RealReal Inc (REAL) 2021 Q2 法說會逐字稿

Good day, and thank you for standing by, and welcome to The RealReal Second Quarter 2021 Financial Results Conference Call. (Operator Instructions)

I would now hand the call over to your speaker today, Mr. Paul Bieber, Head of Investor Relations. Please go ahead.

Thank you. Good afternoon, and welcome to The RealReal's Earnings Call for the quarter ended June 30, 2021. I'm Paul Bieber, Head of Investor Relations at The RealReal. Joining me today to discuss our results are Founder and CEO, Julie Wainwright; and Chief Financial Officer, Matt Gustke. Hopefully, you've had a chance to read our press release and stockholder letter that we distributed earlier today, both of which are available on our Investor Relations website.

Before we begin, I'd like to remind you that we will make forward-looking statements during the course of this call. These forward-looking statements involve known and unknown risks and uncertainties, and our actual results could differ materially. You can find more information about these risks, uncertainties and other factors that could affect our operating results in our most recent periodic report on Form 10-K, subsequent quarterly reports on Form 10-Q and in our earnings release from earlier today. In addition, our presentation will include certain non-GAAP financial measures, for which we have provided reconciliations to the most comparable GAAP measures in our earnings press release.

With that, I'll hand the call over to Julie for introductory remarks, and then we'll go straight to Q&A. Julie?

Thanks, Paul. And thank you all for joining us to discuss our second quarter results. We are pleased to report another quarter of strong growth driven by at-home consignment and our continued retail momentum. We achieved our highest numbers of both new and repeat consigners this quarter. As a supply-driven marketplace, this has resulted in Q2 GMV increasing 91% year-on-year and 53% compared to the same period in 2019. Q2 year-on-year GMV growth also accelerated quarter-on-quarter when compared to both 2020 and 2019. We achieved these strong growth rates while also driving a significant improvement in gross profit per order, a key driver in our path to profitability. Q2 gross profit per order was approximately $94, a $9 quarter-on-quarter improvement.

Q2 was a very busy quarter for us, and we made significant progress with our top priorities. Specifically, getting back at home. Our at-home consignments are increasing as a percent of total consignments and contributed 38% of total units in June.

Our Arizona facility. We accelerated the move to our large authentication center in Phoenix to accommodate future growth. And I'm happy to say that's going very well.

Our neighborhood store expansion. We added Austin, Dallas and Atlanta during Q2. Our retail stores generated 30% of new consignments in the quarter. We plan to open about 2 more stores this year.

And lastly, our technology innovation. Our investments in technology continue to differentiate our business to drive efficiencies in our operations and enable significant future scale. In Q2, we released the next generation of our authentication and pricing engine.

The current trends in our business are strong. We believe they will continue this year and next. Beyond our GMV growth, we made progress with gross profit per order and efficiencies in operations and marketing. All of these elements are key to our path to profitability. The investments we have made in neighborhood stores and our Arizona facility not only support our growth but also create the potential for meaningful leverage going forward. We are focused on achieving profitability and expect to make significant progress over the coming quarters.

As always, I'd like to thank the entire TRR team for their hard work in delivering these strong Q2 results. It is their dedication and commitment that drives our business every day.

And with that, operator, we are ready for questions.

(Operator Instructions) Your first question is from Oliver Chen of Cowen.

On profitability and the opportunity ahead, what do you see as the key drivers in managing that gross profit per order? And also just love your thought on the evolution of L.A. and New York in these markets in terms of supply growth and capabilities.

I'm going to start with the last part, and then I'm going to turn it over to Matt to discuss the key drivers of path to profitability. So interestingly enough, both L.A. and New York, our older stores, are -- continue both in driving new consignors and new buyers and are performing phenomenally well. We are expanding the New York store below. We're going to be reconfiguring the part of the bottom of the New York store to allow us to take an even more consignment there.

But we're -- the stores are a little bit lower to, I mean, especially SoHo. And they're doing -- well, they're achieving plan or better than plan for us. So we are committed there. Our flagship stores, obviously, we are expanding not the flagship stores. We're expanding the neighborhood stores, and those are also doing as well as we thought or better.

Okay. And I'll go to the first question. So I think your focus was really on gross profit per order drivers and within the context of overall profitability, so I'll emphasize that and I'm going to kind of cover the overall framework for how we're thinking about profitability. First of all, we're very focused on it. As you saw, gross profit per order increased $9 quarter-over-quarter, getting into the mid-90s.

From here forward, further improvements can be expected from a full quarter benefit from lower buyer incentives, which exited Q2 at pre-COVID levels. And then beyond that, we do expect over time to see incremental shipping improvements and continuing benefits from a high AOV. So those are really the key pieces that kind of get you from where we are now to a $100 neighborhood and going forward just small incremental benefits from there.

Then the broader context of profitability, that's just one element. So first and foremost is top line growth. GMV is growing well. In July, I think you saw we put a release out that we're still growing 53% versus 2019, which the comps actually are getting more difficult in Q3 as we're lapping our post-IPO quarter. We're comfortable saying that we can sustain 30-plus percent top line growth for the foreseeable future. That combined with gross profit per order increasing to the $100 neighborhood and getting variable expense efficiencies, which we have been doing consistently over time and expect to continue to do so, get us to the point where we're seeing contribution margin per order in the $35 to $40 range.

And then what remains is the controlling of fixed costs. And our fixed costs from this point forward are really not going to increase very much at all as it relates to profitability. Arizona was the last big step up. We've got 2 small stores to come. And after that, you're going to see very minimal fixed cost increases going forward. So the leverage from the top line compounded by the gross profit piece and variable marketing efficiencies and other variable expense efficiencies will carry us to the baseline.

That was very helpful. Last question on the 30% annual GMV growth that you called out in the letter. What are some underlying drivers that give you confidence there? And any further details would be helpful.

Yes. Thanks, Oliver. So I'll start. Maybe you might want to chime in on this. But overall, we're feeling very optimistic about where we are not only coming out of COVID but just overall. Keep in mind that where we are now is just a rounding error in terms of penetration to the overall TAM with significant tailwinds in the resale market overall. We're very well positioned to continue growing well and taking our unfair share of market growth going forward.

On top of that, I think we've derisked some of those assumptions over the course of COVID and still now with more diversified supply coming from more places that just gives us incremental confidence that committing to something like that is reasonable at this point.

And of course, we are always looking at our cohorts and seeing if there's any change in the cohorts. And in fact, we are in very good shape with our assumptions for repeat versus new and also on the consignors and buyers side. So consequently, we do feel confident about next year. And we do recognize we're -- this year and next year, we do recognize the Delta variant is the wildcard. But assuming there are no complete shutdowns again, which we don't foresee, we feel good about our future here.

Your next question is from Erinn Murphy of Piper Sandler.

I've got 2 for Julie and just a quick clarification for Matt. Julie, on the next generation of authentication and pricing capabilities that you're adding, can you share a little bit more about what that should permit you to do over time? And how are you measuring returns there? And then if you could share on the second quarter what you saw in apparel and footwear trends. And then I've got just one clarification for Matt.

Sure. So first of all, I'm going to start with the last. Because apparel is actually up again, so we're very excited about that. In fact, it's actually exceeding our overall growth. So apparel is coming back. And footwear is still not as high, but it's still doing well. And actually, there is -- it's no longer a drag on the business. So we're very excited about that. We're seeing similar trends in July, so this is all good news.

Now when it comes to -- in fact, I can even give you the exact numbers, but someone just [whispered to] me. They are up. Apparel was -- ready-to-wear was up 70% versus year-on-year and footwear was also up roughly 70% versus a year ago -- same period a year ago. So that's really good news and bodes wells for the balance of the year.

In terms of using technology, we use a lot of technology. We've been using it for pricing. And when we talk about pricing optimization, our goal is to get the absolute highest price without a fall-off in velocity of sales. And we've seen, in general, over a $10 a unit price increase by using our -- across the board by using both machine learning and computer vision to help us and then with human oversight. So that continues to show improvement. And obviously, the consignor wins if the prices are higher and certainly to get our fair share of that. So we're excited about the progress there, and that's how we measure that.

For authentication, it's really 2 levels. And first of all, it actually allows us to change the workflow for our team. So we measure both on effectiveness of keeping pace of the market and efficiency of that team on processing units. And it's -- our technology solutions continue to improve. We still are a human-driven business, especially on that side, but it actually is making us more efficient and even more effective.

And then the other thing is we are going to arrange an ops tour, depending on the Delta variant, sometime this year in our new Phoenix facility. So you can -- anyone who comes will be able to see first hand what we're doing, and we can even talk about the changes visually then because I think you'll see a huge impact.

Got it. And then, Matt, my clarification for you, just on the gross profit per order. You talked in the shareholder letter about obviously getting to that 100-plus level. Should we take the language around that to be over the next 18 months, so the exit rate next year? Or is that still potentially in the cards for this year? Just trying to understand the way it was laid out in the shareholder letter is kind of this 18-month guide.

Sure. Yes. We're trying to put very specific time frames in any of this, but by no means are we walking back from anything we've said previously. So I do see the potential to approach the $100 kind of milestone at the end of this year and certainly with comfort kind of getting down on a full year basis next year.

Your next question is from Michael Binetti of Credit Suisse.

Matt, can you connect a few comments on the variable expenses here. I think you noted the variable expenses at the end of the shareholder letter were $79 per order in 2020, and we can see it move around in 2018 and 2019. But maybe you could walk us to where you see the apples-to-apples equivalent of that number this year. And what are the inputs that get you there? And then maybe rank order the inputs for next year as you push towards the -- I think the apples-to-apples number will be $60 to $65 per order to get to EBITDA profitability from the $79 in 2020. I'm just trying to understand the gap there since we've been over the gross profit per order quite a bit.

Sure, sure. I think I tracked all of the parts of your question. So let's start with what the definition, first of all, of variable expenses are. So they include -- the 3 big ones are the cost of our marketing, the cost of our variable operations, so that's both inbound and pick, pack and ship, and the cost of the sales team. And then sort of second tier is our retail, our variable retail operating expenses. To get from here to there, I think 2019 is kind of the most recent good kind of benchmark to work from. So we need to see all of those costs in aggregate improve by about 10% to 15% on a per unit or per order basis to get to the profitability milestone.

The biggest drivers are going to be the first ones I mentioned. That's marketing and our operations variable unit expenses. Market efficiency, we have a long-standing track record of driving marketing efficiency. That starts with really strong cohorts and really strong buyer engagement and retention. I see no reason to think that, that won't continue. And then on top of that, we compound that with improving buyer acquisition costs over time.

On the variable operations side, that's really the product of trying to really leverage the investments that we've made in automation and we continue to make. So we are seeing those benefits now. We expect to continue seeing them going forward. And then as you know, we're kind of -- we're just about opening operations retail stores for the time being. So that, too, will start to generate leverage going forward. So that all adds up to 15%.

I guess to follow that, where do you see the customer acquisition costs going by '22 if we should think about it in those terms? And then one other follow-up on the -- I think you said the direct gross margin was down about 570 basis points driven by the sale of aged owned -- of aged inventory, lower margins. Can you just help us think about what's embedded as we think about 3Q or second half, please?

Yes, sure. Let me start with that and I'll remember your first question. So the direct margin, you're right, was down 4, 5 percentage points on a year-over-year basis. And that's really a function of us deemphasizing the purposeful purchase of inventory as our supply channels across the board have rebounded strongly. So we just don't utilize that as much. So the higher mix of the direct business is coming from the late returns, which is the lowest margin piece. That's important to dimensionalize this thing.

So the direct piece of the business is less than 10% of overall GMV. So small changes on a small base have a pretty significant impact on the surface. Overall, gross profit per order is the metric to look at. That increased $9 quarter-over-quarter, and we continue to see the opportunity for that to go up. What was the other question?

Marketing back.

Marketing back. Yes, I'm not going to provide any long-term guidance on where we see that heading other than to say that we have historically been very nimble and have been very effective at driving back improvements just about every year except for 2020. And I'm confident we can continue being efficient.

Your next question is from Mark Altschwager of Baird.

So another follow-up just on the path to profitability, but maybe from a little bit of different angle here. So I appreciate that the backdrop makes the forecasting difficult. But as we kind of put the pieces together that you are kind of guiding to, GMV to [break] this year, expectations for at least 30% next year. It would seem that you're going to be knocking on the door of about $2 billion GMV run rate by kind of later part of next year. Gross profit per order has bounced back nicely, as you outlined, and there's some room for that to improve. It sounds like you're pleased with some of the efficiency initiatives, especially Arizona.

So I guess as I put that together, I guess, which components on the path to kind of bear the most risk here? Because it seems like you've given us all of the components that are kind of moving in the direction. But you kind of call out not wanting to put a time line on some of these things at this point. So just any further clarity there would be great.

I can start. So we would love -- obviously, I think you're hearing loud and clear our commitments to simultaneously continuing to drive top line growth and driving to profitability. We'd love to be definitive about the time frame that we're going to get there. But the current environment we're in, it just doesn't seem like the prudent thing to do. But we are putting out monthly disclosures, which are a pretty good proxy for providing short-term guidance. The incremental benefit in the short term is certainly negligible. So that's how we're going to approach it and provide as much transparent -- more transparency, not less, on our path.

Fair enough. And just a quick follow-up, Matt, on just the inventory line. It looks like the direct gross margin was weighed down a bit by some working through of aged inventory. Just any more context there. And I think you're guiding to some kind of further inventory build through the remainder of the year. So just any thoughts on how you're kind of managing the opportunities with the direct side with some potential margin risks.

Yes. So let me start with that last part is not what we're saying. So inventory at the end of Q2 was about $60 million. That should be the apex of our inventory balance, at least for the balance of this year. We're not doing that with nearly the emphasis that we were a couple of quarters ago. So that's point number one.

And point number two, we -- the aged inventory part of the comment means older supply that we have from late returns. And that stayed with us over a period of time that we do end up bringing the price down to sell through ultimately. It is not new. That is forever -- that's been the pattern forever. And the late return part of inventory has always been the lowest-margin piece of the direct line and the overall business. So there's nothing really new to comment on other than inventory is not going up. That includes this quarter, we don't expect to see it going up.

Your next question is from Edward Yruma of KeyBanc.

I guess first, any more metrics you can give? I know it's still early days on customer acquisition costs through neighborhood stores. And I know you guys indicated you were going to kind of pause expansion once you've completed this last. So any particular hurdles you're hoping to meet?

And then as a follow-up, we noted that you guys seem to be testing other categories like electronics and sporting goods. Kind of any sense as to how we should think about those tests thus far? And kind of how are you obtaining the initial set of inventory?

All right. So it's Julie. So the interesting thing -- I'm just going to just start with the category plans. We're pretty excited about it. It's early, early days. So we're not going to make any predictions, but we are going into the collectibles in particular. There were large categories in the outdoor. And we are -- we think of ourselves as a luxury lifestyle business. Our consignors were asking us to get into these categories. We researched it for a while. It felt like it was actually going to be a net add to us.

So we just started that in almost the third week of July. So it's very, very new for us. But I would expect it to be -- it expands our TAM. It actually expands our service level. So we feel really great about that. And then do you want to talk about the first part, Matt?

Yes, sure. So to add on to that as well. So I think you also asked them like, how does that change how we get our supply. Not really at all. It leverages our existing sales infrastructure and our existing authentication infrastructure and our retail footprint.

And our retail footprint.

So it's just making better utilization out of those things.

And then the retail stores.

Retail stores, yes. So there's kind of a lot in there. So we're not disclosing specific numbers in terms of store acquisition costs. We have said, and it remains the case, that the acquisition of new consignors through retail is more efficient than through our marketing efforts alone. That remains the case. It's a very surgical effective way to acquire new consignors. It's frankly the tool that we didn't really have previously.

And you're right, we're about to pause. We're doing exactly what we said we were going to do, get to around 10 stores and then give them some time to mature. We're going to learn some things. We're going to optimize them, and we'll come back and reassess what we do going forward. Keep in mind that the majority of these stores have been open, not even at all, I guess, there's a couple of more to go, or less than 3 months. So it's going to take a little bit of time to gather data and come back.

But we've talked in the past that early signs from the stores are quite good. So of course, we're looking at overall value supply -- quantity and value of supply coming in, number of new consignors, the -- of course, the demand that's generated in the stores and then most importantly, the impact of those stores on the market in which they operate. And that can be a small or a large halo depending on the size of the store and location.

But we've seen consistently a halo effect where the growth rate in that market accelerates with the opening of a store and then stays at that higher level and then continues to grow for a longer period of time. So it acts as an accelerant on a market-by-market basis. So we want to see that play out across the entire portfolio before committing to how many were going forward.

Your next question is from Michael McGovern of Bank of America.

I just wanted to ask about the metrics around the retail stores driving 30% of new consignors and then at-home concierge appointments generating 38% of total units in June. Are you seeing that trend continue to improve in July as we've seen COVID cases kind of tick up a little bit nationwide?

And then also, secondly, I just wanted to ask about any other specific cost trends to call out for Q3. I think you gave some high-level framework. Is there anything specific to call out for Q3 that might cause some quarter-on-quarter? Anything to point out?

So in terms of what are we seeing in July, actually, we're seeing an increase in the number of units coming from our in-home experience, not a decrease. We, in fact, have not felt at this point any impact at all from the Delta variant. And I think it's also because we do tend to -- most of our business is urban-centric, and most of the urban areas also have a higher vaccination rate, I think, nationally. So we are an urban-driven business. And getting back in home, people are excited to have us come back in. And we're seeing increasing units, and that continues in August. For the second part...

Yes. The second part was around OpEx. So we do expect to see some OpEx increases on a quarter-over-quarter basis. And that's going to come from -- the general theme is investing ahead of anticipated Q4 volumes and growth. So where you're going to see growth is in our sales team, in our operations to support higher volumes of product coming in and higher volumes of product going out, to some extent with our marketing as we kind of invest into a strong seasonal period. And then in Q3, specifically, we still have redundant or duplicative expenses for our California facility, which is basically shut down at this point, and the cost will be rolling off as we exit the quarter. So that will be clean kind of coming into Q4. So you didn't ask but I'll give you, so Q4 OpEx, given all those different dynamics, should be pretty close to flat sequentially versus Q3.

Your next question is from Lauren Schenk of Morgan Stanley.

Great. I just wanted to ask about third quarter and the fact that you didn't give a GMV guide that you typically did. So just curious if that 30%-plus comment in the shareholder letter should sort of be extrapolated as the general guide for the third quarter or if there's something else that you're seeing there.

Okay. No, I wouldn't -- that's not the interpretation to make. But we replaced giving short-term guidance with providing even more frequent disclosures of actual results on a monthly basis. So the July results are out. You're going to continue seeing those disclosures every month for the balance of the year. So the incremental benefit in the short term of providing guidance, we think, is sort of de minimis.

In terms of what we're seeing, as you saw from July, the year-on-year comp did not decelerate but is the beginning of more difficult comps in 2019. Q3 2019 was one of our strongest quarters ever coming off of our IPO. August was the most difficult comp in that quarter, and September was pretty much up there as well. But we're expecting to see continuing strong growth for the balance of this year, no doubt in excess of that 30% over the balance of this quarter.

Yes. And just in case you missed the press release for July, we did grow 53% versus same period in 2019. So that's 53%, not 30%.

Right. But there's -- I guess the reason for not giving the guidance is just sort of you're just going to give short -- you're going to give monthly updates rather than quarterly guidance going forward. Is that the conclusion?

At least through the balance of the year, that's what we committed just because our business was so impacted last year by COVID. We just thought it would be easier for us -- easier for you to follow the business and its recovery if we give actuals every single month on the top line.

Next question is from Ike Boruchow of Wells Fargo.

I guess, Matt, I did want to dig into the direct business a little bit more. The consignment gross margin looks great. But I guess I'm trying to understand what -- maybe -- what exactly happened this quarter on direct margin. And I bring that up because about 3 months ago, you had stated that you assumed that the margin would sequentially improve and there would be more purposeful direct revenue buying inventory upfront, which gives you better margin. But now it sounds like you're saying you have less of that, and the gross margin decelerated from Q1.

So I guess I'm just trying to understand. Did something strategically happen in the second quarter? And then is there a way we should think about the margins in that revenue base? Just because they used to be in the 20s and now they're closer to 10, I think we're just having a little bit of trouble understanding how we should think about it.

Sure. So yes, something has changed since last quarter, and that is that our in-home or back-in-home and our traditional business of sourcing supply on a consignment basis is growing very well. That takes pressure off to buy inventory to sustain our recovery and our growth. So we made the decision that that's not -- in isolation, that's not how we prefer to use our capital to purchase inventory. So we're just taking our foot off the gas on that for the time being. So that's really what you're seeing there.

But I think we can kind of get lost in some of these -- the distinctions between the different parts of the business. If you just look at overall the consignment business and gross profit per order, that's trending well. The margin in the direct business is just -- like the overall size of the business is too small to really drive things. Over time, that should keep going up because there are, in both parts of the business, elements that are not purely variable. They are semi-fixed, so they get some leverage as we grow. I don't guide to a specific percent, but we'll be very transparent if that part of the business is having an impact on our gross profit per order trajectory, and it's not.

And part of that -- sorry, part of the follow-up is what we're seeing sell through there on the inventory side from a vendor perspective, it's very high-value things. There are lots of high-value watches and handbags. Those carry inherently structurally lower take rates. So that -- you're seeing -- watches are a very strongly growing category recently. So part of that is just systemic in terms of our take rate structure. So that's going to bounce around on a small base of GMV going forward.

Your next question comes from the line of Susan Anderson.

It's Alec Legg on for Susan. Just a bigger overall picture question on the path to profitability and assuming that 30% annual GMV growth. When would you likely need to make additional investments such as opening a new facility or moving into a bigger one? And then what other type of investments do you think you would need to make to maintain that growth?

That's a great question and with a pretty simple answer. We will be well past the profitability milestone at the time we need a new authentication center. With Arizona, we have about 5 years of total growth capacity in front of us. And that obviously -- our warehouse network is a big part of our fixed cost base. But more abstractly, fixed cost overall going forward should not grow very fast from the levels that they're at currently or will exit this year at, I should say. So those fixed costs include our overhead functions, all of our real estate and completing our store portfolio, so that should be a very modest growth rate going forward.

Your next question comes from the line of Anna Andreeva.

Congrats, guys. One question for Julie, I guess, and a follow-up for Matt. So to Julie, the number of new buyers increased nicely again this quarter. Can you maybe talk about the behavior of some of these buyers compared to the previous cohorts? Just anything to call out that's different demographically or regionally.

And Matt, I'm not sure if I missed this. I think you talked about $10 million in the transient costs from the Arizona DC for this year. What should we expect for the third quarter? And just remind us, when should we expect these costs to roll off?

Okay. Should I start with the second one?

Yes.

Okay. Yes, you're right. About $10 million for the full year in nonrecurring costs, that's made up of COVID expenses and Arizona inefficiency. So both overlapping rent as well as duplicative labor costs. The overlapping rent will be done by the end -- exiting this quarter. The duplicative labor also will be done as we exit this quarter. So what's left is COVID. And I wish I could say when those costs are going to roll off. But the COVID costs are $1 million and change per quarter going forward until they're not.

Yes. And in terms of the buyers, we're actually seeing -- it's getting -- our base is getting younger and a little bit more male but not significantly. So we just -- Gen Xs are a little bit more engaged with our -- if you go millennial plus Gen X, it is the majority of our buyer base now. And our -- and then we are getting more males. But having said that, their lifetime value is approaching -- as far as we can tell, the earliest one is approaching pre-COVID level. So on some level, it hasn't changed at all. On another level, it's getting younger, which we like. We do like -- getting younger and more male is also good for us.

Your next question comes from the line of Marvin Fong.

Great. Two for me. Most have been asked, but just wanted to follow up on what [Andreeva] is bringing up that the new buyer growth was very strong. If I look back in history, I think you guys were doing even better numbers like 140,000 or better new customers a quarter. And I noted, Matt, that you said buyer incentives are back to pre-COVID levels. So just wanted to hear your thoughts on, for our models, should we think about your active buyer growth staying about this level where you did in the second quarter? Or could we actually kind of reach those kind of 140,000, 150,000 levels?

And then second question, and just apologies if you addressed this elsewhere, but the AOV for July was down a little bit. Just thought -- just interested in your comments on what drove that. Is it some mix of apparel going up? And just the total AOV trend there, that would be great.

I'll address the latter and then kick it over to Matt. So the AOV was -- it still was a record high for the month of July for us. But it did trend down versus June and May just because we have more apparels and shoes mixed in. But it still was a record high AOV for July, which tends to be -- before COVID hit, tends to be one of our lowest AOV months just because people are buying more apparel and the apparel is less expensive. It's more contemporary in that month. So we did see mix down a little, but it's still at record high levels. And then, Matt?

Yes. I think your question broadly was around buyer growth generally. I think it's a basic function of the metric of active buyers being a trailing 12 months. So you're going to -- you saw that a lagging indicator both on the deceleration during COVID and the reacceleration now that we're kind of pulling out of COVID. I think the most useful use of that metric is to look at GMV per active buyer on a trailing 12-month basis. As you can see, this period, we're up to about $1,700 -- just shy of $1,700, which is approaching where we were pre-COVID.

So overall, the buyer ecosystem, the folks that we have are very engaged. It's very, very healthy. Over time, I don't think our new buyer growth -- active buyer growth should basically track the business. New buyer growth doesn't necessarily need to because we do tend to see a slightly increasing contribution from our base of buyers every year.

Your last question will be coming from Simeon Siegel.

Did you -- any way to quantify how much of the ASP increase was due to mix versus, I think, in the shareholder letter, you mentioned you're already seeing benefits on ASP from the next-gen pricing engine. So I would love to hear about that and then how you're thinking about that as a lever moving forward.

So yes, we've actually got $10 more, and so we're excited about that overall in the business with our pricing optimization. And here's how we literally -- and this is a key area of investment for us because, obviously, you can raise prices. We said it almost -- it's all great. Now having said that, the last thing you want to see is velocity of sales dropping. So it is an iterative process that's ongoing, and we do expect it to continue to see benefit. Sometimes it's only $5. But like I said, overall, it jumped to $10 this year. And hopefully, it will continue to -- it will continue to yield some benefits.

But every dollar is important to us because it does drop -- a portion of it drops right to the bottom line. And it also enhances consignor satisfaction. So those things -- those benefits are really positive for us. And Matt?

Yes. And I think you basically covered it. So I think we've covered in the -- I think it's in the stockholder letter. So ASP was up 17% year-on-year in Q2. And Julie mentioned earlier in the call that in July, women's ready-to-wear and women's shoes were up 70% year-over-year. So starting to mix up in the business. So that implies that the like-for-like prices are up, as Julie was mentioning. Also, units per transaction are very high. So we're seeing a continuance of high-value purchases. The strength there is sustaining. But the items per order has come up to pre-COVID levels as well. So that's the uplift of that -- or the record high AOV that we've been seeing for a while now.

Great. And then, Matt, did you comment at all in terms of the return cancellations, maybe what you're expecting there, the -- if there's an impact from mix and as apparel grows?

No, we didn't comment on it. But typically, what -- it should be pretty stable. I think the abnormally low return rates during COVID have largely normalized at this point, but they're still a bit lower, and that's just a function of category mix. So we don't frankly know exactly what category mix is going to trend over a multi-quarter basis. So they are going to travel together, return rates and AOV, frankly. But then Q4 typically is a slightly higher return rate within the context of the year. But this quarter should be pretty consistent that what we saw in Q2.

There are no questions at this time. We will now transfer it back to Ms. Julie Wainwright.

So thank you for joining our call today. We appreciate your time. We appreciate your questions. And with that, we've got some work to do, and I'm sure you do, too. So have a great week, and we'll talk to you on Q3 results. Thanks. Bye.

That concludes today's conference call. You may now all disconnect. Have a great day.