TDUP - NASDAQ

Good afternoon, ladies and gentlemen, and welcome to the ThredUp's Q3 2025 Earnings Conference Call. [Operator Instructions] This call is being recorded on November 3, 2025. And I would now like to turn the conference over to Ms. Lauren Frasch. Thank you. Please go ahead.

Good afternoon, and thank you for joining us on today's conference call to discuss ThredUp's Third Quarter 2025 Financial Results. With me are James Reinhart, ThredUp's CEO and Co-Founder; and Sean Sobers, CFO. We posted our press release and supplemental financial information on our Investor Relations website at ir.thredup.com. This call is being webcast on our IR website, and a replay of this call will be available on the site shortly. Before we begin, I'd like to remind you that we will make forward-looking statements during the course of this call. Such statements are based on current expectations and assumptions that are subject to a number of risks and uncertainties. Actual results could differ materially. Please refer to our earnings release, supplemental financial information, and our Forms 10-K and 10-Q for more information on these expectations, assumptions, and related risk factors. We undertake no obligation to update any forward-looking statements. During this call, we will present both GAAP and non-GAAP financial measures. A reconciliation of non-GAAP to GAAP measures is included in today's earnings press release and the supplemental financial information, which are distributed and available to the public through our Investor Relations website at ir.thredup.com. Now I'd like to turn the call over to James. James?

Good afternoon, everyone. I'm James Reinhart, CEO and Co-Founder of ThredUp. Thank you for joining our third quarter 2025 earnings call. Today we'll discuss financial results for Q3 and update our expectations for Q4 and fiscal year 2025. I will provide an update on our perspective about the consumer, discuss ongoing innovation in our AI-driven product experiences, and end with a reminder on our compounding competitive advantages in the growing resale market, specifically how we expect new product development will increase that advantage in 2026. I will then hand it over to Sean Sobers, our Chief Financial Officer, to talk through our financials in more detail and provide some guideposts as we look ahead to 2026. We'll close out today's call with a question-and-answer session. First to the results. The third quarter was our strongest year-over-year growth in nearly 4 years and the fourth quarter in a row of accelerating growth. Revenue growth accelerated to 34% year-over-year. Gross margin was 79.4% and adjusted EBITDA was 4.6%, all of which exceeded expectations. Once again, these results were driven by exceptional customer growth and orders in our business. I said this last quarter, and I'm pleased to say it again, we acquired more new customers in the third quarter than at any other time in our history with new buyer acquisition up 54% year-over-year. Active buyers were up 26% year-over-year and orders were up 37% year-over-year. Our approach in 2025 and into 2026 is straightforward: Maintain our gross margin efficiency, gradually expand the bottom line, but largely reinvest incremental dollars we generate back into growing our marketplace through product improvements, marketing spend, and long-term innovation. Turning to the macro. We talked about the impact of tariffs at some length on our last 2 calls, so I will not belabor those points here. Overall, we believe the effect of tariffs and the closure of the de minimis loophole have been a boost to acquiring new customers and could be a structural tailwind going forward as prices rise in the apparel market. Our strategy is to take some price, but largely improve our competitiveness on a relative basis. At the same time, we remain cautious on the state of the broader American consumer and believe that price and value will be of utmost importance this holiday season. While this could theoretically be beneficial to the secondhand market by enhancing the value of comparative offerings, we think a reduction in overall holiday spending or a wallet share shift to new gifts is something we'll have to navigate adeptly. Turning to the product and customer experience. While many of our customer-facing features over the past 18 months specifically drove improvements in our funnel and margins, the third quarter was best characterized as a consolidation and clarification of our mission, vision, and value proposition. In late September, we launched a fully rebranded experience on ThredUp. The unifying theme is "Fashion, Meet Forever, which speaks to our ambitions of building a more emotional long-term relationship with our customers. ThredUp has been a brand mostly defined by logic and quantitative rigor over the past decade. And while we will never abandon that part of our DNA, we know shopping is inherently emotional. And by tapping into our customers' hearts through storytelling and cultural relevance, we can elevate both our brand and secondhand shopping to new heights. We saw the green shoots of this in October as it was the best month for new customer acquisition in our history, up 81% year-over-year, driven primarily by historically low acquisition costs. Of course, as a customer-assessed company, we did not miss the chance to launch our rebrand alongside 2 powerful new product features, the Daily Edit and the Trend Report. With the Daily Edit, every customer will receive a newly personalized feed of 100 items that are refreshed daily. This was a major technical advancement in our personalization capabilities, powered by AI models we've trained in-house that can generate real-time user and item embeddings, allowing us to better understand each customer's style preferences and serve them a fresh curated feed every day. The Trend Report is using AI to combine macro and social trends alongside internal search and customer trends, then generating imagery and style feeds in real time that help customers shop what's on trend. Now let me turn to selling on ThredUp. Since the founding of ThredUp more than 10 years ago, we've maniacally focused on building competitive advantage in our supply chain. Our investments in infrastructure and data have been central to the success of our marketplace, expanding ways we can process clothing at ever-increasing levels of scale and profit. Our investment in building a novel, dynamic, and robust data layer for secondhand clothing has enabled us to develop additional ways to compete in the evolving resale market. The first new supply growth vector we built on top of this core infrastructure was our Resale-as-a-Service business or RaaS, which now powers resale for dozens of brands. This month, we are launching RaaS programs for New York & Co. as well as Cotopaxi, a brand that is near and dear to my heart as someone who loves the outdoors. It's the first large brand to launch after our RaaS strategy shift 6 months ago, and it's just one of many expected to come over the next few months. Our Cotopaxi launch is a showcase of the suite of services we can power for brands, including take-back programs and resale shops as well as cash out programs for customer acquisition and bulk consignment for inventory management. Earlier this year, we launched our second supply growth vector, The Premium Kit. With virtually no marketing investment, this product was an instant hit with sellers and has grown to be more than 20% of the supply in our marketplace. Premium kits deliver superior monetization for sellers, access to in-demand products for buyers, and accretive margins to ThredUp compared to our regular kits. Today, I'm excited to announce the third vector of growth, which is the launch of direct selling on ThredUp, often known as peer-to-peer. While currently in a closed beta, given the way direct selling is expected to impact buying and selling on ThredUp, I thought it important to detail in advance our approach to serving this large part of the resale market. We have been working on the launch of direct selling of ThredUp for more than a year, but I personally have been working on this strategy for many years. I felt strongly there was an opportunity to serve this market as resale became more mainstream, mobile technology matured, and our operations hit a level of scale and margin where we could build a superior, differentiated customer experience. That time is now, day 1 of direct selling. But let me explain. The problem for sellers in the peer-to-peer market is the friction that still exists in listing, pricing, fulfilling, and servicing the items available for sale. Many items don't sell. Those that do don't always touch the right price and post-purchase management of returns and seller reputation becomes an ongoing headache. The result is that most casual sellers participate for a while or they mix and match across peer-to-peer platforms, but they never love the experience. While public data is hard to find, our longitudinal research has suggested that the majority of items listed on current peer-to-peer platforms never actually sell. For buyers on peer-to-peer marketplaces, it's very much buyer beware, a lack of quality merchandising and curation, low trust or buyer recourse in the event of a bad transaction keep many buyers from shopping more than periodically. For platforms, the incentives are to race to the bottom on fees to acquire sellers and to encourage as many listings as possible. This leads to rampant product pollution, limited curation, and the flea market quality that leads to short-term success, but long-term value erosion with weak network effects and limited moats, a new peer-to-peer marketplace pops up every 5 to 7 years, skinning buyers and sellers off the top with the renewed promise of that it will be better this time, join us over here. The fact is that this is a big market, and we believe it's mostly broken. Against that backdrop, here's our new approach. First, our marketplace will focus on casual sellers, the exceptionally large long tail of sellers who consistently get crowded out. The number of items the seller can list will be based on their selling success. Flooding the site with low-quality items will not be an option. Second, sellers will be independently verified so that buyers will be able to shop with total confidence. We plan to mitigate the potential for fraud at every opportunity. Third, sellers will not pay fees to list items. ThredUp will provide premium listing, merchandising, and photography tools that make the sellers' life easier. We believe if done right, that suite of tools will be worth paying for over time. Finally, and unique to ThredUp, sellers will have a seamless experience to choose between direct selling and the Clean Out Kit to meet their needs at any point in their selling journey. ThredUp is now a one-stop shop for most apparel selling needs. Turning to buyers. We are excited to solve the most important parts of buyer friction. First, returns. We believe the single biggest challenge with the peer-to-peer model is seamless returns. Leveraging our decade-long investments in our supply chain and infrastructure, we can now see this as an option to buyers given our power to resell return items in our marketplace. Second, trust. With every seller vetted and ThredUp's brand and customer service standing behind our sellers, buyers can shop with confidence. Third, we will bring standards of merchandising, listing quality, and curation to the peer-to-peer buying experience. We will bring a new wave of merchandise to buyers, but in an organized and thoughtful way backed by the Generative AI products we launched over the past year. And we will be methodical in our rollout, opting for quality and long-term defensibility over quantity. We acknowledge we're in the early days of this new vector for growth, but we are excited to bring our experience, expertise, and unique assets to solve this large customer opportunity. We believe the supply and demand we can unlock in this effort will further accelerate our flywheel for years to come and that this launch couldn't be more timely given the economic uncertainty present for many American households. Finally, before I turn it over to Sean, let me place some of the work in Q3 into the context of our longer-term strategy. On our last call, I discussed in detail the three important competitive advantages we've been building. First, our operational infrastructure and supply chain continues to prove a defensible asset. Having invested more than $400 million in infrastructure, software, and data to invent how a managed marketplace can work at scale, we are now capable of building customer-facing experiences more rapidly on top of it. Our RaaS business, our premium kit, and now the next generation of direct selling are examples of business lines built on top of this core infrastructure. Second, we believe the investments in a unique proprietary data layer have helped us build a direct listing beta product that can work better for sellers while providing endless ways for buyers to shop well-curated merchandise. Third, marketplaces are hard to build and sustain. But when you get the flywheels going, they are very hard to stop. Our marketplace has exceeded over prior years, primarily through building a quality transactional experience. By updating and elevating our brand, we have the potential to deepen customer attachment and stickiness, making ThredUp a household name for years to come. In expanding ways that customers buy and sell on ThredUp with the launch of direct listings, we believe that over time, we can increase our wallet share as well as widen the moat in our marketplace. With that, I'll turn it over to Sean to talk through the financials in more detail.

Thanks, James. I'll begin with an overview of our results and follow-up with guidance for the fourth quarter and full year 2025. I will discuss non-GAAP results throughout my remarks. Our GAAP financials and a reconciliation between our GAAP and non-GAAP measures are found in our earnings release, supplemental financials, and our 10-Q filing. We are extremely proud of our Q3 results in which we accelerated revenue growth and exceeded our adjusted EBITDA expectations. For the third quarter of 2025, revenue totaled $82.2 million, an increase of 33.6% year-over-year. Our performance was driven by investments into marketing and inbound processing that drove our marketplace flywheel. As we discussed on our last call, we started spending on marketing and processing earlier in the quarter, which allowed us to generate our significant top line beat. These investments, coupled with improved buyer metrics resulting from a series of new customer-facing products we've rolled out over the past 18 months, generated our fourth consecutive quarter of accelerating growth. These drivers resulted in another record quarter for new buyer acquisition with new buyers up 54% year-over-year. We also benefited from repeat purchases by new buyers acquired earlier in the year as well as improved conversion for both new and existing buyers. We finished the quarter with 1.6 million active buyers for the trailing 12 months, up 25.6% over last year, while we had 1.6 million orders in the third quarter, up 37.2% over the same time period. For the third quarter of 2025, gross margin was 79.4%, a 10 basis point increase versus the same quarter last year. Our outperformance versus our expectations was largely a result of higher average selling prices due to the rapid growth in our premium supply offering. Adjusted EBITDA was $3.8 million or 4.6% of revenue for the third quarter of 2025. We improved adjusted EBITDA margin by 410 basis points over last year as we leveraged our multiyear investments and benefited from our revenue outperformance while still making investments into marketing and inbound processing in order to drive our top line. Turning to the balance sheet. We began the third quarter with $56.2 million in cash and securities and ended the quarter at $56.1 million, reflecting just $100,000 in cash use. We are proud to have generated $2.4 million of free cash flow for the quarter and $3.4 million year-to-date. We continue to expect to be free cash flow positive for the year. We spent $3.7 million in CapEx in Q3 as we made several opportunistic investments in order to support automation. We now expect CapEx for the year to be closer to $10 million, similar to what we expect in 2026. Now I'd like to provide a bit of context for our updated guidance. Though we remain cautious on the current consumer environment heading into a highly competitive holiday season, we are pleased to be raising our top line expectations for Q4 to align with the positive trends we are currently seeing in the business while maintaining our Q4 EBITDA margin outlook. As has been our strategy throughout this year, we see continued opportunity to invest in marketing and inbound processing to drive growth. With contribution margins in the low 40% range and healthy LTV to CACs driven by scale and recent improvements to our product experience, we plan to flow any incremental dollars above our guide back into our growth-driving opportunities. With all of this in mind, the fourth quarter, we now expect revenue in the range of $76 million to $78 million, representing 14% year-over-year growth at the midpoint and $3 million higher than our previous outlook. The sequential step down reflects the expected seasonal slowdown in resale around the holidays, combined with our planned pullback in marketing dollars as CAC spike during the highly competitive period. Gross margin in the range of 78% to 79%, adjusted EBITDA of approximately 3% of revenue in line with our previous expectations and basic weighted average shares outstanding of approximately 126 million shares. For the full year of 2025, we now expect revenue in the range of $307 million to $309 million, reflecting 18% year-over-year growth at the midpoint. This updated view is $8 million above our previous guidance, incorporating our Q3 beat and the raised outlook for the remainder of the year. We are raising our gross margin range to 79% to 79.2%. Adjusted EBITDA of approximately 4.2% of revenue, this incorporates our Q3 beat while maintaining our Q4 outlook. And basic weighted average shares outstanding of approximately 122 million shares. As we look into next year, our planning process contemplates 2026 revenue growth in the low double digits, in line with U.S. online resale industry growth expectations. On EBITDA margins, we are planning for slightly better expansion than we currently expect in 2025. Since we are planning to iterate on and roll out direct selling methodically throughout 2026, our current forecast does not include this new growth vector. We will provide a more detailed outlook on our Q4 earnings call in March. James and I are now ready for your questions. Operator, please open the line.

[Operator Instructions] And your first question comes from the line of Ike Boruchow from Wells Fargo.

Congrats. A quick clarification and then a question. Sean, on the comment you just made on next year, just to make sure I heard you right. So initially for next fiscal year, you're assuming revenue can grow low double digits with similar EBITDA margin expansion that you're putting up this year, which I assume is 4.2% relative to the U.S.' 3.3% last year. So basically 100 bps margin.

Yes. So we said a 90 bps expansion we expect for '25, and we'll do slightly better than that in '26.

And then, James, maybe just to talk about the revenue. And again, I'm not nitpicking it all. I'm actually trying to understand how you think about revenue growth because it's been a roller coaster the last 12 months and a good way for you. But just how are you thinking about the compounding effects of customer acquisition, the experience improvements, all intertwined with what is a much lower growth rate next year that you're starting to plan? And then really just longer term, I mean, this business IPO-ed doing, I think, 25% plus and then it went negative last year and now you're in the 30s. So I guess just what's a sustainable growth rate for this model as you kind of see it over a multiyear period?

Yes, I mean, I think we've been pretty consistent saying we want to be a Rule of 40 company, which our long-term EBITDA model is 20% to 25%, which implies growth in the high teens to 20%. And I think that's the path that we're on. I think the guide for this year is 18% coming off a flat '24. I think given it's the first week of November, as we look to '26, I think starting with kind of low double-digit industry growth rate is probably a good place to start. And look, I think the plan for '26 and then into '27 and beyond is very similar to '25, which is let's methodically expand EBITDA. As Sean mentioned, we're going to continue to expand rates similar to last year, if not more, and then flow those dollars back into the growth rate. And I think if you look at the sort of anatomy of 2025, that's exactly what we did, which is we took dollars and we flowed them back through, and that produced 4 quarters of accelerating growth. So I think that's the same playbook for '26. I just think given all the uncertainty in the economy, let's turn over some more cards, right, before we guide the full year in '26. But I think we feel very good about '25, and I think we feel very good about the multiyear path to being a Rule of 40 company.

And your next question comes from the line of Bernie McTernan from Needham & Company.

Maybe just a couple on peer-to-peer to start. How will these products look on your website relative to goods coming in from the Clean Out kits? I guess, would be question one. And then maybe second on that, if you just talk to the unit economics of $1 of the expectation for what the unit economics of the peer-to-peer sales will look like versus the traditional Clean Out kits?

Yes. Hello, Bernie, yes, I mean, the product displays will actually look pretty sharp. I think we've done a lot of work using AI to produce high-quality imagery. So I think you're going to see best-in-class imagery for how they look on site. And in early beta, we're already seeing that come true. I think you really do see a rich set of products. So my guess is that consumers are actually going to really appreciate the diversity of the imagery and the quality product experience that we've put forward. So I actually feel very good about that. On the unit economics, I think we have built a variable unit economic model that supports peer-to-peer being a strong long-term EBITDA driver, right? And so typically speaking, you have lower top line, like revenue from peer-to-peer, but it generates superior margins because of the way that it flows through on a variable basis. You don't touch all the items in the same way. Sellers make more money, right? Buyers are happy. So I think, Bernie, if you kind of look across similar models that have been around a long time, they generally generate superior long-term profit pools. And I think we can build a superior customer offering and benefit from great economics. So I feel good about both areas, but I would caveat all that to say that we're early in the journey. And normally, we wouldn't talk about this for some time. But given how much it will impact the customer experience, it will become obvious to anybody browsing the site, the difference. And so wanted to be more detail oriented and explain it in advance.

And your next question comes from the line of Bobby Brooks from Northland Capital Markets.

I was hoping to get a little bit more granular on the buyer growth. So overall, up a very healthy 26% year-over-year, but with new buyers up 54% year-over-year. So I was just wondering if we could hear of that 320,000 or so buyers you added year-over-year, maybe what the mix of that was towards new buyers? And then secondly, could you discuss how your marketing approach may differ between getting prior buyers back on the platform versus new ones?

Yes, sure. Hello, Bobby. Yes, I mean, as you know, the active buyers is a trailing metric, right, versus new buyers is a quarterly metric. Generally speaking, about 1/3 of the total new buyers -- the total buyers that we're adding at any point, 1/3 of them are customers that we had previously who had churned that we resurrect as customers. So think about it as 1/3, 2/3 is customers we've seen before. And I think as we go forward, I think we'll continue to focus on driving new buyer growth. And with the rebrand, I think we are expecting that as we move up the marketing funnel a little bit, we'll actually capture more lapsed buyers in there. And so I think we're optimistic we'll actually recover resurrect customers who've churned maybe who don't get our e-mails or push notifications the way they do today. So that's kind of the approach to the marketing mix. Hopefully, that's helpful.

Yes, that's definitely helpful. And then I was just -- I was surprised that in your opening remarks, you mentioned how you won your first new large RaaS partner since the shift in that go-to-market strategy for RaaS. Could you help me understand why there was kind of a large lag between the change in the go-to-market and the new partner that joined this quarter? And also point it seemed like your commentary, you had some pretty good visibility or confidence on new partners joining that channel over the next year or so. Could you just discuss what's driving that visibility and confidence?

Yes. I mean it's -- we announced the new strategy 6 months ago. Most of the contracts that you have with these brands are multiyear contracts. So it's really just a lag effect, Bobby, of how contracts come up for renewal. Think about it more like enterprise. And so I think now we're getting into renewal season for retail brands end of this year and into next year. And so the pipeline feels very good for some of these brands to either sign for the first time or switch over. But that's the way I think you should think about it. It's the lag around the contract renewal process.

And 6 months isn't a long period for an enterprise transaction.

Yes. Yes, exactly. I'm actually delighted about the speed upon which a lot of customers are reaching out to become part of the RaaS portfolio.

Yes, I agree with that. I had a -- I thought it was a little bit long, but yes, 6 months to turn around on an enterprise channel is quite quick. Just any more on -- so it just seems like general contract timing is giving you that confidence in landing some more new ones?

Yes. Yes. No, definitely. I think we feel like the pipeline is good, and we'll see how Q4 kind of concludes, but I think you'll continue to see momentum with us launching new brands.

Fair enough. Congrats on a great quarter I'll turn in the queue.

Thanks.

And your next question comes from the line of Dylan Carden from William Blair.

James, sorry if you said this. So will the direct -- the peer-to-peer product be listed alongside the consignment? Is this sort of a separate site? And I'm just kind of curious a broader discussion of sort of the synergies that you see between these 2 businesses, I guess, beyond just simply sort of the captive audience.

Yes, Dylan, you'll be able to browse them together or separate, right? And so very much if you think about Amazon, right, the ability to shop 1P or 3P consistently. I mean I think this is a very well-established convention in commerce these days. So consumers will be able to flex in and out of it however they like. And what was your second question, sorry, Dylan?

Kind of the synergies between the 2 platforms and well…

Yes. I mean the primary driver is the feedback from sellers. So we've heard consistently for some time how many sellers are sending some stuff to ThredUp but then selling high-quality stuff on other platforms, specifically peer-to-peer platforms. And so in the research that we did, we found there was really a compelling opportunity to centralize all of the sellers' needs and give them that flexibility to do both. So I see the opportunity to really consolidate selling of secondhand online through the ThredUp platform. And then I think there's huge benefits on the buyer side, buyers get greater selection. I think that can help drive customer acquisition efficiency. And then in our DCs being able to leverage our supply chain and logistics network. So I see synergies on both sides, but primarily, it's all about sellers and supply over time. And I think this just gives us another tailwind in that market.

Awesome. And then for either of you, the deleverage -- or sorry, rather the leverage in the marketing line item is kind of impressive. I'm just curious if you could speak to sort of efficiencies you're seeing in marketing. And then maybe just sort of the lag effect in your customer acquisition given kind of the active customer growth versus the revenue growth.

Yes, Dylan, we continue to see sort of historically low CAC. I mean I think it's a combination of the product experience being better. We talked about the conversion rate last quarter. That has continued to improve. I think the ad market, we've continued to find opportunities buying ads, whether it's on Google or Meta and just really taking advantage of some soft spots as buyers now -- sorry, as other brands sort of navigate tariffs. And then from a lag effect, yes, I mean, the new buyer growth continues to be exceptionally strong, and you'll see active buyers sort of trail that. But we feel very, very good about our ability to be aggressive in market acquiring customers. And I think you should see that this year. And there's no reason, frankly, we can't continue to acquire customers next year at a similar or better rate, given that we're going to be spending more dollars in marketing next year over '25. And this year, we spent more than '24. So I think the trajectory on the acquisition side is as good as it's ever been.

And your next question comes from the line of Dana Telsey from Telsey Group.

Nice to see the progress. Can you expand, James, on the premium selling kits, what you're seeing there and what percent of the mix do you think it could become? And also on the AI investments that you've made, how that's leading to conversion? Is that a step-up from last quarter? What are you seeing there? And then I just have a follow-up.

Yes. Sure, Dana. Premium has really grown nicely this year from really 0 to north of 20%. I think there's more room to run on it, Dana. I mean, certainly, the buyers that we have are really drawn to the premium mix. And so I think we're continuing to invest in scaling that. I think it'd be premature to know what to predict like what the steady state rate is of premium, but I would say it's probably higher than it is today. And what we're seeing is the premium brands out there that customers are loving are the ones that we're seeing the fastest growth. FARM Rio, Mac Duggal, Vuori, like those are all brands that I think are hitting the sweet spot of premium, and we just want to get more of them. And then on the AI side, I would say the product conversion rates continue to trend positively. I think the rebrand launched September 22. Alongside of that, we launched a new personalization strategy that was very powerful. And then we launched this Daily Trend Report that also, I think, is capturing what's in demand and using our AI tooling to deliver that in real time for customers. So you have better personalization plus better curation and trend forecasting on top of that having superior products flowing in. I think that's a recipe for the success that we saw in Q3. And I expect that to continue not just in Q4, but into '26.

And then the new buyer growth, which is very impressive, demos of the new buyers and what you're seeing? And then just after that, just marketing spend, how do you see marketing spend in '26 compared to '25?

Yes. I mean I think on the marketing spend side, we're going to continue to spend marketing at higher rates than we -- on a percentage basis, the same, but more dollars overall. I think Sean said a number of times, we think it's the last thing we'll probably try and leverage in the business as we pursue the growth strategy.

And the demos of the new buyers?

Demos remain the same, remain the same. Yes, it's been a very similar story. I think it's probably the third quarter in a row that we've been talking about record buyer growth and the demo of the buyers is consistent with what we've seen previously.

And your next question comes from the line of Matt Koranda from ROTH.

Nice job. I guess I just wanted to hear a little bit more unpacking of what you think is enabling the large acceleration in sales in the third quarter. Would you say it's the new tools that are available? Is it sort of the new buyer growth that you've alluded to? Are you getting more repeat from existing core customers? Maybe just unpack the trends that are driving the acceleration in the third quarter? And then also just curious on the fourth quarter growth that you guided for 14%. I guess, is that what you have observed actually quarter-to-date? And what causes sort of the deceleration there relative to the 30-plus percent growth you've been on?

Yes. Let me tackle the fourth quarter piece. We've baked in everything we've seen to date in the guidance. So you can do the math how you want on that. But the midpoint of the guidance is 14.5% on Q4.

And typically, Matt, like October remains very strong, but then the minute you switch to holiday, you start to see wallet share shift to new gifts. So I would say that October year-over-year was stronger than the 14%, but we tend to see November, December be a little softer.

I would add in the difference between like Q3 and Q4 is the comps that we're comping off of last year because last year's Q3 was a minus 10% growth and the Q4 last year was plus 10%. So if you think about it on a 2-year stack, actually, Q4 is growing really nicely.

Yes. Now I think I understood it.

And then as far as like your first question on what's driving Q3, you sort of hit the tools, the buyers and the macro. I would say that the -- generally speaking, you're seeing consumers looking for value. So at the macro level, I definitely feel like we are being sharp on price and the value proposition. And I think probably on average, drawing more customers in with that approach. And then I think the tooling that we've built is improving conversion. And so I think we're having success in the story that we're telling in the market around ThredUp has great brands at great value. And then when customers are getting to the site, they're converting at higher rates. And I think that's been driving the flywheel for a few quarters now, and I think really worked exceptionally well in Q3, and we're optimistic that will continue.

Makes a lot of sense. Maybe just one on the direct listings. Exciting to see that development, and I see the betas on the site right now. I guess how will the process work for sellers to begin to be vetted? And then I noticed it looks like no fees right now for sellers. So how do you envision, I guess, layering in seller fees over time as you get the volume ramped up in that channel?

Yes. I think on the vetting side, we're going to do a couple of things. So one is we have more than 0.5 million sellers on ThredUp today that we have already vetted, right? So the people who are already sending us Clean Out kits are vetted for peer-to-peer in advance. So I think we have a huge head start. And then on top of that, we're going to do -- we're either going to work with some third-party vendors to do vetting ourselves or just take an extra step for customers, whether that's making us scan a picture of their driver's license, right, or scan a QR code that we send to their house. But we're going to take seriously this idea of ensuring that these are high-quality sellers. I think it's become such a problem in the broader market of fraud and low quality. So we're going to take that seriously and probably invest on average more than other peer-to-peer markets might. And as for fees, don't get me wrong, we are going to monetize the transaction, but we'll generally be charging buyers some percentage of the fees. And then for returns, which we've talked about, we're effectively launching an insurance product, which is for buyers who want to be able to return items they're going to be buying an insurance like ability to return, and we can price that based on what we're seeing in the market. And I think for sellers over time, while I don't anticipate us charging fees to sellers, I do actually think we're going to build a lot of tooling that will help make their lives easier in selling items. And I think if those tools are high quality, you can imagine sellers subscribing to a suite of tools to improve their listing, their merchandising, all the things that make the seller process robust. So I see many, many ways that we can monetize this stream of buying and selling on ThredUp. And the market is so large, I think there'll be lots of ways to do it.

And your next question comes from the line of Oliver Chen from TD Cowen.

Nice job. Congrats. So on the revenue beat in Q3, what was driven by repeat versus new customer acquisition? And as we look ahead to 2026, how are you thinking about how those drivers interplay into your guidance? Also on the peer-to-peer model, which is really interesting and obviously very important. How would you compare and contrast on Poshmark or Mercari? I know it's been a difficult market in terms of profitability and fees. And it sounds like you're balancing control and scalability versus curation and fraud. And third question, Generative AI is something you've been very, very good at. On the peer-to-peer model, what role will that play there? And then we're doing a lot of work around OpenAI. Your thoughts on Agentic and the evolution in terms of brands and long tail, just different characteristics of Agentic and conversational commerce continues to be an important growing traffic consideration.

Thanks, Oliver. You got a lot in there. So if I miss anything, you let me know. I mean, I think on the Q -- on the revenue beat, consistently, existing buyers are still the driving force. Historically, it's 80% of our revenue is coming from existing buyers, and it hasn't varied that much to date. It's still -- while we're acquiring lots of new customers, the bulk of buyers on ThredUp are existing customers. And so you're even seeing the new customers we acquired over the past couple of quarters becoming repeat customers at higher rates in Q3. On the peer-to-peer piece, you asked on the competitive set. Look, I mean, frankly, there's been very little innovation and product work done by a bunch of these other peer-to-peer platforms. I think they've lost the plot a little. And so I actually think we can build something that's far superior to what's out there today. And so I see huge opportunity to build something that sellers and buyers love. I think it's clear there's a market out there. And the question is how to serve that market really effectively, and I'm confident we can do that. On the GenAI piece for peer-to-peer, yes, a lot of the tooling that we've built over the past year is being used to deliver a superior direct selling experience. That's everything from listing photography, curation, merchandising to how we price, to how we display to buyers. So I don't think we could have done this over a year ago, right? So a lot of this has been in development based on technology shifts in the market that I think have been pretty profound. Just a nice segue to your last question, which is on OpenAI and Agentic commerce. I think it's a big part of what's coming. I'm convinced that agents is going to be a part of how people shop in the future, but I can't tell you when. So I'm quite confident that we're in the middle of the change, but I'm not sure of the timing. And because I think shopping is a little different -- shopping for fashion is a little different than buying peanut butter. And so I think the peanut butter use case is a little bit more well defined. I think in fashion, it's going to take a little bit more time. But we're staying close to all of the large players in the space. Anybody who's building sort of customer-facing chat clients, we are in conversations with. Today, if you go to OpenAI and talk and ask ChatGPT about selling used clothes, we're at the top of that list. And so we're going to keep investing to make sure that, that stays true.

And what -- on the peer-to-peer angle, James, what will be your competitive advantages? It sounds like customer engagement in the existing sellers. But what would you say? Because it's been a race to price in that marketplace, but it does sound like there's new technology now and you've investigated P2P for a long time. And then, Sean, as we think about CapEx in the forward years, is there anything we should know about like in terms of how that interplays with some of these new endeavors?

Yes. On the selling piece, I think if you look at the market for sellers, really, it sort of has lended itself to this professional seller network, which has crowded out your casual seller. And I actually think the most interesting part of the market is the long-tail casual seller. And so right now, the incentives for some of these platforms because the product experience isn't great, is to just get flooded with stuff. And so I think our approach is a much more measured, curated experience in direct selling that I think will benefit sellers by driving liquidity and sell-through for them and also delighting buyers with a better experience. I think for buyers, in particular, returns is a huge piece of friction in this market. Trust is a big piece of friction in this market. And I think we are delivering something that, I think, a far better experience, both on the trust and safety side as well as the ability to do returns and remove that big piece of question -- that big question mark among buyers of like, can I trust what I'm going to get, who stands behind it. So I think there's actually big advantages we're bringing to the market, and I'm excited to kind of keep going.

And Oliver, on the CapEx, like I said on the call, we'll do $10 million about this year, and that will be consistent for 2026. And then once we get to 2027, we're probably at the point where we're filling in the Dallas DC. So we'll give you guys more of a view there, but I'd expect it to be more than the $10 million, but we'll give you information as we go along there.

And finally, James, we've talked about AI together a lot. As you think about like first-party data as well as LLMs and partnerships with different LLMs, like how do you see that evolving in terms of your competitive mode in AI and how AI has a lot of open source. However, the proprietary tools that you develop are quite necessary since a lot of this is also not very generalizable.

Yes. I mean I think that we are benefiting from being a technology company and an infrastructure company at heart. I think all the tooling we built I think, has allowed us to move faster and stay ahead. And at the end of the day, I think Amazon has proved this out time and again that having the right products and being able to deliver them to customers is of utmost importance. And so in secondhand, I think we've got an incredible product selection across our DCs that's improving every day. And I think then when you add in direct selling, you just are compounding that supply advantage. And in resale, supply is the name of the game. And I think we're continuing to distance ourselves from others and having the best supply out there.

And there are no further questions at this time. I will now hand the call back to James Reinhart for any closing remarks.

Well, thank you all for joining our call today. We set out on a mission to inspire the next generation to think secondhand first. And I think this year's results so far are just beginning to show what's possible in the years ahead. It's such an incredible time for ThredUp right now. I want to thank all the teammates for being a part of this journey and look forward to sharing further progress next quarter. Thanks.