Blend Labs, Inc. (BLND) Earnings Call Transcript
November 5, 2026
Earnings Call Speaker Segments
Good afternoon, and thank you for joining Blend's conference call covering financial results for Q3 of 2026. I'm Meg Nunnally, and I serve as Blend's Head of Investor Relations. With me on the line are Nima Ghamsari, our Co-Founder and Head of Blend, along with Jason Ream, who leads Finance and Administration. Ahead of getting started on today's call, please be aware that we'll cite some non-GAAP metrics, and reconciliations of those to GAAP metrics appear in today's earnings release as well as the appendix of our supplemental slides. Non-GAAP metrics should not be viewed as a replacement for GAAP results. Except where noted otherwise, every financial metric we cover today, profitability included, is presented on a non-GAAP basis. In addition, some remarks made on today's conference call concerning Blend and how it operates — specifically our guidance for Q4 of 2026, additional commentary about 2026 and 2027, and our expectations regarding markets, strategic investments, product development plans, operational targets, and the macroeconomic environment — could be treated as forward-looking statements under federal securities laws. Be advised that forward-looking statements carry substantial risks and uncertainties, and numerous factors, many outside the company's control, might cause actual results, events, or circumstances to differ materially from what those statements describe. Please review the risk factors identified in our most recent 10-Qs, our 10-K for fiscal year 2025, and our remaining SEC filings. We make no commitment to update these statements should conditions change, except where the law requires it. Financial information shared on this call reflects continuing operations, and earlier periods have been recast so that operations now classified as discontinued are excluded. Finally, a copy of our prepared remarks will be posted to our website by the time today's call concludes, and an audio replay will be accessible shortly afterward. I'll hand the call over to Nima now.
Thanks, Meg, and hello, everyone. Let me give you the straightforward take on the quarter. Conditions in the market became tougher, yet our pace of execution has never been quicker. Revenue landed in the upper half of the range we guided to, and non-GAAP operating income finished above the top end. Three points for today. First, we just completed our strongest sales quarter in over 3 years; second, Autopilot grew from 6 customers to 10, and the pipeline behind it is now large enough to measure and still expanding; and third, the market may offer no tailwind next year. Growing means converting that Autopilot pipeline. Starting with sales: during Q3 we signed 26 new deals and expansions; measured in dollars, that is over 2x what the first 2 quarters produced together. In August, I shared that we had a late-stage pipeline expected to convert into signed deals during the coming quarter, and that is precisely what occurred. Several deserve mention; we brought on a new logo with a top 10 bank for mortgage and Rapid Home Equity. We also cross-sold Rapid Home Equity into a different top 10 bank and into a top 20 mortgage originator, and 5 of the 26 deals were Autopilot deals. Two forces are behind this. Rapid Mortgage and Rapid Home Equity are what lenders are buying today because those products make it simpler for their customers to do business with them and reduce their cost to originate in a market where every basis point matters. And Autopilot has shifted how the biggest lenders view us. To them we are no longer merely a point of sale; we are a system that processes and moves the loan in real time, which leads to my second point, Autopilot and the pipeline behind it. That Autopilot pipeline right now includes 3 of the largest 20 banks, 2 of the largest credit unions in the country and one of the largest 20 largest retail mortgage lenders. The deals we are working to close in the fourth quarter alone could add $5 million in incremental revenue to 2027, and the pipeline stretches well into 2027. For existing customers, Autopilot operates within the customer's current Blend environment. No integration project is required. Once a customer signs, we switch it on and the revenue and the value begin almost immediately. Contrast that with perhaps our Rapid deals or new logos, which can require multiple quarters of integration before appearing in our results; this is a very different motion for us. So, if we execute against this Autopilot pipeline, we believe a very real upside scenario exists for 2027 despite the macro. And that pipeline does not exist in a vacuum. It exists because of the adoption and the customer outcomes underneath it. Let's discuss adoption, the simplest measure of whether this is working; in August, I told you Autopilot had worked on more than 40,000 loan applications since it launched. Today, that number is more than 90,000. That figure has doubled since the previous call. We now have 10 customers under contract for Autopilot, compared with 6 back in early August. The roster spans everything from a top mortgage servicer nationally to community banks and credit unions. And although the bigger banks are collaborating with us to finalize the agreement, they still need to clear AI governance. So we anticipate a few of those closing during the current quarter. The reason customers sign with us is that the results are visible; the numbers are the same preliminary ones I cited in August — a 10% to 15% gain in pull-through rates, 2 to 4 days of cycle time improvement on loans, and roughly 4.5 hours of fulfillment time automated per loan. For a loan officer, processor, or underwriter, that means the difference between chasing down documents and having a personal conversation with your customer, the borrower. Yet customers are equally enthusiastic about Autopilot because of what we're developing around it. The largest new addition there is Navigator. With Navigator, Autopilot sits in front of the loan officer as an assistant — one that's built into Blend. So a loan officer can ask which loans in my pipeline need attention today? How do I get a pre-approval letter for Nima? And Navigator handles all the work they previously had to click through multiple systems to do, spending tens of minutes or hours, much like tools you might use personally such as Muse or Claude Duet. But it lives inside the lenders' own products, with their permissions, the guidelines they've already configured, the integrations they've already set up — all the consequential work they can do today is now doable through Navigator; and anything important requiring the loan officer's attention waits for their approval. That keeps them in control, and the architecture remains permissioned; we launched on desktop first in October at the MBA Annual Convention, and we unveiled Navigator Mobile — the same assistant, but in your pocket for people on the go. Plenty of loan officers are in the car between appointments, working with Realtors, and we launched that earlier this week for our customers. I'd like to play a short video of it for you here now. < > The response has been fantastic. At MBA, a day ahead of the event, we arranged a preview lunch on short notice and 17 lenders turned up at that lunch in a small room while I demonstrated this together with the Autopilot team at Blend. So it was truly an amazing experience watching them arrive and get enthusiastic about something new that could help make their lives and their teams' lives easier. With Autopilot handling the labor involved in the loan, and Navigator returning time to the loan officer so they can engage with their customers, that's the second rationale for lenders to take up Autopilot. And when a loan officer conducts their day through Navigator, they're still conducting it through Blend, but in a way that's 10x better — quicker, simpler, and easier to consume — which fuels utilization and fuels retention. It's also a solid lever for us to keep expanding alongside our customers and generate value, and in the end it's that value that will fuel retention for us and for our customers. In August, I also outlined 4 advantages that we think give Blend the right to win in agentic lending: where we sit with the data — 15 years of data — where we sit at the front end of the process, when data and documents are arriving, the harness that Autopilot operates within, meaning all of the integrations and all of the compliance embedded in our platform, and on top of that, tremendous relationships with lenders who genuinely want a system that can drive outcomes for them. And we've delivered that for them historically. None of that has shifted, so I won't go through all of it again. But I do want to stress one point again, because we have fresh evidence for it this quarter: Autopilot becomes stickier and more future-proof as customers build on top of it.
Three months ago, in August, I stated our plan to expose our infrastructure through Autopilot MCP, enabling customers and partners to build atop our platform rather than working around it. Let me share what has happened since. Two large lenders have developed voice agents powered by Autopilot MCP. One large fintech lender is building inside its own mobile app and website, incorporating capabilities from Autopilot MCP together with Blend integrations and intelligence beneath. In the past, that presented them with a major trade-off: whether to use Blend UI or their own UI. That trade-off no longer exists. Additionally, a community lender is working with a lead generation partner to move a pre-application through a chatbot and open a file straight into Blend. Four lenders, three quite distinct use cases—and through Blend, every one of them will operate more quickly and effectively thanks to greater connectivity. That's the core idea: it transforms Blend and Autopilot from a mere interface into an actual engine. The underlying models will keep improving—becoming faster and less expensive—and that applies across the board. Yet what resists commoditization is the layer that understands how to handle loan context, maintains the integrations, enforces the guidelines, and preserves human oversight over the regulated choices that count. Building a harness of that kind generates lasting value. Whenever a customer adds something on top of it, that harness grows more valuable to them. So customers end up holding onto it for a longer stretch and getting more use out of it. Compared with where we stood just a few years back, it's practically a different business. On the pace of Autopilot — a big part of why we're able to go this quickly is that, starting in March, we've rolled out Autopilot, Autopilot MCP, Navigator, and Navigator Mobile. There's additional stuff in the pipeline for later this year as we work toward Blend 3.0, which fits into our push to become an agent-first company. That concept became generally available for desktop in September, then a mobile app we demonstrated in October, and it's now generally available in the App Store for our customers — all within weeks and months, thanks to an incredible platform and harness that already carries these capabilities. The work centers on constructing the agent layer around it, and doing so with agents too, so our team can accomplish these things faster than maybe ever before. Speed, in my view, is an advantage that compounds. Eventually, we anticipate that this will be reflected in operating leverage, and you'll be hearing additional details on that when we outline and communicate our 2027 plans. Let me now wrap up with a few closing remarks regarding the broader growth trajectory. Back when we had our conversation in August, mortgage rates stood at roughly 6.8%. Following the Fed's September rate hike, they climbed to a peak exceeding 7.5%. As one would anticipate, applications declined in tandem—for both purchase and refinance. Refinance applications in particular now sit at approximately 60% below their levels from earlier in the year. Since over 75% of outstanding mortgages hold rates under 6%, refinancing demand at current levels remains fairly constrained. This is arguably the hardest origination environment we've faced, certainly in 2023 and possibly ever. I won't pretend this doesn't matter—it does, which is why we had to update our Q4 figures. But the macro is the one factor in this business we've never had control over. And over the last 4 years, we've worked to ensure our company isn't dependent on it. We're developing new products we can monetize. We're expanding our customer base in ways that let us do more and deliver more value to them. Looking at this quarter's 26 new deals, it's clear we're moving in a favorable direction no matter what the macro environment does. In May I said Autopilot and our internal agents would produce significant growth next year, and here's the status of that. The Autopilot contracts are real. The biggest lenders are taking action. And as I noted earlier, large lenders put AI through model risk, compliance, security, and procurement reviews, and those steps take months—but these are among the largest financial institutions in the country. Even though those clients are actively transitioning and often pick us as their preferred provider, additional effort remains before everything is finalized and they can fully activate it for their own operations and their end users. Those sizable agreements therefore continue to advance, with the associated revenue arriving in 2027 when they finalize — nothing sooner. I want to be candid about that, though collaborating with major financial institutions genuinely requires patience. Here's my current view of how 2027 will unfold. Jason will share our volume outlook shortly, and the base case is a market shrinking again next year, with overall market size and a business that must earn its growth against that backdrop. Turning the autopilot pipeline I sized for you earlier into reality is what converts that baseline into growth—turning those Rapid deals into products deployed at scale for our customers. Those are levers within our control, and that's where the company is focused. Another point I want to stress: Autopilot revenue from existing customers adds to what we earn from them on every loan. Same loan, same customer, more revenue per loan. Rapid follows the same pattern — it carries a higher rate per funded loan because it's a fuller offering capable of real-time approvals. So when those Autopilot deals in the pipeline close and Rapid gets rolled out, our revenue per funded loan rises even if the market doesn't move, since we're delivering more value to customers. Jason will walk you through the trailing 12-month trend; that figure has fallen over recent years for reasons we've covered many times and that you already know. What I care about, though, is the road ahead. In the fourth quarter it's trending up, and we expect it to climb further as we head into 2027. Naturally, nobody can forecast the future with complete accuracy. And that's with only a small portion of the pipeline I just sized for you. We aim to be deliberate regarding what goes into our forward-looking statements and what stays out; that's the key point I'd like you to remember; in 2027, revenue per loan can carry the load even when market volume isn't growing. There's no need for us to depend on market volume growth in order to keep expanding our business with our customers; so in summary, the market became more difficult, yet we still moved faster. Autopilot shifted from something we were launching into a pipeline we can size, with excellent customers building on top of it, and we aren't counting on any assistance from the market next year.
Our task is to convert that autopilot pipeline, get these Rapid deals deployed, and help us grow despite the market. With that said, I'll hand things over to Jason.
Thanks, Nima, and I appreciate everyone else who's on the call with us today. Third-quarter total revenue came in at $32.9 million, which is up 1% year-over-year and sits in the upper half of the $31.5 million to $33.5 million range we guided to. Mortgage suite revenue was $17.4 million, a decline of 2% year-over-year, and that lands inside the negative 4% to positive 3% range we talked about on our August call; on our platform we saw roughly 194,000 funded mortgage loans, which is down 4% year-over-year and a bit under the 200,000 to 210,000 range we had expected entering the quarter. As Nima explained, the macro downturn began hitting us in September and caused that shortfall. Per funded loan, economic value came in at $81, which sits at the high end of the $80 to $81 guidance range we gave in August. Revenue from our Consumer Banking suite was $13.2 million, a 5% increase year-over-year, and that's above the top of the negative 5% to positive 1% range we discussed on the last call. Professional services revenue came to $2.3 million, matching our expectations. Turning to profitability: non-GAAP gross profit was $25.4 million, with non-GAAP gross margin at 77.2%, compared with 77.7% in Q3 2025 and 78.3% in the preceding quarter, putting us at the low end of the 77% to 78% normalized range we've referenced. The main reason for the drop from last quarter is the expense of Autopilot models, and that expense rises as the number of loans Autopilot handles increases. Back in August I flagged these costs as modest yet expanding. At this point they've begun to affect the margin line, but we don't anticipate anything material for at least the coming year, and we're monitoring how this develops. Non-GAAP operating expenses came to $20.5 million this quarter, including the sequential increase from our yearly Blend Forum customer event in August. OpEx for Q3 fell 3% versus a year earlier, in line with the guidance we gave in August. Non-GAAP operating income came in at $4.9 million, which is above the top of the $3.5 million to $4.5 million guidance range we had provided. That works out to a non-GAAP operating margin of 14.9%, roughly 2 points better than what we saw in the third quarter of 2025. Free cash flow during the quarter was basically flat, bringing the year-to-date figure to $14.3 million. We ended the quarter holding $44.2 million across cash, cash equivalents and marketable securities, with no debt, naturally. I'd also like to address churn, because I know it got some attention following our previous call. Churn notices during the third quarter amounted to approximately $2 million in annual revenue, which is better than the roughly $3 million we recorded in the quarter before. Looking at the year so far, and factoring in a Q1 with even less churn, gross revenue retention comes in around the mid-90s when measured on a churn notice basis — a very strong outcome for an enterprise software business. As before, the bulk of what we've observed this year involves smaller customers moving toward lower-tier solutions, plus a handful of accounts that shut down or merged amid current rate conditions. To be clear, no new entrants are reshaping the competitive landscape in any meaningful way. The churn I described stems from pain the market is feeling due to the extended period of elevated interest rates, rather than any shift in how we compete. Now, on to guidance. During our August call, we anticipated that industry volume for the fourth quarter would fall by about 8% to 13% compared with a year earlier. However, the latest forecast from Fannie Mae, released in October, now projects a 23% year-over-year decrease. Building in a measure of conservatism, we project fourth quarter market volume of 850,000 to 910,000 units, which represents a 28% to 33% drop from a year ago. As has been evident over the past 2 months, refinance activity responds strongly to rate changes in either direction, so conditions remain in flux. Against that backdrop, here is our outlook for the fourth quarter. We anticipate total revenue of $27.7 million to $29.7 million, roughly 8% to 15% lower than a year earlier. For the mortgage suite, we're projecting a year-over-year revenue decline of 23% to 28%. Again, the difficult macro environment we're dealing with is the main driver of our expectations here; we anticipate economic value per funded loan of roughly $86 to $87, compared with $81 in the third quarter. As a reminder, that per-loan figure moves inversely with volume because of some of the fixed fee arrangements within our customer base. So lower volume in the fourth quarter mechanically lifts our evPFL calculation. Roughly $3.50 of the step-up we expect between Q3 and Q4 comes from this mechanical lift; for the fourth quarter, we expect Consumer Banking suite revenue growth year-over-year of between positive 2% and positive 10%. For Q4, our non-GAAP operating income should land somewhere in the range of $2.3 million to $3.3 million, which works out to a non-GAAP operating margin of roughly 10% at the midpoint. As you can probably guess, that figure sits below what we had anticipated prior to this macro downturn, though part of the revenue pressure does carry through and reduce operating income. One more thing to keep in mind: the fourth quarter outlook reflects only autopilot revenue from contracts already signed as of today. Nothing from the pipeline that Nima walked through is baked into these numbers. Beyond the fourth quarter, I'd like to address two items tied to 2027 — volume and evPFL. Where volume is concerned, the first quarter faces an unusual comparison. Q1 2026 was a robust period for the sector, since market volume rose 37% versus the prior year, and the rate conditions we discussed previously continued right into what is already the seasonally weakest quarter. As a result, we anticipate that industry volume in Q1 2027 will fall roughly 30% year-over-year, a drop at least as pronounced as in Q4; looking at the entire year, our present macro outlook calls for 2027 industry volume to decline by a high single-digit to low double-digit percentage relative to 2026, and the bulk of that year-over-year decrease should land in the first half, given that we are comping against the stronger early portion of 2026. Regarding evPFL, that quarterly number fluctuates with seasonal volume, and for precisely that reason we anticipate it peaking in Q1. That's why a trailing 12-month perspective now appears in the supplemental slides, and I'd urge you to use that as your reference point. Looking at it on a trailing 12-month basis, evPFL stood at roughly $90 a year back and sits at approximately $82 now. Two factors we've previously covered account for the bulk of that decrease: losing a single large customer, and our verification and insurance products shifting to a partnership structure under which we record net revenue instead of gross. Core pricing remained steady across that entire period. Going forward, we anticipate the LTM evPFL figure climbing starting in Q4 of 2026 and carrying on through 2027. Part of that reflects the volume arithmetic I mentioned earlier, while the remainder is genuine economics. Autopilot contracts already signed are converting into revenue, customers keep adopting Rapid and Close, and renewals are being priced with greater discipline. For the full year, we anticipate 2027 evPFL rising by mid-single-digit percentages year-over-year, and we view our estimates as having more upside than downside. Before concluding, I'd like to make a point about capital allocation. We completed nearly $37 million of share repurchases during the first half of the year, though no share repurchases occurred in Q3. There's still $13.2 million available under our authorization, and we feel good about our current cash balances. As we've consistently stated, operational liquidity comes first for our cash, yet buybacks stay the next destination for excess capital, particularly given recent trading levels, and we'll keep assessing that opportunity; to sum up, we produced the third quarter we targeted, revenue landed in the top half of our range, profitability came in above the high end, and sales had a better quarter than we could have imagined. Conditions outside our company have become more difficult than they were last quarter. We've tried to give a straightforward and cautious view of the volumes we expect through Q1 and on into 2027. That said, the factors we can influence — turning our pipeline into closed business, uptake of autopilot, and the expense restraint we've maintained — are what will determine the longer-term result, and each of those is moving in a positive direction. With that, let's start taking your questions.
Nima, you've been upfront that Autopilot deals are being signed for one year at a time, at least until a longer-term value can be settled on with customers. So once customers start really looking inside the Autopilot box and seeing the $600 in lower cost per funded loan, pull-through rates that are 10% to 15% higher, and cycle time that's 2 to 4 days lower, where would you place customers' view of Autopilot on that metaphorical spectrum — scooter versus a Ferrari — as of today?
Yes. It's a good — I mean, great question; and we kind of went into starting to monetize this in July. Our customers have partnered with us extremely well, since nothing, in my view, begins life as a Ferrari, despite all the promise behind it, yet working alongside those customers has produced a great deal of improvement as well as new products. So I would put it this way: I don't wish to take it for granted, but in my own view it is already a Ferrari. And I believe further enhancements to the product will, I suppose, help the world recognize that. And certain customers, the ones nearest to us who have been working alongside us, already view it that way. I don't want to talk up the product too much, yet at the same time I don't want to downplay how big a difference it can make—particularly as we build out additional partnerships around it.
All right, that helps. And then, Jason, could you speak a bit about what's really driving RPO? It rose sequentially, yet it's down 7% year-over-year. How should investors think about how much of that gap reflects demand versus perhaps a shift in sales strategy, where RPO now is of higher quality and will turn into revenue on a more defined timeline, whereas previously there was always the question of over what period RPO would convert into revenue?
Honestly, I'd encourage you to not really devote much attention to RPO. As you're aware, we don't run a pure subscription model; ours is usage-based; and although we do carry certain commitments and some revenue does get deferred. RPO isn't truly meaningful in our case, since different product types carry differing commitment levels from the customer. So it isn't a metric we track, and I'd encourage you not to track it either.
I suppose, regarding Autopilot, with the pipeline having matured as it has, is it fair to say you're now able to offer a bit more detail about the pricing uplift an individual customer might see versus baseline evPFL? And on 4Q, Jason, you mentioned there'd be some revenue contribution there. Could you put a size on it? I'm just hunting for numbers here that I know all of us are eager for, though I get that you'd rather keep this under wraps at this point.
Yeah. That's actually a terrific question. Yeah, I was about to say, I think picking up on that — the prior question was along much the same lines. When we first set pricing on this for those 1-year deals, we deliberately made it less Ferrari and more scooter, for 2 reasons I wish I'd mentioned on the earlier question. First, we're clearly still developing this product, and although things are going well, we want to sustain momentum behind it. Second, the market needs time to digest the changes and to understand and trust them. So honestly, driving adoption of this product is my main focus at the moment. That's why we set it as a flat fee, a total price fee for lenders in year 1, to say, hey — look, we know this is worth more to you, but we don't expect you to commit to some big, huge per funded loan, long-term contract before you genuinely believe in and see the value. So although we don't have specific numbers on what that amounts to, I mean, pricing tiers run from $100,000 or even less, all the way up to $1 million for our customers. It simply depends on their size. And sorry, Jason, for cutting you off.
No, honestly, I think you covered everything I intended to mention and then some. In short, we aren't sharing the specific financials at this time. I'd say our attention is much more on what Nima described — adoption, traction, and product improvement; though, as Nima noted during his prepared remarks, just from the Q4 deals already in our pipeline, it could contribute $5 million to 2027 revenue; so it might be meaningful, but I don't think we're at such an early stage that we're withholding any specific financials for Autopilot yet.
Understood. Next, when it comes to the Consumer Banking suite, could you describe how the home equity product has performed as rates have climbed? What I'm wondering is whether you might be seeing some tailwinds from the substitution effect against cash out refi getting better, or whether home equity remains somewhat vulnerable to demand softening when rates rise?
Right. In Q3, we referenced 26 deals for Q3. For us, the Rapid Home Equity product stood out as a bright spot during Q3, since home equity is a competitive market, and many of the biggest home equity lenders nationwide are on our platform; and the question now is how do you give customers and consumers certainty about home equity lines and loans more quickly, upfront in that flow, with an actual decision — or at minimum a largely confident decision — so the consumer can feel assured that you can assist with their renovation and help them consolidate their debt; and so yes, I believe, yes, definitely, a tailwind exists there, and market demand is strong, including from small lenders, medium lenders and some of the biggest lenders nationwide, to offer something that helps counterbalance their business; so that was a major bright spot for us in Q3. And we mentioned the new top 10 bank logo. They signed up for Rapid Home Equity. We cross-sold another top 10 bank and a top 20 originator with Rapid Home Equity. And so yes, I think that contributed. And rolling these things out is going to be, I think, great for — I mean, for the economy in general. People need to be able to tap their equity in a tough financial environment for consumers, and our customers see that, and they feel that's part of their mission to help them do that.
Congratulations on everything you've achieved with Autopilot. I'm curious—as it matures, a follow-up question: do you see any adjustments to the overall pricing approach from here that might help drive more sales across the whole suite, with Autopilot as the lead? And second, considering the industry outlook going into next year, how are you thinking about your cost structure as we head toward 2027?
Right. Our pricing approach for Autopilot hasn't changed—we've essentially said that for now we'll do these 1-year flat fee deals with customers. That could change next year, and we might even offer some kind of introduction. I really believe there's substantial demand for agentic products in the market, including among lenders who aren't yet on Blend. It pains me to admit they're not yet on Blend—we'd love to have them on Blend. So finding ways for them to come aboard quickly and have certain pieces of Autopilot available when they sign up with us, so they can see the value and get excited about it. I really think it can serve as a market share catalyst for us if we choose to move in that direction. So while historically I haven't said much in these forums about how we'd use that to drive customer adoption, I do believe that as customers get excited about an agentic product like this, it will also pull their peers who may not yet be on Blend toward us. And I'll let Jason address the cost structure question.
See, Joe, we — as you're aware, we've stayed really disciplined when it comes to our cost structure, and we're going to keep at it, I suppose; and clearly, we aren't giving guidance for 2027 at this point, but I think we're — that muscle has been built, and we won't allow it to go to waste.
I've got two things. First, could you elaborate a bit more on what was behind the outperformance in the Consumer Banking part of the business, and what's baked into the guidance for 4Q? And also, any thoughts on how sustainable the growth you're seeing in that segment is, in your view?
Yes. Michael, look, we have a solid customer base in Consumer Banking. And clearly, that held steady in Q4, and that helped us. As the previous question pointed out, home equity is a strength for us too. I wouldn't claim it fully offsets the macro pressure on the mortgage side of the house, but it has performed well. And I think that's likely more due to the strategic push we've made around home equity. But I think that was a major contributor to how Consumer Banking performed.
And I do appreciate the volume — the early volume outlook for next year, and I appreciate some of the forces beyond your control that you're thinking about. Jason, in that volume scenario you're laying out, could you just speak to how much you're able to protect margin and the cash flow profile of the business, if that is indeed what we see in the first half of the coming year?
Yes. Look, as I mentioned in our Q4 guidance, the downside of having a profitable model is that when revenue declines, obviously you're going to — take a bit of a hit on margin; so I can't claim we're immune to revenue fluctuations causing some hit to the bottom line. That said, like Joe's question around cost discipline and expense management, we feel very strong there and able to control our own destiny. I think we're also seeing fairly strong momentum in our customer base, primarily around Autopilot and Rapid, which leads us to think we have real revenue growth opportunity. Macro environment aside, obviously that's a headwind for us, but there are — we have our own tailwinds that we're trying to create to help us counter those.
If we look purely at the deal volume this quarter, would you describe it as a perfect storm of several things — past go-to-market investments, rising awareness, or referenceability for Autopilot? Is there anything particular you'd highlight for how we should think about deal volume momentum going forward, and how these deals will layer into the model?
I wouldn't call it a perfect storm or luck. I've been pointing to late-stage pipeline for a couple of quarters now. Before that, I said the pipeline was growing. Then lenders, I think, saw our innovation progress, both in our core products and in the Rapid and Autopilot suite; and Rapid Home Equity, as someone asked earlier, is very popular in the macro right now. Autopilot changed the conversation. The largest lenders no longer see us only as a point of sale. One point I wanted to make about using — working on Autopilot and pricing and packaging to get new logos. One of the interesting — one of the most interesting things to me this past quarter was the sort of fintech bank using Autopilot MCP to integrate this into their own experience, which isn't something we've done historically, because it was sort of you use our front end or you don't use Blend, and now they don't have to make that trade-off. So I think the combination of those things, plus the fact that we were talking to one of the large industry players about just how much volume is going through Autopilot — it's a lot. In an environment like this, 90,000 applications worked in Autopilot is a meaningful amount of the loan applications that are out there, and it's such an early product. So I don't want to give a forward-looking statement on anything but the pipeline I already talked about earlier — I don't think this was luck. Some of these deals were years in the making and some were months in the making; and some of this is just — it's a timing game, and you keep showing our execution; we keep showing the focus we've shown from a financial perspective, profitability, and driven confidence in our business. So we still have a lot of pipeline ahead of us. On top of that, a lot of innovation that we hope and believe will drive a lot more pipeline in the future. So I don't think we're even getting — I just feel like this is a new era; this agentic world was in a lot of ways a gift to the lending industry, which has historically been very manual and very paper-based. And this has always been something I wanted — to make this industry feel like the most consumer-friendly one out there, so that the people who work in it aren't stuck sitting there buried in hundreds or thousands of pages of documents they have to wade through just to get someone successfully approved for a loan. I feel like as a team we're finally in a position to deliver on that. I don't think we've delivered yet, not by any means, but this is our moment. This is where execution truly counts. That's why I've been spending so much time with the team on execution. I do believe we've been executing very well on the go-to-market side and in pushing the pipeline through to completion. We've been meeting our customers at their sites, at conferences, and at our forum, which took place a couple of months ago. And I think all of this together, over time, adds up to what was a really good quarter for us. And hopefully there are more of those ahead in the future.
That appears to be all the questions we have. Today's call has now come to an end. We appreciate your participation. You're free to disconnect at this point.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete Blend Labs, Inc. transcript - plus 131,000+ transcripts from 7,000+ companies, speaker segments and full-text search - through the EarningsAPI REST API or hosted MCP server.
Get an API key View API docs →For developers and AI pipelines
Programmatic access to Blend Labs, Inc. earnings transcripts and 131,000+ others is available through the
EarningsAPI REST API and the hosted MCP server.
Quarterly plans from $162 - full transcripts, speaker segments, full-text search,
and the /api/v1/transcripts/recent polling endpoint for ETL pipelines.