great potential over to help the partner unlock new ways to monetize its extensive customer base. In a way that has not been done before. It’s important to note that we are not just processing customers who raise their hand and ask for a home finance product. Rather, we are fully integrating Tinman into both of our partners’ customer data loads and CRM systems. This allows us to algorithmically mine customer data attributes and property data attributes for these customers, match them to products and investors on the Tinman platform, and use our AI to recommend the most applicable offer directly to the customer. We are also completely agnostic to the user interface, be it an iPhone app, or a human loan officer in a branch.

We serve all of them. Third, we partner with Finance of America, an industry-leading reverse mortgage lender with access to millions of senior customers who are typically home equity rich but cash flow disadvantaged. Together, we are launching the first HELOC and HELOAN product offerings to their customers powered by our Tinman AI. What’s more, leveraging Tinman, we have developed a senior second lien HELOAN product that specifically addresses the debt-to-income challenges that limit traditional HELOC products from being offered to seniors and that you typically see securitized by the incumbent players. Together, we believe these new partnerships demonstrate our evolution in powering the home finance ecosystem as a full suite platform and software well beyond our direct-to-consumer origins.

These partners are now live, and we look forward to sharing updates on our subsequent earnings calls as these partnerships ramp. In addition to our newest partnerships, we continue to make great progress growing our existing Tinman AI platform Neo powered by Better, local loan officer teams across the US experiencing rapid growth. The Tinman AI platform approach to local retail mortgage loan officer teams is similar to how Amazon opened its D2C model to a third-party seller marketplace. Similarly, Better is enabling retail mortgage lenders to build their business on the Tinman platform. And in doing so, we provide the compliance and licensing engine, loan origination system, and capital markets marketplace. We have near zero customer acquisition cost on this channel, and as partners fund loans on our platform, we earn a platform fee and a share of profits.

We’ve grown this channel from zero just nine months ago to now approximately 40% of our total revenue. The Tinman AI platform enables retail loan officer teams to originate more loans, serve more families, and lower their cost of funds, dramatically increasing their profitability and throughput versus traditional platforms that these loan officer teams have been on for decades. These officers are transitioning from dated expensive tech stacks where origination of a loan could cost over twice as much as Tinman to Tinman where the cost is just a fraction of that, at approximately $3,000. Savings go straight to their bottom line, allowing them to reinvest in their customers, offer lower rates, and close more deals within their local markets. Further, we’ve designed an optimization path to retain customers entering through the direct-to-consumer channel, who we might otherwise lose to an outside local loan officer.

By identifying customers who would benefit from more personalized local support, we connect them early on with a partner loan officer instead of losing them to competitors later on in our direct-to-consumer flow. This approach significantly boosts conversion rates amongst these customers and in turn strengthens our overall unit economics. During the third quarter, we funded approximately $483 million in funded loan volume for 1,148 families on the Tinman AI platform, an increase of 13% respectively compared with the prior quarter. And coming back to our multipronged distribution, we are also serving the customer by powering banks, credit unions, and other large mortgage originators that are seeking to license our Tinman AI software to either enter or reenter the mortgage business.

As our Tinman AI platform approach is like Amazon’s third-party marketplace model, you can think of our Tinman AI software channel as Amazon’s AWS software model. A lot of banks and credit unions are taking a refreshed look at the mortgage space as a regulatory environment is becoming increasingly favorable. However, bank origination of mortgages has largely been unprofitable given their high cost to originate. This is where our Tinman AI software comes in. Our Tinman AI software essentially provides mortgage in a box, enabling banks to not only use our software but also gain access to underwriting resources and sales resources if they so desire. And while the broader software industry charges clients on a per-seat basis, we have a disruptive pricing model of charging on a per-funded loan basis, or outcome as a service, which is very similar to what a lot of the leading AI companies in Silicon Valley are doing.

Over time, we expect this channel to be the most profitable of our three channels with SaaS-plus level margins since most of the costs associated with this initiative have already been spent on developing Tinman internally for our direct-to-consumer business. Our existing bank partner on the Tinman AI software platform is ramping as we power its mortgage origination business from click to close across multiple products and across multiple channels. And we expect revenue from this partnership starting in Q4 2025, with SaaS-level margins. Our overall partnership pipeline is robust. We are focused on aligning with companies that are leaders in their respective verticals, those with large customer bases and where the Tinman AI platform clearly outperforms legacy systems.

Our strategy is simple yet powerful: capture a leading player in each vertical, empower them to scale their mortgage business and home equity business with Tinman, and then expand outward across the ecosystem as others follow suit. Land and expand. Verticals that we are interested in include fintechs, BNPL providers, traditional mortgage lenders, and servicers. Each of these verticals represents hundreds of billions of dollars in annual mortgage originations. So by first securing a partner who is a leader in their vertical, we establish credibility, create momentum, and open the door to broader adoption across that vertical. Looking ahead, the opportunity has never been more exciting. We continue to make great progress towards our goals of driving increased volume and revenue, balanced with ongoing expense management and improved efficiency.

We remain focused on enhancing our go-to-market strategy with growth being our North Star, alongside continued expense management and channel diversification. We all with the goal of reaching breakeven on an adjusted EBITDA basis by 2026. Our path to adjusted EBITDA profitability will be multifaceted, driven by volume growth in both our direct-to-consumer and Tinman AI platform channels, unit economics or per-loan contribution margin continuing to improve as we further lean into AI efficiencies, the scaling of higher-margin partnership channels, including Tinman AI platform and Tinman AI Software, pricing improvements, and continued corporate cost reductions. While our unit economics are already profitable at the contribution margin level, increasing volume will allow us to offset additional corporate expenses.

Note that these growth opportunities come with varying levels of expansion and profitability profiles and will change based on the broader macroeconomic trajectory. As a result, our path to adjusted EBITDA breakeven is unlikely to be linear on a quarterly basis. We do not anticipate the same level of burn reduction each and every quarter. During the third quarter, we had an adjusted EBITDA loss of approximately $25 million, down from $27 million last quarter and $39 million one year ago. In particular, for the three large partnerships we signed, we had a significant amount of resources in sales, operations, and technology dedicated to launching those partners that were not revenue generative but will create significant growth in the years ahead.

As these partnerships launch and start to generate revenue and contribution profit, we expect burn to come down more dramatically in the coming quarters ahead. 2026. Now to touch briefly on our balance sheet and capital positioning. We ended 2025 with $226 million of cash, restricted cash, short-term investments, and assets held for sale. In addition, we continue to maintain strong relationships with our three financing counterparties, which provided a total capacity of $575 million as of September 30, 2025. We expect that our recently announced partnerships will require us to increase those warehouse lines meaningfully to accommodate the expected funding demand. On capital positioning, we rightsized the capital structure earlier this year, retiring approximately $530 million of convertible notes for a $110 million cash payment and a $140 million note, generating $211 million of positive equity.

As announced in our 8-Ks, our CFO, Kevin Ryan, will be concluding his time with us. We are so grateful for everything Kevin has done for this company, taking us public, rightsizing our capital structure, and building out our finance and accounting function. We wish him the very best in his new endeavor and are excited about the strong candidates in consideration for the CFO role. We hope to share the outcome of our new CFO search with you soon. In the UK, we were pleased that Birmingham Bank grew its loan book by 44% in the third quarter sequentially. Versus 2025 as we have implemented our technology stack into the bank and, in doing so, enable the bank to become the fastest-growing specialist mortgage lender in the UK. With respect to our non-core UK assets, we continue to exit those positions and expect these divestitures to continue to benefit our adjusted EBITDA through the remainder of 2025.

Turning to our outlook, the Tinman AI platform loan volume continues to grow rapidly, and we expect over $600 million of AI platform originations in Q4. This would be growth of over 24% versus Q3. For the full year 2025, we expect total funded loan volume to increase year over year driven by tailwinds from growth initiatives, including Tinman AI platform, offset by continued macro pressure and the loss of our Ally business, a roughly $1 billion headwind. We expect further improvements to adjusted EBITDA losses for the full year 2025 versus the full year 2024 through a combination of AI-driven improvements in conversion rates, efficiency gains, and continued corporate cost reductions. In the medium term, while we expect D2C to continue to grow nicely, we expect it to become a smaller part of the total revenue mix as our partnership channels scale faster.

We spent the past three years building for this moment. Our platform is proven. The housing cycle is turning. Our AI is scaling, and our partnerships are just beginning to ramp. About a year ago, we met the NEO Home Loans team, and I saw firsthand the experience they were able to deliver in a branch or over a Zoom call. And I thought to myself, how can we make this accessible to everyone? How can we take this four seasons experience that the Neo team delivers and deliver it at a four points price to the average consumer? That got us thinking about the experience of being in a branch or on a Zoom call, with someone with ten, twenty, thirty years of experience with a deep knowledge base, someone who’s done billions of dollars of loans and one who can walk you through every aspect of the process that is life’s biggest financial transaction.

One, American families are still fundamentally insecure and unknowledgeable about. We’ll be launching the AI mortgage adviser in a prerelease. And you can sign up to join us at better.com/tinman. Please take a look and let us know what you think.

Tarek Afifi: Hello.

Ryan Grant: My name is Ryan Grant. How are you doing today?

Vishal Garg: Woah. Am I talking to an AI or is it you, Ryan? What happened here?

Ryan Grant: Hey. It’s actually me. Ryan’s AI mortgage adviser. Kind of like having Ryan’s brain and Better Mortgage’s tech all rolled into one. Ready to help you anytime. What can I do for you today? Woah. Woah. Woah.

Vishal Garg: Like, what did you do with Ryan’s brain? Like,

Ryan Grant: Think of it like this. I get it. It sounds wild. Basically, I’m Ryan’s expertise combined with AI tech. So I can help you anytime with mortgage questions, calculations, and advice just like he would. But instantly and around the clock. No worries. Ryan’s still around. But I’m here to make things easier for you. What would you like to know or talk about?

Vishal Garg: Okay. I mean, that’s interesting. Like, you tell me a little bit about yourself? Like, do you have a lot of experience doing this, or do they just get, like, some guy off the street?

Ryan Grant: I’m built on Ryan’s experience. He’s personally funded nearly $2 billion in mortgages and helped thousands of homeowners. So I bring all that know-how plus Better Mortgage’s smart tech to guide you through your mortgage journey with confidence and clarity. Think of me as Ryan’s brain. But always ready to help you anytime you need. What’s your biggest question or goal with your mortgage right now?

Vishal Garg: I mean, I think rates are coming down, and I got this mortgage two years ago. And I think I can save some money. Right?

Ryan Grant: That makes sense. Refinancing to save money is a smart move if rates have dropped since you got your mortgage. To see how much you could save, I’d need to know your current interest rate, your remaining loan balance, and how many years you have left on your mortgage. Can you share those details?

Vishal Garg: Yeah. No. So I think I have $500,000 left on my mortgage. I got it two years ago, literally, like in September ’23. And I think like, you know, I don’t remember exactly what my rate was, but it was like, in the 7% something range. Maybe you can help me, like, figure that out. Like, yeah.

Ryan Grant: Got it. You have about $500,000 left and your loan started around September 2023 with a rate around 7%. I can help estimate your savings based on that. Do you happen to know your current monthly principal and interest payment? If not, no worries. I can estimate that too.

Vishal Garg: Yeah. I know. It’s like on autopay, so I don’t know exactly what it is. So, yeah, if you have an estimate, that would be really great.

Ryan Grant: Your current loan is about 7% with a $500,000 balance and twenty-eight years left. And your monthly payment is roughly $3,398. Oh, that’s close. Got it. Let’s start with the option that gives you the lowest monthly payment by paying points upfront. This option lowers your interest rate, so your monthly payment goes down. But you’ll pay some cash at closing. Sound good?

Vishal Garg: Yeah. Yeah. Let’s find out. Sure.

Ryan Grant: The lowest monthly payment option with points. Your new rate would be about 5.13%. Your monthly payment would drop to around $2,722, saving you about $676 each month when it comes with points costing about $12,907 upfront. Would you like to hear about the par rate option with lower closing costs too?

Vishal Garg: Yeah. Yeah. Why not? Like, you know, I’m not sure if I’ve got just $12 lying around.

Ryan Grant: The par rate option means no points upfront. So your closing costs are lower. Your rate would be about 5.99% with the monthly payment around $3,053, still a bit lower than your current payment, saving about $345 each month. It’s a bit higher monthly, but you keep more cash at closing. Sound like something you want to explore more?

Vishal Garg: Yeah. No. Like, isn’t there a way, like, I heard, like, you can just roll the closing cost into, like, the mortgage, like, at closing? Is that possible here?

Ryan Grant: Yes. You can often roll closing costs into your new loan balance, which means you don’t pay them out of pocket at closing. That usually comes with a slightly higher interest rate, like the par rate option I just mentioned. So your monthly payment might be a bit higher, but you keep your cash upfront. Does that sound like a good fit for you?

Vishal Garg: Yeah. No. I think I’m gonna go with a lower monthly payment. Like, $700 a month of savings is like a lot more than $350 a month of savings. So I think that will be really good for my cash flow. So what do I need to do next? Like, does this work?

Ryan Grant: Great choice. Saving about $700 a month can really help your cash flow. The next step is to start your refinance application so we can lock in that lower rate and get the process going. Could you please spell your first and last name for me?

Vishal Garg: Yeah. Yeah. Wait. Wait. Wait. Can I talk to the real Ryan too? Like, or is it just gonna be me and you all the time?

Ryan Grant: I totally get wanting to chat with the real Ryan. He’s a great guy. I’m here to help you anytime with quick answers and guidance. If you ever want to connect with Ryan directly, I can help make that happen once we get your application started and things are moving along. Sound good?

Vishal Garg: Sounds great. Alright. Listen. You know, I’ll get the application going, and then why don’t we, like, set out some time to talk to the real Ryan? It’ll be interesting to see whether he knows as much as you do.

Ryan Grant: That sounds like a solid plan. To get started, could you please spell your first and last name for me? That way, I can set up your application, and we’ll be on our way.

Vishal Garg: Okay. It’s Vishal Garg. Why don’t I do the rest of this online, and then we’ll, you know, connect later? Alright? Talk soon. Bye.

Q&A Session

Operator: Thanks, Vishal. We will now begin the question and answer session. Please limit yourself to one question and one follow-up. If you would like to ask a question, please raise your hand now. If you have dialed into today’s call, please press 9 to raise your hand and 6 to unmute. Your first question comes from the line of Owen Rickert with Northland. Your line is open. Please go ahead.

Owen Rickert: Hey, Vishal. Thank you for taking my questions here. I guess quickly, can you just dive a bit deeper into the three recent partnership announcements and how you expect each of these to ramp as we head into 2026?

Vishal Garg: Sure. So with respect to the large financial services platform, we expect that to ramp over time, over the next six months, specifically as we increase the penetration of their users in their app that see the offers from us and the number of users every day that they drop into Tinman to surface offers for. And those offers will be sent via stories in their app, notifications, text messages, things like that. And so we’re just gonna increase those. And we’ve created a specific pod for this partner because it’s such a large partner. And we need to staff into that pod. And so, while we expect the overall size of the partnership to manifest itself into, you know, multiple billions of dollars a month, it’s going to take a bit of time for us to also ramp up and see what labor is gonna be required, what percentage of that partner’s customers are comfortable talking to an AI, what percentage of that partner’s customers need to talk to a person.

So we’re working all of that out. With respect to the other partnerships, the large mortgage originator, we’re gonna start first with their direct-to-consumer team, then we’re gonna start rolling it out to the team that does MSR and MSR recapture. And then from there, we’re gonna start rolling it out to their loan officer teams all around the country to market HELOCs and HELOANS too. And so there’s gonna be a ramp in that regard as well over the next six months or so. And then, with Finance of America, we are launching the HELOCs and HELOANS first to their customer base, then to their partner originators, and then across to their wholesale channel. And so, I think that is also gonna take, you know, another three to six months to fully ramp up.

As well as the second, you know, the reverse second lien HELOC product, which we are rolling out in beta right now, which we’re going to then ramp up across their entire network.

Owen Rickert: Got it. Thank you. And then secondly, you did hit on this pretty early on in the prepared remarks, but how would you characterize the future partnership pipeline right now? And what does that look like today? And maybe how has this pipeline evolved over the last few months?

Vishal Garg: I think as our partners are able to see how fast we’re able to implement some of the earlier partners that we have now launched, the quality of the user experience, the ability to, you know, get approved for a mortgage programmatically, the ability to take something that traditionally has been very passive and sold passively by these partners and then have that be done in an active algorithmic way, the partner pipeline has really quite frankly exploded. And so we are seeing a lot of demand. The other thing just from a macro perspective, the largest incumbent solution has been forcing, has been going through an SDK change and has been forcing reintegrations with all of its partners, for its clients. And so it’s been an interesting moment where a lot of people are very, very frustrated with the incumbent solutions that are out there and are looking for something new.

And so I think, you know, it’s sort of like, luck is when preparedness meets opportunity. And I think, you know, we’re pretty thankful to be in the position that we’re in now.

Owen Rickert: Great. Thanks, Vishal.

Operator: Your next question comes from the line of Brendan McCarthy with Sidoti. Your line is open. Please go ahead.

Brendan McCarthy: Great. Good morning, Vishal. Good morning, everybody. I really appreciate the demo there with Ryan. I thought that was great. I just wanted to start off circling back to the new partnerships, particularly the one with the top five US personal financial services platform. Can you give us detail on what the ultimate volume opportunity is like there? You know, 50 million customers is obviously a huge number. Just curious as to what you think the addressable market is in terms of volume.

Vishal Garg: Yeah. So, I mean, if you go and try GPT and type in what is the mortgage penetration rate for a financial institution in the United States with 50 million customers, it will tell you that, you know, it ranges from 10 basis points of that customer base to 15 basis points of that customer base. So let’s use, like, you know, a low average 12 basis points. You multiply 12 basis points by 50 million, that gives you 60,000 originations a year. 60,000 originations times an average balance of, like, $400,000 gives you about $24 billion. So I don’t know exactly what the number is. I’m not committing to that number, but that’s sort of, you know, if we were able to just do it in an average passive manner at some branch, what we think we can achieve could be multiples of that if we’re able to sort of algorithmically mine and surface offers directly to consumers in their mobile app.

Brendan McCarthy: Understood. That’s very helpful. Thank you for that. And next question, just looking at the guidance, you know, really implying strong growth there, I think, from, you know, the 500 million monthly loan volume run rate to about a billion. It’s just really a step up there. What’s really underpinning that outlook? Is it, you know, just strictly the partnerships? Is it a growth in D2C? Is there any interest rate assumptions there? Just curious as to what’s underpinning that.

Vishal Garg: No. We’re assuming interest rates stay the same. And, yes, I mean, as you can see in D2C, we have been focused on making more money per loan in D2C rather than growing volume, though volume has grown pretty substantially, especially if you take out, you know, on a quarter-on-year basis, if you take out the Ally volume that we had last year, you know, organic growth has been over 50%. And, you know, when you layer that on, if there’s a rate cut, I think, you know, D2C is going to fly. But other than that, like, we’re just assuming that the rates stay the same. And so the numbers I’ve given you and I’ve indicated assume the interest rate environment doesn’t change.

Brendan McCarthy: Great. Great. Thanks. That’s all for me.

Operator: Your next question comes from the line of Kartik Mehta with Northcoast. Your line is open. Please go ahead.

Kartik Mehta: The press release indicated that you anticipate about a billion dollars of loan volume in the next at the end of six months because of these partnerships. Does that assume that each of these partners will be fully ramped or get fully integrated? Or are you anticipating the ramp to take longer? So, really, the billion dollars could be a lot more once the partnerships are fully integrated.

Vishal Garg: I think it could be a lot more once they’ve fully integrated.

Kartik Mehta: And then just, you know, the per funded contribution margins increased significantly. The one volatility is in the CAC. So I’m just curious, what’s your anticipation for CAC as we move through 2025 and then 2026? I’m assuming they’ll start trending lower as the partnerships become a bigger part of the loan volume. But I wanted to get your perspective on that.

Vishal Garg: Yeah. I mean, with the partners, there’s no CAC. Right? There’s, you know, no upfront CAC. The D2C CAC remains quite high. It, you know, purchased, you know, which remains a challenge in this market environment. You’re spending money this quarter to book loans in six months, twelve months, eighteen months, when the consumer actually buys the house and books a loan. I think one thing that may be underappreciated about Better is over the past three years, we’ve given out over a million preapprovals to consumers, and those consumers have not been able to find a house or it’s been too expensive for them to find a house. And so that CAC, you know, that you see there is elevated because for all the consumers that are not able to find a house or that they want to buy, you know, basically, we eat that CAC in that specific quarter, then when that consumer finds the house they want to buy, then when they come through, then, you know, it shows as lower CAC.

So the mortgage industry CAC acquisition cost problems are even further compounded by the long gestation cycle of, you know, consumers on the Internet and when they get preapproved and when they actually find a house. So, you know, we do expect as rates, if rates come down, that, you know, the CAC will come down materially across the board for purchase or for refi. I mean, just to give you some context, when in the last rate cycle, when rates were coming down, our CAC on a refi was $1,000 a loan. And so there’s a lot of positive convexity in the CAC as consumers, you know, as the rate environment changes and consumers’ propensity to get preapproved and then actually fund increases.

Kartik Mehta: Perfect. Thank you very much. I really appreciate it.

Operator: Your next question comes from the line of Bose George with KBW. Your line is open. Please go ahead.

Frankie: Hey, good morning guys. This is actually Frankie on for Bose. Nice to see you. Start with, can you just walk through the ways in which AI efficiencies increase revenue per funded loan? In slide 16, you noted that this will be driven through enhanced sales and operational performance.

Vishal Garg: Yeah. So I think what you’ll see is our revenue per loan is continuing to grow up. Right? And I think the reason for that is Betsy’s able to supplant the loan officer whenever the loan officer is not able to, you know, either pick up the phone, answer a question, turn around a new preapproval based on, you know, data that the consumer has provided. You know, Sunday afternoon, 04:00, they want to put in an offer that they saw. You know, Betsy’s there for them in a way that, you know, traditionally, your human loan officer isn’t able to be. So that’s enabled us to, one, you know, make our competitive pricing, and, you know, slightly less competitive and, you know, increase the gain on sale. Number two, as our volumes are going up, and, you know, it allows us to not have to staff up with many people.

I think as you can see, like, on a year-on-year basis, volume and revenue went up substantially over 50%. And expenses actually, you know, stayed the same. And, therefore, the burn came down substantially by, like, about 35, 40%. And so that’s sort of how Betsy’s allowing us. It’s allowing us to be more responsive, which means lower discounts, superior service, really build a service offering for consumers. And then on the flip side, you know, not have to hire as many people as we scale volume. And automate the processes, like processing loans, underwriting loans, closing loans, that traditionally have been done by people.

Frankie: Great. Thank you. That’s very helpful. And then can you just help us understand what types of incumbent solutions you’re replacing in your partnership? Is it both the LOS system and POS system?

Vishal Garg: Yes. So we have integrated with a number of POS systems that are out there. Where, let’s say, if our client wants to keep the POS that they’re using today, that’s fine with us. You know? We’ll take all of the other stuff. We generally do replace the incumbent LOS. And in many cases, we replace the POS, the LOS, the pricing engine, the CRM system, the document generation engine, the notary, you know, and closing engine, and the warehouse, you know, the warehouse software. So when the client signs up with us, we might replace as many as eight to 10 different systems that the client has.

Frankie: Awesome. Thank you. That’s all for me.

Operator: Your next question comes from the line of Mikhail Goberman with Citizens JMP. Your line is open. Please go ahead.

Vishal Garg: Hi, Mikhail. You’re on mute. I think you have a question.

Mikhail Goberman: I was. Sorry about that. Thank you. No problem. Good morning. Thanks for taking the question. If I could ask about expenses and I appreciate the comments prepared remarks about the expenses and how you’re planning to deal with the partnerships with regard to that going forward. I believe you mentioned a target for the first quarter of next year. Is there any sort of a number or run rate that we can put on that?

Vishal Garg: No. I think we, you know, we’re hoping that, you know, within the next six months, we get to a billion dollars a month origination run rate. I think, you know, we’re hoping that we continue to have scale in our expenses. We’re hoping that we continue to drive a lot of corporate cost reductions forward. We’ve been really busy this last quarter, so I think I personally wasn’t able to pay as much attention to some of the, you know, legacy contracts and things like that that we need to kind of continue to still beat out, you know, three or five-year contracts that we signed back in 2020, 2021 that we’re, like, working to sort of, you know, reset, with more AI-driven type solutions. I think there’s still a lot of cost savings left. Which is why we continue to drive to, you know, achieving profitability. While, you know, growing scale at the same time, you know, by Q3 2026.

Mikhail Goberman: Great. Appreciate that. And if I can fit in one more, just your general thoughts on the stability and strength of the mortgage industry in general given where we are with interest rates and sort of wobbles, I guess, you could say, with the economy a little bit. Just your general thoughts on the borrower and the consumer and how the whole system is developing going forward. Thank you.

Vishal Garg: Yeah. No. I think, look. I really believe that we’re headed into a recession. I believe that, you know, that’s gonna result in a couple of things from a macro standpoint. I think there’s, you know, you would think that heading into a recession, purchase mortgage would be disadvantaged. But there’s millions of people who have wanted to buy a home over the past four, five years who missed out on the 2019 to 2021 rate environment. And they have been building up their savings, and they’re looking and, you know, many of them who have owned equities in the past couple of years, they have been building up wealth to go and buy a home. So I think that you’re gonna see purchase mortgage origination stay at sort of where they are.

You might not have, like, the boom that you did in 2020 and 2021. If we have a real recession. And then on the flip side, you know, there’s like 20 million people that can start to save money as rates go below 6%. If we do actually enter into a recession. And I think that that’s pretty significant. And then lastly, in the current period, let’s assume we just stay in this sort of muddled medium inflation, 6% plus interest rate environment, you know, home equity origination is still such a small number compared to what they were pre-global financial crisis, you know, or where they are, you know, relative to the total size of home equity that people have in their homes, which is now, I think, $22 trillion of capital home equity, according to the latest TransUnion report.

And I think for us, you know, we have both the secular tailwinds of a very competitive business model in D2C that we are now, you know, continuing to improve the conversion rate on. I think as you might have seen, like, in the earnings release, like, we talk about the conversion rate going from 3.3% over 6%, like an 81% increase. Right? That’s just, like, grinding out, like, putting the AI in places where the humans are not able to do as good of a job. Right, to satisfy the consumer, just keep on grinding away at that. And so that, I think, is super meaningful. And will continue to drive both unit economics and growth in the D2C channel. And then when we’re taking on partners that we’re taking away from incumbent platforms, they are, quite bluntly, we’re stealing market share.

And so, you know, and then that’s the fastest-growing part of our business. And so, you know, the mortgage market stays the same, if it’s whether it’s a trillion and a half in originations, 2 and a half trillion in originations, of course, we’d love it to be 2 and a half trillion in originations. But, you know, where we’re moving partners from incumbent solutions that are, you know, built in the eighties, nineties, February, onto our tech stack, they’re, you know, we’re relatively agnostic to the cycle. And if the cycle comes our way, then that’s even better.

Mikhail Goberman: That’s great color. Thank you very much.

Operator: Your final question comes from the line of Doug Harter with UBS.

Doug Harter: Hi, Doug. Thanks. Vishal, I was hoping you could talk about, as you’re guided to getting back to breakeven and to profitability, what type of volumes do you need to accomplish that?

Vishal Garg: I think depending on the mix, you know, I think we get to a billion plus, and, you know, we have a good shot at it. Obviously, the margins in our partnership business are higher than that in our D2C business. But, you know, even D2C is getting to a place where the margins are pretty healthy on a contribution margin basis. But, yeah, you know, I think we get to a billion plus. And then, you know, depending on the mix, we get to beyond that. I think, per month, I think you have a very, very, very good business that’s driving towards breakeven.

Doug Harter: And then can you talk about is there different revenue that you’re generating with partners for home equity origination versus a traditional first lien mortgage?

Vishal Garg: I think home equity originations, I mean, the, you know, the loan amounts are much smaller. But the gain on sale is higher. And, you know, between the gain on sale and the fees, you know, you’re making, you know, on the mortgage side, you’re making maybe $8,000 a loan. In the home equity side, you’re making, like, $6,500 a loan. I think it’s very important to remember in both of these cases, we’re not retaining mortgage, the MSR, we’re not taking credit risk, we’re not taking prepayment risk. We’re not taking any of those risks. In home equity, we have yet to scratch the surface on what scale looks like. There are other people in the home equity market selling their loans at 107 or booking a gain on sale at 107.

We’re at 103 and a half. So there’s a long way to go in bridging that gap. But, you know, when those people are booking those loans at 107, they’re taking principal prepayment risk. They’re taking credit risk. They’re, you know, booking residues, all that sort of stuff. If you like comparing apples on apples basis on a pure marketplace basis, I think we’re getting a pretty good deal, but I think we probably still have another point or two that we can squeeze out on our home equity originations.

Doug Harter: Great. Appreciate it, Vishal. Thank you.

Operator: There are no further questions at this time. This concludes today’s call. Thank you for attending and you may now disconnect.

Vishal Garg: Thank you, everyone.