Good afternoon. Welcome to Affirm Holdings second quarter 2023 earnings conference call. Following the speaker's remarks, we will open up the lines for your questions. As a reminder, this conference call is being recorded, and a replay of the call will be available on our investor relations website for a reasonable period of time after the call. I'd now like to turn the call over to Zane Keller, Director, Investor Relations. Thank you. You may begin.
Thank you, operator. Before we begin, I would like to remind everyone listening that today's call may contain forward-looking statements. These forward-looking statements are subject to numerous risks and uncertainties, including those set forth in our filings with the SEC, which are available on our investor relations website. Actual results may differ materially from any forward-looking statements that we make today. These forward-looking statements speak only as of today. The company does not assume any obligation or intent to update them, except as required by law. In addition, today's call may include non-GAAP financial measures. These measures should be considered as a supplement to, and not a substitute for, GAAP financial measures. For historical non-GAAP financial measures, reconciliations to the most directly comparable GAAP measures can be found on our earnings supplement slide deck, which is available on our investor relations website.
Hosting today's call with me are Max Levchin, Affirm's Founder and Chief Executive Officer, and Michael Linford, Affirm's Chief Financial Officer. With that, I'd like to turn the call over to Max to begin.
Thank you, Zane. We appreciate everyone taking the time to join us. I hope you've had a chance to review our letter to shareholders as it contains a great deal of detail. Amidst increased macroeconomic headwinds, our fiscal Q2 had mixed results. Revenue was at the low end of our expected range, and adjusted operating income came in better than expected. On the other hand, gross merchandise volume was short of expectations, as was revenue less transaction costs, as our mix shifted to more interest-bearing loans and we retained more loans on the balance sheet. We once again reported excellent credit performance as delinquencies fell on a sequential basis. Our continuing vigilance and attention to credit outcomes allowed us to meaningfully increase our funding capacity in January. We also acknowledge a tactical error on our part that hurt our results.
We began increasing prices for our merchants and consumers later in the year than we should have, as this process has taken us longer than we anticipated. This is a lesson we will not soon forget, though it does not change our long-term outlook at all. We have taken appropriate action from implementing pricing initiatives which are gaining traction to refocusing our product development effort on margin optimization and core growth to the most difficult decision of all, reducing the size of our team by 19% today. I believe this is the right decision as we have hired a larger team that we can sustainably support in today's economic reality. But I am truly sorry to see many of our talented colleagues depart, and we'll be forever grateful for their contributions to our mission.
With a smaller, therefore nimbler team, we are focused on achieving profitability on an adjusted operating income basis as we exit fiscal 2023 by executing on three key initiatives. Accelerating GMV growth while optimizing RLTC, engaging consumers to drive greater frequency and repeat usage, and growing Debit+. We continue executing on our strategy to scale our network, make disciplined, high conviction bets in our most promising opportunities, and capitalize on our massive secular tailwinds. If anybody wants to ask me about the recently proposed rule on late fees, please go for it. We'll head to Q&A now. Back to you, Zane.
Thank you, Max. With that, we will now begin our question and answer session. Operator, please open the line for our first question.
Thank you. If you would like to ask a question, it is star one on your telephone keypad. You will receive a confirmation tone as you enter the question queue. You may press star two if you would like to remove yourself from the queue. For participants using speaker equipment, it may be necessary to pick up the handset before pressing the star keys. Our first question is from Ramsey El-Assal with Barclays. Please proceed.
Hi. Thanks so much for taking my question, this evening. I was wondering on the new pricing actions that rolled in a little bit late, what do you see there typically in terms of attrition or other impacts, kind of downstream, when you go about rolling those in? Is that a risk factor for later, or do you have a pretty good idea in terms of what to expect as you roll those pricing actions in over the course of the next few months?
We have seen zero attrition that I can think of. My colleague Trevor will correct me, but it is not a matter of risk of implementation, but it is very much a matter of timing. The process is a little bit more complicated than, in some cases anyway, than simply notifying someone because for a large percentage of our merchants, they utilize something or anything in our set of offerings as far as buying down rates is concerned. The conversation isn't just, "Hey, we need to raise prices on consumers," or, "You need to pay us more MDR." It's inevitably the conversation about how the programs change, what buy downs will look like going forward now that there's a different construct in front of the consumer.
For example, you might see we now have a significantly more visible set of 4% and 5% APRs, not a product or not a program that we featured last year at all, et cetera. It's a matter of underestimating complexity on our part. The other unfortunate reality is that having these conversations in calendar Q4 with merchants is just not something that happens very quickly. We, we don't have much risk in those conversations, but the timing made a little difference.
I think it's also important to know that from a consumer price standpoint, we continue to believe that there is very minimal elasticity. In thinking about the impact on the top line and the top of the funnel, we don't think there's a measurable impact there.
Okay. One quick follow-up from me. I also noted that more of your GMV was coming from interest-bearing loans, and as you called out, the highest ratio in corporate history. Can you just kind of comment on how we should expect that to trend going forward? Is that a rate that should continue to increase? I've noticed that I've seen some, for example, some 0% loans on the Amazon website that I hadn't seen in the past. Could we expect that to kind of come in?
I think it's, generally speaking, reasonable to expect as the Fed rate continues to go up or at least remains high or elevated relative to last year to see more interest-bearing loans versus zeros. That said, the subsidies to reduce the rates or eliminate them entirely come from both merchants and platforms as well as manufacturers, et cetera. Overall, the trend should be expected to be towards more interest-bearing loans. We're certainly still very much in the business of finding ways of offering consumers magical deals that contain no interest at all, which is obviously far more valuable now that the overall borrowing cost for consumers went up a lot.
Our next question is from Rob Wildhack with Autonomous Research. Please proceed.
Hi, guys. The new guidance, especially in the second half of 2023, points to, you know, lower volume and revenue growth and RLTC that's actually gonna be down year-over-year. I know you stuck to the profitability target. How are you thinking about the longer-term margin and profitability of the business? How do you get there given that the growth seems to be slowing before you've really hit escape velocity?
Yeah, it's a great question. We continue to believe that the long-term range of revenue less transaction costs as a percentage of GMV should be in the 3%-4% range. I think what you have a couple of factors going on in Q2 with respect to the timing of how we earn the revenue and how we recognize the expense that distorts it. Given the what we think is a 1-time step-up in loans that are held for investment through our warehouse financing growth, we think that obviously will weigh down the full year number, but still allow us to end up in the 3%-4%.
The reason for that is, as we talked about before, the business is, you know, really a mix of Split Pay in 4 volume, which, you know, has margins that are much lower and very profitable longer-term monthly installment, and the two mix in a way where we can pretty reliably predict that 3%-4%. Additionally, I'd point out that we feel really good about the quality of the assets that we originated this quarter. That is to say the economic content there is really good. That hasn't been a primary driver. Most of what you're seeing is, again, how those yields flow through the P&L.
Thanks. If I could just follow up there. you know, similar question, but more from an operating profit standpoint. The long-term target, I think, used to be a 20% or 30% operating margin when GMV growth was below 30%. You're kind of there now, but still have, you know, a lot of fixed costs to scale. From an operating profit standpoint, how do you think about the longer term margin here?
Yeah. We've not given any update to the framework that we laid out last year. I think that we would probably say we're in the midst of one of the biggest kind of moments of volatility from a macro sense. Not sure that we would hold ourselves to the framework that we outlined a year and a half ago in this very moment. I think part of the reason we laid out our profitability commitment to the end of the year was a reflection of the fact that we were wanting to get ahead of that from a framework standpoint.
Just for the record, this is not the growth rate that I personally like. We intend to grow the business faster. The expectation of we're there now is not the expectation that I have for this business. That said, we will manage credit most importantly as job number zero. Like, we will never allow growth trump the necessity of managing losses and yields, et cetera. There's absolutely no reason to believe that having taken over a quarter of U.S. retail sales, we're gonna attenuate and match some steady-as-she-goes growth rate.
Our next question is from Dan Perlin with RBC Capital Markets. Please proceed.
Thanks. I wanted to explore the long-term profitability question again. From the, you know, the viewpoint, I feel like I've heard you say at conferences, like one of the biggest toggle points is really kind of the human capital aspect of your business. You just, you know, you obviously just did a very large reduction of force here. My question is, you know, that seems to be helpful today, but to the extent that you're able to sustain long-term profitability, are you gonna have to lean into something that requires a lot more automation on your part, and are you doing that? Or are you just trying to match? Now obviously right now, obviously you're kind of matching an expense versus a downdraft in the top line. I'm thinking about this longer term from a sustainability perspective. Thank you.
Couple of different thoughts on that. The RIF is very unfortunate and certainly saddens me greatly and the rest of the team. It is an economic reality that we have to live within our means and match growth of headcount with growth of revenue. Just for the record, what we've done is we've rolled back six months of engineering hiring. This is not a, you know, everything's gonna be replaced by robots, and we're writing a lot of code and we'll continue to do so. We have definitely shifted our geographic hiring focus to Poland, where we've been able to attract exceptional talent at significantly lower cost than Silicon Valley, for example. We have a lot of things that we want to build, and we'll certainly expect ourselves to build it and deliver it.
What we're doing right now is not building all those things concurrently. What we've really done is reduce the service area of engineering projects we're allowing ourselves to invest in, which a year and a half ago or two years ago was exactly the right strategy, and I stand by those decisions. Today, it's a little bit tougher to justify having things that will create the next billion-dollar business three years from now built today. We'll have to build it a year from now.
Got it. If I could just do a quick follow-up on the on the pricing initiatives. The question here is really, you know, as you increase, you know, APRs up to 36% is the cap, and then you're also, I guess, toying with the idea of increasing MDRs on the 0% APRs, which kind of puts a burden on merchants. I'm just wondering, do you kind of foresee any, I guess, diminishing returns associated with that, you know, from a merchant perspective? I know you have to get some approvals, it sounds like, in order for you to actually take those caps up. I'm just wondering how those discussions have gone and what that kind of feels like from a from a merchant perspective.
I think everyone, merchants and Affirm alike, are keen on more volume. As I've repeated often, and I'll say it again, we are fundamentally governed by yield and risk management, so we must maintain the risk frameworks that we've signed up with our capital partners. If we are able to increase the compensation we get for taking the risk, and we really do think of it in terms of MDR APR trade-off. There are many situations where the merchant is more than willing to pick up the increased cost because they want to pass the savings on to the consumer and attract them this way. That obviously works really well for folks with direct-to-consumer brands, where maybe manufacturing is partially owned or fully owned.
In other situations where the brand is already paying us minimal possible and is unable to shoulder any more, then it's the consumer that has to see increased rates. In either of those two cases, if we are able to raise the rates, we will increase approvals. Like, this is fundamentally not about expanding a margin. We're quite comfortable with the margin structure that Michael outlined. We continue targeting those RLTC percentages. Being able to talk to our merchants about helping them sell more in a period of obvious consumer slowdown is a pretty welcome conversation. It is not Not in every case anyway, is it a, "Hey, we're just gonna go do this," because again, we run complex programs.
Part of why this business is so defensible is because vast majority of our merchants have significantly more to do with us than just showing up our logo on their checkout. If you look at their product details pages, you'll see that there are pre-quals and various forms of finding out the true cost to the consumer, which has to be updated for regulatory and just marketing purposes, et cetera. So it's a more complicated thing to execute than perhaps meets the eye, but it is not a difficult conversation with the merchant.
As Michael pointed out, we've run the 36 versus 30 sensitivity tests last year and found that our consumer is actually smart enough to realize that when there are no fees, there are fixed schedules, and there's no compounding difference between 30 and 36 on a $240 loan over 12 months is $0.70 a month. The true cost to consumer is practically de minimis on a cash basis, and our merchants are smart enough to understand that as well.
I think it's important to note that our direct-to-consumer channel, where we have complete control, is probably the best, insight into where the, like, the structural economics are here. That's our most profitable product and channel. We have a very efficient way to engage and monetize that engagement
In our app, I think Max's opening remarks pointed out that one of the ways we're getting to our goals is by driving that engagement back to our own services where we can very profitably engage consumers and we're in full control of that experience.
Our next question is from Jason Kupferberg with Bank of America. Please proceed.
Thank you, guys. I just wanted to start on GMV. I'm just looking at the new outlook there. You talk in the shareholder letter about some slowdown in discretionary consumer spending. Just wanted to take your temperature on how much of the lower outlook on the volume is that versus other factors, whether it be competition or just some tightening of the credit box. Then if you can just talk about what you see ultimately re-accelerating the GMV growth, 'cause I think the math suggests you'd be down around 13% or 14% for the next couple of quarters. Thank you.
It's a great question. The, you know, the discretionary spend is down. We get a pretty good preview of what that looks like, especially around Christmas shopping and Black Friday. From our seats, electronics were down about 11%. Homewares and sports equipment in particular were hovering in the negative high 30s. There's quite a lot of, you know, I'm not sure the right word to use, but folks are digesting the purchases they made during the pandemic. I think those are not transactions that will disappear forever, but I think they're probably gonna remain muted for we expect at least a few quarters of that. On the flip side, credit is always an input.
We set the loss rate that we're willing to live with and our capital partners are willing to live with, then we manage everything towards that. That's why the delinquencies are as good as they look. We have total control, and we are willing to compromise GMV, although don't have to compromise too much of it to maintain industry-winning loss rates. I'm not sure I'm prepared to give you a breakdown, but those are the two fundamental reasons consumers are pulling back their spend. Every time I talk to my friends, CEOs of broad line retailers, they tell me that discretionary spend is down. There's quite a lot of movement into things like consumables, and obviously food prices being higher does not help either. To the re-acceleration point, obviously, we've talked for a long time about Debit+.
I'm sure somebody will ask me, and I'll give you a full update on what's going on there, but I remain very, very bullish on that. We've worked really hard over the last six months. It's hard to overstate just how much work went into the product just over the last couple of quarters. We feel very good about its state of readiness, and we'll start finding out just how much of that food and consumable spend we're gonna be able to shift to Affirm. Our consumers still love us. They still come back to us. You can see that the frequency per user is rising. The network activity is extremely healthy.
I think probably the set of metrics that I would direct all of you to look at if you wanted the how does Affirm win story spelled out very clearly. There's a almost a 40% growth in active consumers year-over-year, almost 40% growth in transactions, three and a half transactions per year per active user, 51% growth in transactions themselves, and 86%, up slightly from last one, is the repeat transactors. So the network itself is increasing density, and that is fundamentally the long-term play. Like, if we are able to pick up more and more of your transactions, we will ultimately have a really good shot at also helping you buy groceries. That is transactions that do not, generally speaking, diminish much in the positive and negative economic environment.
That is where the real re-acceleration will come from. We're also selling to more merchants and launching new projects and new products with them, that too will bear fruit. In terms of new categories, offline and lower AV transactions offline in particular is where I'm most excited about.
Just quickly on Amazon, I think the exclusivity provision of the contract expired January 31. Just any updates there? Thanks again.
You know, I think the world where you can lock up your partners with a 10-year contract and not do much after that is, that's left to card-issuing banks. We're not one of those. The way we maintain our partnerships and hopefully have a right to continue showing up is by showing up and delivering real value every day. I think we feel very good about state of all of our enterprise partnerships.
Yeah. Just real quick. In our queue, you'll see we are breaking down the concentration that you see for Amazon. I think we have a meaningful exposure there. We are a little over 20% of our GMV that is still under-penetrated versus Amazon's share of e-commerce. We still feel like we have a lot of room to grow there. Nothing happened to our business, on to Max's earlier point, on the day the contract terms turned over.
Our next question is from Rayna Kumar with UBS. Please proceed.
Hi. Thanks for taking my question. Just looking at your FY 2023 guidance, you're calling for a cut at the midpoint on revenue of 8% and transaction costs to be cut by 2% at the midpoint. Just wondering, you know, why there's that big differential? Any call-outs there?
Okay. Sorry. Is your question why are we able to...
Why are you cutting, revenue more than, the transaction costs?
Okay. I think the guidance for the back half of the year transaction cost does reflect continued some volatile macroeconomic conditions, including and especially the capital markets, where we would continue to expect there to be a lot of pressure on the yields that we need to generate for our capital partners. I think that's just reflected in the guide, and that's the thing that's happening to us. The thing that we're doing about it is the pricing initiatives that Max alluded to. I think we'd feel better about the world had that, had that already been in the ground and reflected in the mix of GMV that we originated. We do expect that continue to be a source of pressure on us in the near term.
Got it. It's really helpful. Just one follow-up. If you can just provide us an update on how the Shopify partnership is ramping and how that runway for growth looks like from here.
We're very happy with the Shopify relationship. The headline answer is these things take a long time to build out. Again, I love being just a little bit about the complexity of this business as a moat, but it really is that. It typically takes us 2 to 3 years to get to kind of a full deployment because it's such an interesting beast. You know, you have to figure out how to promote the product the right way, and yet you can't over-promote it because then you're gonna be pushing people into debt where they shouldn't be. There's a lot of finesse to figuring out how to get to a full, sort of a fully deployed mode.
You know you're there when you're seeing kind of a 1% improvements and not better than that. We're still in a really happy position where we can roll out an improvement or a project with Shopify and see meaningful improvements come out in GMV or in profitability of the program, et cetera. We're still very much at work. We have significant percentage of our effort dedicated to what we call PBA, Powered by Affirm, and that's the componentry that powers both Shopify and several other platforms for us. We're still very significantly invested in building that out, and there's still quite a lot of opportunity there. Generally speaking, very excited, great relationship. Spend a lot of time talking to my counterparts there.
Our next question is from Andrew Jeffrey with Truist Securities. Please proceed.
Hey, guys. I appreciate you taking the question this afternoon. Michael, you know, by all accounts, it would appear that capital markets are maybe healing a little bit and equity as a percent of the total funding platform is up, you know, pretty substantially quarter-on-quarter. I guess a couple questions. One, you know, how do you sort of assess the state of capital markets from a funding standpoint? I notice you expanded capacity. Two, do you think you're gonna be able to stay below that sort of 10% pre-IPO equity funding threshold through the cycle?
Yeah. The first question first, the markets are healing. I think that, I think that the new year did an awful lot for the debt capital markets broadly. You're seeing the ABS market open up. You're seeing much more constructive conversations with forward flow partners. Max and I spent a lot of time over the past couple of weeks meeting with capital partners of all stripes. The tone is just markedly better than where it was as the volatility appeared to be reducing and the new year really did help. We feel much better today, and yet we are still very much humbled around just how difficult it is to execute and how volatile and uncertain it remains.
You saw the whipsaw this week in and around the Fed's meeting, I think that kind of volatility is something we're prepared to and comfortable with navigating, it does reflect in us being very thoughtful and careful about how we run the business. With respect to your second question, absolutely we will stay below 10%. We think this is near the high watermark for where that number should be.
We think that the seasonality of our GMV, specifically the holiday shopping season late in the quarter and then, of course, in the quarter itself, causes an increase, a pretty big step-up in total platform portfolio that we don't think will continue to grow as quickly through the back half of the year, which means that our funding mix will probably be very stable through the back half of the fiscal year, so you wouldn't expect any meaningful increases in that equity capital required. We would continue to feel confident in our ability to execute both securitizations like we did earlier in January as well as net new capacity with forward flow partners.
We feel good about our ability to do that right now as we sit here today, but nowhere near 10%.
Okay. As a follow-up, just wondering about... Pardon me. I've lost my train of thought. Oh, yeah, on the loan loss reserve, I know you've admonished us not to necessarily consider that as we would a more traditional financial, but can you just discuss sort of the 5% reserve and where you think that goes in the current environment? Should it fall given the slowdown in growth?
I think 5% is a really good number. I think it's. It is obviously linked to delinquencies. Again, apologize if this comes off as admonishment, it's really not. It's just a chance to learn about how this business works. We have a chart in my letter I would really encourage everyone to look at. It shows the delinquency trends at Affirm as compared to some of those traditional players whose measures I think some folks are wanting to apply to our business. We're the only player with the line on delinquencies pointing down, okay? Some players are not as high as others, but the directionality is very different. That's because our asset turns over so fast that you're not building for losses for loans that you have.
You know, it's somewhat of a cheeky statement, but we can't build allowance for loans that we don't own. We can't build ahead for originations that haven't happened yet. What you see here then isn't a judgment about how the back book will deteriorate in the macroeconomic environment. It is a reflection of the quality of loans that have originated recently, given the velocity of the book. What you should interpret it as the 5% is very much connected to that declining delinquency trend that you see. That's a reflection of extremely strong credit performance, much more so than anything else. Lastly, I think we included a chart in the supplement that I would encourage folks to look at.
That just breaks out where the allowance bridged to from September to December and then again where the last 12 months have gone. You'll see the allowance build is a reflection of both growth in assets, but also the actual charge-offs in the period.
Yeah, just I don't mean for it to be admonishment either, but I will attempt to say exactly what Michael said in a probably less careful way. I think it's really, really important to understand the whole point of including this chart. It's not as though our consumers are experiencing less or more stress on average, is that we have through the really short-term nature of the products that we print and the fact that we decide every loan individually, where we think we are not able to take the risk, we don't. The downward slope of delinquency is a direct result of our action. We changed our credit posture sometime starting maybe nine months ago, and we've done it again several times since. Sometimes with finesse and other times somewhat more actively.
The point is, we are in control of the credit outcomes, and we'll continue controlling them. That's really, really important to understand. We're not building allowances for the mistakes we made that we couldn't have predicted three years ago by giving someone a credit card. We make the decision every single time they choose to transact. Has a direct consequence of GMV might be lower because we decided the GMV that is coming to us is higher risk than we want to take on. We do rank risk really well. Reducing GMV a little bit eliminates a tremendous amount of potential loss, and we are in total control of what kind of loss we take on. That is the reason we included that is to just drive the message home.
We're not interested in building up giant piles of cash for losses that we'll make from loans from three years ago because we don't really have a whole lot of loans left from three years ago at all. I hope that didn't sound too admonishing.
Our next question is from Bryan Keane with Deutsche Bank. Please proceed.
Hi, good afternoon. I guess just thinking big picture here, Michael, what surprised you versus the guidance you just laid out, you know, last quarter? Was it the pullback in consumer spend? Was it that you thought the pricing would get all pushed through? Was it the mix of loans? I'm just trying to get a handle on the reduction in the guidance going forward and kind of the surprise in that caught you by surprise.
Yeah. I think it really is the overall consumer demand, which shows up both in the aggregate GMV but also the mix underneath that. I think there's, you know, some good progress that we made. For example, we were pretty happy to get our business with Peloton to actually be ahead of where we thought it was going to be. There's a lot of strength of that program as they returned to some of the programs that we had from years before. Then there was a lot of real legitimate slowdown in the broad line merchants that we are very proud to partner with in some more of the durable goods categories. These are the larger considered purchases. I think we were surprised about that.
Then frankly, we continued to manage credit very tightly and we were probably, and continue to be, as Max alluded to, you know, we're gonna manage credit first. That shows up on the positive side with really excellent credit performance, which ensures that we continue to access capital, and our capital is not a constraint on the growth in our business. Yet it does create some short-term top-line headwind.
Got it. No, that's helpful. Then Max, I'll take the bait and ask about Debit+, the rollout there, and the prospects of profitability. I know there was some hesitation worried about the profitability of Debit+. Maybe you can update on that as well.
You could not have set me up better for that one. Thank you. Sorry? All right. I'll spare you the long story, but sometime about seven or eight months ago, we rolled out the first kind of a seriously sized batch, if you will, of cards to our existing users and began observing. Obviously, you're rolling out a completely new set of credit programs. You're taking overnight or multi-day risk on pay-now transactions. There's a whole bunch of different things that we needed to watch, and it's the kind of thing that you can't really model because you just don't have any real background information. We did that.
Much to my chagrin, sometime by midsummer, we knew that transactions we knew how to do, which is longer term interest-bearing and short-term Pay in 4s, were generally performing fine. We encountered a whole bunch of types of transactions, and I certainly won't get into the details, but there are multiple modalities of using the card that were just fundamentally unprofitable. As we were looking at the usage and the fact that the product is so sticky, consumers would literally shift from using Affirm in any other mode to using the card the second they had access to it, sort of debated the responsibility of rolling out a product that was inherently less profitable and in some modalities unprofitable to users that were very hungry for it, but were not going to transact with our other product.
We spent the last six months just drilling into profitability of Debit+. There are people who know who they are, so I'm not gonna name them and embarrass them, but they spend an uncountable number of hours figuring out how to optimize it. This is primarily machine learning work, where you're figuring out things like probability of insufficient funds in someone's settlement account and that. It's a major body of work that was actually in the end, faster than I expected. The punchline is that I'm very happy to report that now every class of transactions in Debit+ is profitable. To an enormous amount of optimization, again, one of these things where you look back and say, "No one else will go through the trouble.
They'll just print out some revolving line and move on." Our consumer doesn't want a revolving line. They want Debit+. Very excited that this thing is profitable now. The other thing, a little bit less, but kind of even more in the weeds, we saw really good stickiness of the product once you comprehended the value proposition. There's a bunch of wrinkles in onboarding in particular that lost too many people as we were trying to onboard them. We spent the remainder of the time in the last six months just figuring out how to make it as easy to get live with a card, like eliminating a huge number of steps while not losing anything in KYC and all the other things that we have to do. Both of those projects are basically completed.
The way you will know that we are mashing the pedal through the floor, as Michael likes to say, is you'll see Debit+ cease to be its own separate application. Up until now, it's been a standalone app that you have to download, so we purposely put in a bunch of friction so that we would be able to control the spread. Now that we feel very good about the economics and the comprehensibility of the product, we're actually going to integrate it directly into the mainline app. I am not going to make the same mistake I did in the past and put a number out there. I will do that internally, but the team knows exactly the pressure and excitement we have for the product. So extremely bullish. You will see it in your app soon.
As the rest of the team is looking at me angrily, so that's all I'll say.
Remember, the Debit+ product is another one of these channels that we control entirely. Some of the profitability of the product is a function of that, and we feel really good about where that sits right now.
It took a while.
Our next question is from Moshe Orenbuch with Credit Suisse. Please proceed.
Great. Thanks. Most of my questions have been asked and answered. Could you talk a little bit more about the, you know, how much of your GMV you think will be on the balance sheet? You sort of talked a little bit about some of that, versus, you know, what you'd be able to sell and, you know, and the idea of, you know, what in those discussions you were talking about that you're having with your capital markets partners, how much is their pricing to you changing and how much, you know, do you need to raise pricing to, you know, to keep them, you know, perhaps where they had been, you know, prior to this range of, you know, of interest rate increases?
Yeah. It's two factors. There's interest rate increases, and then there is credit. I think we spent a lot of time on talking about why controlling credit is so important for the yield those investors get. That is a point of pretty big differentiation when thinking about us versus some of the other alternatives that some of these forward flow partners could be buying. As that differentiation grows, we know we'll get rewarded for it. Yet, also we do feel the need to increase the revenue content in the loans that we're selling in the form of higher APRs, for example. I think the... We're not giving specific guidance to the balance sheet or to the funding models, through the back half of the year. We do expect the mix that we saw in our second quarter to be pretty consistent with the mix that you would end the year at. You know, I think it's safe.
Safe starting point is to assume that we're flat, flattish, which means we still have our largest funding channel is still going to be the forward flow market. That's where the most of the total platform portfolio will be sitting as a single channel. We did the securitization in the beginning of this quarter, which will allow us to grow that line item and yet that still is on the balance sheet with slightly more leverage to get with the warehouses. Anyway, I would assume flat within certainly within modeling error is a good assumption, which means that our forward flow partners are at the table and still dealing constructively and maintaining their level of commitment throughout the back half of the year.
Thanks, Michael. Maybe just as a follow-up. When you gave the guidance for revenue less transaction costs, did you factor in, you know, kind of a better, worse or comparable level of gain on sale on the assets that'll be sold?
That's a great question. It's worse. We've contemplated that we would continue to have yield pressure with respect to our forward flow partners. We saw that in the pricing conversations that we've had and been having. While we're very confident about our ability to control the asset yields, as Max talked about, it is the case that the rising rate environment has put the yield threshold, you know, higher for all of these programs.
All right. Thank you.
Our next question is from Kevin Barker with Piper Sandler. Please proceed. Kevin, please check and see if your phone line is muted. We will move on to Chris Brendler with D.A. Davidson. Please proceed.
Hi. Thanks. Good afternoon. I wanna try the volume question one more time just to make sure I'm understanding correctly. Just relative to your expectations three months ago, is it fair to say that, you know, there was probably a little bit of consumer moving away from discretionary items, especially discretionary items that would be of a ticket size that would make it an Affirm product? Is that? I would say that's only part of it. What about the sort of the offset of increased consumer demand in a stressed macro environment? Is that still, you know, a factor or is it overall a net negative the way consumers are choosing to spend today? I have a follow-up.
It's a great question. The pullback from discretionary spend is exactly right. I think I already rattled off a bunch of category drops that we are seeing year-over-year, that's certainly factual. We do expect it to continue. Nobody knows when the trough of consumer demand is hit, I don't feel like people are running out and buying couches all of February or all of January. The demand for the program, I think I dropped a juicy stat in the opening part of my letter. We see about $1 billion of demand every week.
I think that's not the same thing as, "Well, great, why don't you guys take it?" Because each one of these loans or each one of these applications has to be underwritten for through the lens of what's responsible for us to take a risk on and what's responsible, frankly, for this person to borrow. Increasing consumer demand is certainly there. I think if we were careless, we could probably grow GMV to astronomical numbers quite quickly. That is most certainly not what we're going to do. We are unique in a sense that we don't charge late fees, we do not profit from delinquencies, we do not celebrate late fee increases, and I'm glad there's some pressure downwards in that particular part of the world recently. Hopefully, the playing field is getting a little bit more level.
The demand is a good thing to have. I think we are now big enough where the overall consumer sentiment makes a difference to our business a little bit more than it used to. We're still growing three times the U.S. e-commerce rate, as people walk away from buying more TVs, for example, it will have consequences. So long as we are responsible lenders, we will feel a little bit of that.
Okay. Great. The follow-up would be sort of a clarification of sort of putting all these factors together, correct me if I'm wrong here. It sounds like, you know, consumers aren't experiencing BNPL burnout. You know, to the extent that your last question or last answer suggests that there's still very much a lot of interest in BNPL, especially in this macro environment. It's not really consumers tiring of the product. It is more your line of scrimmage calls on making sure you're originating profitable loans. Because of the higher interest rate environment as well as higher sensitivity to credit costs, you pulled back maybe potentially at the bottom of the funnel, not at the top of the funnel, but more at the bottom of the funnel.
As you make pricing changes, you know, that you could see a better conversion rate in the next fiscal year, potentially. Is that fair?
That's exactly right. Actually, Michael, I'm stealing his line in this one, but he loves to remind everybody that our loans are ranked in profitability the correlation between profitability of our loans and the internal credit score or even FICO score, if you will, more or less are tightly correlated. In other words, the highest credit quality loans are also most profitable for us. We are not using, pardon the crass statement, poor people to subsidize great deals for rich people. We're actually attributing the cost and profitability quite directly, which means that any time we need to or we decide to improve the profitability of the book, we end up taking slightly less risk at the very bottom. The overall demand for the product is still very strong. We're not seeing any. You know what?
I've had enough BNPL during the pandemic back to my... I'm not sure which credit card I'll jump on here, but no, not seeing burnout. If anything, on the margin, I feel like there's demand for more flexibility. I think the one thing that we're probably seeing, this is a little bit more anecdotal, so take it with a little bit of grain of salt, but in the experiments during Debit+, we looked at sensitivity and consumer demand for longer terms. Obviously people always want longer terms because just a little bit less cash flow hit on the monthly basis. As the overall economic environment softened and consumer pulled back, it seems that at least part of the pull back is actually cash flow dependent versus kind of a general decluttering trend, which is also, by the way, happening.
Yeah. Again, we don't talk enough about this, but we should. We're growing at somewhere between two and three times, we estimate the U.S. e-commerce growth rate to be, and that's despite posting 115% growth in GMV last year. I think it's easy to think about, thinking that the industry has slowed down, but that you have to put into context the overall scale that we got to and how we kinda got there a little bit more quickly. We still feel like it's under-penetrated, we'll get to the kind of numbers that we talked about. I think some of the quarter growth rate numbers are a reflection of the comps.
Again, the growth rate last year was buoyed up by the launch of three major programs all happening at the same time. We're quite proud that we were able to do that, but that does mean that some of the growth rates need a little bit wider aperture to get an understanding of what's really going on. The fact that transaction counts are up 50% year-on-year suggests that consumers are not at all being burnt out by it. They're in high demand for. We are figuring out a way to profitably serve those transactions.
Should have just quoted the transaction growth. I think that's the single highest growing metric actually in this quarter.
Our next question is from James Faucette with Morgan Stanley. Please proceed.
Thanks. wanna ask a couple of follow-up questions, particularly to the one that was just asked. You know, it's understandable in terms of, you know, tightening the bottom of the funnel, as you said, a little bit, where appropriate and to manage that. Any sense for how much that then costs you or introduces incremental friction to bring those people back, whether that be first time applicants or people that have taken out multiple loans in the past and for whatever reason, just don't meet the criteria you're looking for for that incremental one?
That is a great question and something that is extremely top of mind for me. I spend a lot of my time staring at re-engagement stats, which is why you see it in my top three priority, both in my letter and certainly communications to the company. The good news there. First of all, you're exactly right. If you tell someone, sorry, no loan for you, and it's their first transaction, that is not a great first impression, and we will have to work harder to get that consumer back. Perhaps even worse, you can imagine I've been a loyal customer for a very long time, and now you can no longer serve me.
We invest tremendous amount of resources to both the communication of declines and also trying to make sure that we can bring folks back where, you know, appropriate. Part of why we know so well that the rate sensitivity is not actually a major problem for our consumer, certainly at the bottom part of the credit funnel, is because we tested tremendous amount of those communications and just various forms of reengaging the consumer in our own surfaces where we have total control, where, of course, we are able to raise prices and ask for significantly higher down payments and optimize the overall experience. Instead of saying no, we can say yes. The long and short of it is the results are good.
Probably not worth getting into without a whiteboard, perhaps when we see each other in person, I'll show you. We'll probably have to publish a chart for everybody to see. We've tested what happens when we recontact a consumer that we have declined, and what do they do when we tell them, "Hey, you're now approved," or when we tell them, "Here's a different form of transaction that we can approve you for." They're very encouraging in the sense that consumers, especially those that have used Affirm before, are not particularly hurt or offended by the decline because we think we do a pretty good job explaining what happened, and are quite willing to come back and reapply.
I'm, on the margin, confident we'll be able to continue engaging those consumers, and you'll see us actually invest quite a lot in products that enable that re-engagement. That's a huge push within the product roadmap in the next couple of quarters, really. It is very much top of mind and not something that we think is just gonna, you know, be available to us and take for granted. It's an area of extreme focus for me.
Then just as a follow-up is, you mentioned, you know, a couple of times, particularly as it relates to the changes in pricing, et cetera, that some of those implementations took or price changes and work with the merchants took longer than expected, and there was more back and forth there maybe than you had anticipated. What are you being able to do so that in the future, like you've got more flexibility there and can move more dynamically, you know, as rates, you know, you've got to imagine in the very long run, we're gonna move around quite a bit? I'm just wondering how you're approaching that.
Yeah. This is if I have some egg in my face to clean off still, this is the one. I thought myself a bit of a payments expert going back 30 years, and I think I still am. But I completely forgot the part where Visa and Mastercard, whenever they change rates, where any network rule changes, they always operate it in a 6-month schedule with a 6-month notice, and then there's a 6-month implementation window. Late summer, we decided this is what's gonna have to happen, and it in fact takes six months. The way you do this right is you structure these into the contracts. You make sure that these contracts stipulate both what happens when the rates go up and when the rates go down.
You make sure that you know what the transition periods look like, et cetera, et cetera. This is, you know, again, somewhat embarrassingly our first effort of that kind. This is the lesson that we've all now learned here. Certainly I am first in the line of the lesson learning. I think you should not expect us to have to have another one of these apologizing sessions where we say, "Oh yeah, we were gonna change prices, but we took a lot longer than we thought." We've now figured out or we've learned how to price in the right amount of time.
Our next question is from Eugene Simuni with MoffettNathanson. Please proceed.
Hi, guys. Good evening. I wanted to ask about the merchant counts actually. See the data on the slight decrease in the total merchant counts and understand that's driven by smaller merchants as you're showing in a very helpful disclosure on what's happening with the larger merchants. Even with the larger merchants, there's a pretty noticeable slowdown in kind of the incremental merchants added to the network. Just wanted to ask about that, what's sort of driving that in your view, and what's your expectation for that trend going forward, how important it is to the overall growth of your platform for that kind of number of larger merchants, let's say, to keep growing at a decent pace?
First of all, merchant count is a little bit of a vanity metric at our scale. I think Michael has a few promises in the letter. We're going to publish a slightly different metric to make sure it just shows kind of a true state of penetration. Huge merchants, quote-unquote, is a very countable set, and we are very well penetrated there, as I'm sure all of you know. Mid-size merchants are important because these are kind of leaps and bounds of volume that we're picking up, and that's where majority of our, if you will, hand-to-hand sales combat takes place these days and probably has been in the last few years. Those are important. Little merchants are a little bit different. They sometimes become inactive, especially at the really small scale, excuse me, become inactive over the course of a quarter.
The true count of installed merchants or activated, but not necessarily active, is significantly larger than the number we published, and it'd be easy to sort of have an even more inflated vanity metric here. We're trying to be transparent here. The growth of merchant base is kind of a set of weights for the weighted average total of GMV. Obviously, GMV growth is what we are entirely focused on.
Yeah. Got it. Okay. That's helpful. Thank you. Then a quick follow-up. wanted to ask about the Affirm app and kind of the transactions that are initiated through your, through your app, through your website? Think if I'm kind of looking at these numbers correctly, the proportion of that has declined a little bit sequentially. my question is really broader, sort of what are the, you know, what are your initiatives around improving the engagement with the app and again, how important it is to keep investing in it and what are you doing to keep driving traffic through it?
Yeah. To be honest, I don't have it off the top of my head whether it declined or... Oh, yeah. I guess, it's slightly down quarter-over-quarter on a percentage basis. It is super important to us. For the avoidance of doubt, that is a thing that I care tremendously about. There's a little bit of equivalence between transactions that happen in our partner apps. For example, as we grow within Shopify, you know that or imagine you know that you can service those transactions both inside the Shopify app and in our own app, and we always route the consumer towards the most likely place of repayment because, again, job number one, job number zero is making sure that credit metrics remain excellent. The overall re-engagement within the network is what we care about the most.
Our app and our site and our extension and a bunch of other things, and most importantly, to me, at least right now, a card, is where we are investing a huge percentage of our engineering cycles. The app is a required companion to the card. In fact, the functionality is in the app. You're borrowing money in the app. You're not ever borrowing money in the card itself. That is where we think a lot more of this re-engagement will take place. There's some really cool experiments taking place there. I'm staring at a spreadsheet titled Max's Top Twenty, which has 35 elements in it right now. Which are all the projects that we've put in the motion over Christmas break with our head of consumer products, and we're shipping
A couple of those every week now. I'm very happy with the velocity of the experiments we're doing there.
There's just also some math that's really important when we are scaling programs like we are with the largest platforms, you know, e-commerce players and Amazon and Shopify as an example. All of that growth, all those transactions are obviously not starting on our platform. A lot of growth is coming from there, which means that, like, I'm actually very impressed we've been able to hold it as constant given the rapid rise in growth in these partners. I think to Max's earlier point, the health of the network is reflected in those engagement stats and the user stats and growth in transaction counts. The fact that we're holding serve on the level of engagement through our properties today is a really encouraging sign given the growth that's happening away.
As those growth rates attenuate, you're gonna see our share pick up.
Our next question is from Andrew Bauch with SMBC Nikko Securities. Please proceed.
Hey, guys. Thanks for squeezing me in. Just wanted to hone in on one of the comments in the shareholder letter, where you say you're redirecting R&D efforts towards margin-improving projects. I mean, I guess how much of that is tangible in the current guide, and is it fair to assume that that is a bottom line margin improvement variable rather than, you know, any kind of shift to potential long-term RLTC numbers?
Yeah. The short answer is, a lot. When we talk about some of the speed of, for example, taking pricing, we would have a lot more of the impact in this year's guidance had we acted earlier. Especially true given the size of the balance sheet that we're sitting on right now. We feel like there's a lot of very long, very big initiatives to improve the margin, and that is up in the revenue less transaction costs line item much more so than in OpEx.
We feel like there's a lot of very big and structural improvements that we're gonna be able to make. They don't really show up in the vertical periods in Q3 and Q4, just given the timing of so many of the flow throughs for any loans on the balance sheet, for example. So that's where the focus is. The effort is reflected in our guidance, you're not gonna see the full benefit of all of the efforts until the quarters play out in the back half of the calendar year.
Makes a lot of sense. The sensitivity that you guys have historically provided, of RLTC relative to rates, any material change in that to what was provided last quarter? I know that heading into fiscal first quarter, you know, you had narrowed the range of impact, you know, given where rates could potentially go, but just wanna double-check on that one.
No, no material change. I think we're thankfully staring down what looks like a more flat rate curve, which I think is allowing us to focus our efforts on making sure we create profitable units at the kind of peak rate curve environment. Further stress above that would continue to have the same reaction in our framework.
Got it. Thank you, guys.
That is all the time that we have for the Q&A today. I would like to hand the conference back over for closing comments.
Thank you, everybody, for joining today. We look forward to speaking with you again next quarter.
Thank you. This will conclude today's conference. You may disconnect your lines at this time, and thank you for your participation.