The RealReal, Inc. (REAL) Earnings Call Transcript & Summary

Great. And here we are at the next, one of our slots here for a fireside chat. Thanks, everyone, for tuning in. It's my pleasure to host the team from The RealReal for a fireside chat about everything that's going on in their business right now. We have Matt Gustke, the CFO. Paul Bieber, I think, is also here from Investor Relations. Matt, Paul, hope you're both well, and thank you again for attending the UBS Global TMT Conference for the second year in a row. And hopefully, next year, we'll have it live again.

Yes, hopefully. Yes, thanks for having us. And good to see you.

Thanks, Eric.

Great to see you guys as well. So I think what I always like to do is for those who don't know the business as well, Matt, maybe you can just set the table, what is the business? What is the unlock that Julie and the team are trying to solve for over the medium- to long-term? And how are you seeing the company evolve?

Sure. So The RealReal started in 2011 by our founder and CEO, Julie Wainwright. On the aha moment of seeing a close girlfriend of hers go into the back of a high-end boutique and shop exclusively stuff that was on consignment. So Julie's insight was like "geez, there should be unlimited demand for this type of product, high quality, authentic, good condition, luxury brands and exceptional value." But the problem was there was no good service model to bring that to people at scale. The incumbent model was the consignment store, and there are a lot of inherent sort of disadvantages to that. And then otherwise, you're kind of in this world of unregulated sort of peer-to-peer marketplaces, where there is a trust deficit for buyers. So from day 1, the whole premise of the business was established trust absolutely for buyers by building an authentication infrastructure and then do all the work on behalf of consignors to take friction out of the process and get more supply unlocked. It's not -- in a lot of ways, it's kind of what -- and I've spent pretty much my entire career inside of marketplaces and I find the most interesting ones are those that obviously are selling something that people want, but that are not just disrupting the business that they're playing in, but growing the market, not -- ridesharing is an obvious one where it's much more prevalent than the industry that it replaced -- or is replacing. So here we are kind of almost 10 years later, we're clearly, the leader in the space of authenticated luxury resale. Still pretty much a U.S.-only business in terms of supply acquisition. We do ship a bit internationally. But the addressable market for what we're talking about here, the trapped value in people's homes in the U.S. alone in the categories that we participate in today is about $200 billion. So we were about $1 billion of that last year, probably going to be about the same this year. So very, very early innings of developing into that market and really developing -- penetrating into that TAM.

That's great, Matt. And I do want to touch upon the year that's almost behind us. It's obviously been a year with a lot of volatility, a lot of curveballs. And it's interesting. When I talk to a lot of investors about The RealReal, I think they are beginning to understand that luxury is in the process of moving online. But then there's still a fair bit of confusion about how you source supply, and that might have been volatile this year in the environment we found ourselves in with the global pandemic. So give us a little bit of a step back on how this year transpired? What you learned? And sort of how it sets the company up for going forward?

Well, I don't think I'm alone in wishing this year to be behind us.

I don't know anybody who's holding on to 2020.

Yes, fair enough. Well, what COVID showed us is that our business is not immune to global pandemic. Who would have thought. So here's what happened in the early days of COVID-19, most people are probably familiar with the story, so I'll kind of speak through it. The vast majority of our supply was disrupted overnight. Half of our supply came from us going into people's homes, one-by-one individual consignors, white glove service. Another 20% came from our retail stores. So [indiscernible] that's gone. On top of that, then a bunch of restrictions were imposed on us by kind of local governments that limited our ability to operate. So even if we could supply, we had nobody to process it. So we first worked through that problem and put in place a bunch of health and safety programs, worked with governments, got open again. Then that was the precursor to seeing stability in the business in the beginning of a recovery. So I'm going to kind of go through then what do we do to stabilize the business, start to see recovery? What's in our control at this point? What's still not largely? And then what do we do about that? So in our control was pivoting from in-home consignments to something else. So we pivoted to virtual, and that largely worked pretty well. I mean it's -- the analogy I'm using is kind of like -- it's kind of like we're playing goalie. So we're not catching every puck because not every person who had consigned in a white glove environment is moving to virtual, but it is a suitable replacement, and we're getting a lot of supply that way. So we're still learning our way through it. Still not sort of perfect in how we execute that part of our supply acquisition, but it's getting better, and it's showing it's got a permanent role for us. So in the past, you had in-home or we send you a shipping label. And it was like the distinction of what channel was appropriate for you as a consignor was pretty arbitrary. It's just based on whether we thought it made sense since the consignor really wanted one way or the other. I think going forward, virtual is going to fill out the spectrum, it's going to kind of sit in that sweet spot in the middle, it's going to so be the best channel for perhaps the majority of consignors, where there's still a human element, a high-touch element, but is more productive and efficient. And a lot of people -- better for a lot of people. They don't have to have somebody come to their house and whatnot. So I think there's a permanent place there for that. So that's good. Then there's retail. So once we opened retail after a couple of starts and stops, but by July, we were largely open in retail, and it started building. And what it shows, people did want to drop things off in a brick-and-mortar environment. But it took us a while to realize that we needed to tell people about it. So in October, we started messaging it and promoting, come to the store, drop off consignment. We had our best month ever in October in our stores. Our supply grew 48% year-over-year. And right now, flat to new growth. So growth is incredible. So we're very encouraged to see that. And that continued in November even once we turned the promotion off, so the supply grew 51% in the stores. So behind the scenes, we've been doing a lot of analysis on what the role of retail is for us and what is the true incrementality of a brick-and-mortar presence. One could argue that in a business like ours, where everything we bring in sells that demand incrementality is kind of a tough sell. There are some good things that happen on the demand side, higher selling prices, higher velocity, lower return rates, stuff like that. So there's goodness. But what is becoming increasingly clear is supply is truly incremental. About half of the consignors who come into the store to drop off are new to us. And when we look at our entire portfolio of stores, what appears to be the case is that consignor acquisition is both more cost effective and faster. Immediately after opening a store, we see an acceleration in consignor growth in that market. And that goes for a while and it doesn't peak and come back down. It just goes to that higher level and then continues to grow. And then as we further parsed our retail data, one stood out, and it was the -- our Madison Avenue store. It's our small footprint stores. So most of our stores are like 8,000 to 10,000 square feet. This one is like 2,500 square feet, but it brings almost as much supply in as the big ones. So we said, well, why don't we run a test of a small number of consignment-focused, small footprint locations where people are going to them into their neighborhoods. The first of which was we launched a few weeks ago in Palo Alto, California, and I'll be the first to admit, I didn't think that the Palo Alto store was going to crush it. I thought it would've been -- it's a good location, and it would have been fine, but it's just kind of blowing our -- it's doing better than the Madison Avenue store did in its first couple of weeks, and we're in the middle of a pandemic. So if that plays out, then we're going to see -- now we're going to have a marketing lever. So in the past, we [ spent ] on marketing to drive demand to supply, but demand comes along with it. We just kind of get this bundle of about 4 buyers for every consignor and several blunt instruments. But if retail works like we think it does or can, it actually gives us a target marketing vehicle that can target supply acquisition relative to demand. So that's particularly encouraging. So that's something we can control. Third thing we can control is our vendor channel buy. So historically, we've used that channel to fill in product gaps, high value, in particular. We've used it for handbags, jewelry and watches. It was 5% of the business. It's now 10%, and it continues to grow. Fortunately or unfortunately, that's the product that we need the most. During COVID we've seen our demand mix toward those categories. So we need even more of it so we can lean in there in a targeted way to try to acquire that type of supply. The overall opportunity for businesses, and these are all kinds of different businesses, but predominantly, resellers are who don't have their own demand channel, but they do have access to high quality supply. There's a ton of it out there, and it's better than in a nonpandemic world because it's not overstock. It's stuff that retailers and brands can't sell if locked in their stores or locked in their warehouses. So they need distribution for this that creates an opportunity in the near-term to get that supply, but also to justify making some investments so that, that can become a permanent part of our business at an elevated level, maybe 15% to 20%, who knows? Time will tell. But bigger than 5, right? So you can see that we need to do some things to enable that, but it's worth our why. So that's all good. That ends the positive part of COVID story than the stuff that is not really in our control. Today, we went down to lockdowns again in California, and that's happening kind of broadly across the country. Now our stores are limited to 20% capacity. They were 50% not long ago. So we're adapting, we have to stay around that and figure out how do we capture as much of the supply as we can, still allow people to shop and still have employees in the store. So we're adapting to that, but it just is what it is. And then in-home, that's -- we, a few weeks ago, started reintroducing in-home selectively in New York and a couple of other markets. And overnight, we were getting people with 50, 100, 200, 500 unit consignments. But then COVID spiked and we had to shut it down. Consignors didn't want it. Our employees didn't want to do it. But what that told me is that, that group of -- 2 things, that group of consignors is doing nothing with their stuff. They're not like, hey, let me try eBay or Craigslist. They're going on that stuff. And two, in-home white glove is the right channel for them. That's the only thing that's going to get them off the sidelines because they're not going to sit on the other end of a Zoom call and go through 100 items one-by-one with their salesperson. It's just not the right fit for them. It's too much friction. So if that group of people, again kind of your top consignors is 20-plus percent of your supply, that's just sitting there and growing. And it's really just a matter of COVID playing out before we can get a real unlock of all of that. [indiscernible] but probably correct.

Okay. No. So there's a lot in there, and I want to sort of probably piece out a handful of themes in there. I'm going to ask 1 short duration question and then more of the longer duration themes, I think, were more interesting in what you said. Short duration, you did provide the market with an update last week, talking about trends in October, November. Wanted to know sort of what the main messaging was about, what you're seeing in the business quarter to date just so investors who maybe missed that update a week ago better understood how the company was communicating that.

Sure. So yes, we provided an update right before we did another one of these conferences. This one is far better by the way. So the update was, things are going pretty much as expected. Demand is holding up really well. New buyer growth is accelerating. We had our best month of new buyers ever. New buyers are up 24% year-over-year. That coincides we starting to see the benefits of marketing. Order in our Black Friday weekend, Thanksgiving weekend was comparatively a little bit better than the rest of the month. Order growth was [indiscernible]. However, what was down was average order value and ASP due to those dynamics that I was talking about earlier, not having enough of the high-value supply. If what's coming in is consistent, but what's going out is skewed toward high value, you're just replenishing the stock in the high-value stuff. So we're starting to see the compound impact of that. And we're going to have to get targeted in how to bring in that high-value supply to get AOV back up and get supply back in balance. So those are the kind of key messages. Demand is great. Supply overall is trending well. It's positive. The next step is just getting the type of supply that we need, and that's the final piece in getting back to growth.

Okay. So in your answer about the learnings from COVID, you threw out a couple of things. When we think about you sourcing supply, is there something you're trying to solve for from an optimum mix standpoint over the medium- to long-term? Some people will go in-homes, other people will be offered virtual, other people will take advantage of real estate and then real estate could take a couple of different examples, big store, small store. How do you think about where you see the business trending and the trajectory of getting the right supply at the right mix? And what it might mean from a cost and a growth standpoint longer term?

Yes, sure. So there's 2 sides of that. That's what's optimal from our perspective and then what's optimal from the consignor's perspective. Fortunately, those are basically the same thing, right? So if you have 2 items to consign, you don't really even want to make an appointment to have someone to come to your house and get them unless they're big pieces of art or furniture, though they don't take much furniture anymore. And if you have 100 things, you probably do want someone coming to your house. So fortunately, there's a high correlation of what's best for us and what's best for the consignor. But generally speaking, we're going to let the consignor lead, but we're not going to do things that are sort of silly and not sustainable. Virtual kind of fits a nice sweet spot in the middle. I think -- as I think about longer term, how this is likely to play out. I've got some instances because we just did a pretty good consignor survey. And what people -- first of all, we confirmed a lot of what we kind of already knew. Our biggest source of growth in the biggest pool of new consignors and buyers are younger, they're millennial, the Gen Z. And they overwhelmingly prefer drop-off in a brick-and-mortar location or to do a virtual appointment. Third place is in-home. So they're saying even post-pandemic that's what they want to do. So I think those 2 channels are likely -- well, virtual is here to stay exactly where it goes. I don't know. Stores certainly going to increase their contribution to the overall pie. In terms of -- the good news is those 2 channels are the most cost effective for us, right? The people sitting in the stores are not our most highly compensated salespeople, and they're incredibly highly productive. And in a virtual world, our salespeople can handle many more appointments per day because they don't have to spend their time driving in between appointments and dealing with the products once they get them. So we just have somebody sweep by with a van, pick it up and send it to the warehouse. So that is a path to higher sales productivity from the formerly white glove team. Now [ I'll be clear ] to say, we haven't realized that increased productivity yet because we're still in this zone where we're getting back to sustained growth in supply, and we're hiring into the sales team to anticipate significant growth going forward. But at least that -- we see that potential, better consignor experience, better sales economics. Those are the wins, if you think kind of across the things that have been strategically most important to us in sales and operations, in authentication, et cetera, et cetera, that serve a dual purpose of improving experience and quality and driving down costs.

Got it. Okay. That's super clear. I want to stick for 1 more minute with the smaller store format. Do you as a company think you already know enough based on what you've seen through a handful of tests that you could see your real estate strategy changing now over the next 3 to 5 years? Or do you think it's still an open-ended question?

Definitely an open-end question because I think we've got like one small footprint store that's been around for about 1.5 years and then another one that's been around for about 3 weeks. So we are -- we're pulling the trigger to get as -- like as many as we can, but as many as 10 of these between now and some point in the second quarter of next year, and we would move much faster than we have been, but it's really the speed of the landlords. Because frankly, there has not been this reckoning where a retail landlord is accepting that they can't get pre-COVID rents and pre-COVID lease terms. We're going to be opportunistic about it. Because it's a test, we don't want to sign up to anything. There's more than a couple of -- 2, 3 years in duration because we're going to learn things and we're going to adapt our strategy. If we learn nothing new other than what we've learned already, this is the right strategy, and we're going to do more of them. Because again, it's more cost effective and faster to unlock supply. And we're confident that once we bring it in, the demand is going to be there.

Got it. And then 2 more follow-ups on the supply side. The virtual white glove appointments. Do you see those now being a real part of the arsenal in terms of sourcing supply now? And even if we go back to a world that's somewhere between where we are now and pre-COVID, that, that still remains part of the go-to-market strategy in terms of growing supply? And what would that virtual supply growth or virtual consignor growth means, again, for sort of thinking through margins and cost structure in the business long term of the stores, the cost of sourcing supply?

Yes, yes. So yes, we think it's here to stay. Exactly how large it is, is hard to say. It's kind of like what percent of investor meetings and investor conferences are going to remain virtual after the pandemic. It's like some. Exactly how much? I don't know. There's -- like there's a role for in-person and there's a role for virtual. They play distinct roles. So it's absolutely here to stay. We're going to invest behind it to get the experience better and to train our people and hire people and specifically who are more adept at that sort of style of interaction. It gives some -- there's a range of potential outcomes in terms of sales productivity and cost savings depending on what that penetration is and depending on what the productivity is. But I'll give you sort of directional sense. Our good white glove salespeople pre-COVID were doing about 4 appointments per day, and they were bringing in 17, 18 items per consignment. So here is like about [indiscernible]. In the virtual world, there's no reason that they can't do many more appointments, at least 8 a day. That's assuming that every appointment takes an hour. And that's not going to be the case. So we've actually got to get the data in how long these appointments actually take. But let's just say for sake of argument you can do 8 a day when you're fully booked. You can double your appointment productivity. And what we've seen so far in virtual, there's not much dropoff per consignment supply -- per volume or per consignment. You may be lose a unit or 2. So almost doubling of per person productivity per day applied to X percent of supply, right? So all you did was took a share of formerly white glove. White glove was 50%. So if you take half of that, I'm just making up numbers, right, so you can double -- almost double the productivity in about 25% of your supply. Of course, there's other moving pieces, stores going up, vendor going up, all sorts of stuff, that muddy the math. But boil all down, all things steady state, you could see a path to maybe 10%, 15% sales productivity all-in increase through stores, virtual, vendor. And the cost of the sales effort on a per order basis is about $20 today. So you take that out -- we're talking a few dollars of contribution profit increase.

Okay. All right. Maybe 1 more on the supply side. You talked about the B2B vendor channel rising as a percentage of the mix. Two-part question. Number one, what do you think some of the things we should be watching for as to the pace and the why that, that percentage moves up over time? And second, what output could we expect in terms of potentially improving your ability to authenticate in total supply on your platform at a lower total cost structure. Because some of that -- you would hope some of that, if not most of that inventory would be coming in very, very authenticated already.

Yes and no. So not so much. We're still going to authenticate every item because even -- frankly, even if the supply comes directly from brands, fakes make it in, right? So there's the one tactic of buy something new in a store, return a fake, right? So we get that too. So we still have to be diligent and go through our process.

I wish I had that much time on my hands. People have [indiscernible] time on their hands. I know you and I are very busier than being able to do that. That's right.

Right, but it happens. Anyway, so -- but on the sales side and the acquisition, it can be much more efficient. So we've got literally a handful of people on the sales side, bringing in 10% to the GMV of the company. So we're going to add another half a dozen of those as quickly as we can, who are going to bring in an established set of relationships. They tend to be category specific. They're in the jewelry business or they're in the watch industry or they're in the handbag business or they know a series of wholesalers and middlemen that are kind of across categories. They bring relationships and then like they know them, so we can accelerate that supply coming in. So that's, part 1 is just bringing in the resources and expertise to get access to the supply, similar to like a T.J. Maxx model at smaller scale, right, where we have buyers who have directed access to sources of supply. That's part 1. Part 2 has been scaling out our operations infrastructure. That's a little bit different when you're dealing with business sellers and when you're dealing with individuals. When you're dealing with businesses, obviously, they're in the business to make money. So they are more concerned about how we're going to price things. So there's a lot more back and forth on things like pricing before we take them. And we get these line sheets of lots and lots of product that we have to go through and figure out what we want and then we agree on price. There's just a thing that happens there. So we need to build that out. In the final -- we can do that, and that can help us continue scaling the business little bit by little bit. The bigger unlock is rolling out some technology that allows us to merchandise these products in a natural way so -- like our entire platform and businesses are built on the idea that every item is unique. In this channel, that's not always the case. We can have multiple quantities of the same thing or multiple sizes and colors of the same SKU. Our warehouse doesn't recognize quantities. A location points to an item, right? So you have to rebuild your warehouse management system. And in the front-end, you have to do the same thing. So say I got 5,000 of the same Gucci handbag. Today, you'd see 5,000 separate listings on the website, which is obviously no [indiscernible]. So how we handle that is our merchants launch a few to the site. Once they sell, they launch a few more. It's incredibly manual and not very scalable. So we need to roll out just what we call multi-SKU technology, what the rest of the world calls an e-commerce platform. So we need to build that over the next 6 to 9 months, which is kind of the bigger unlocks how we can sustain that business [indiscernible]

Got it. One more big picture push -- sorry, go ahead, Matt.

One other thing. You also mentioned kind of what does that mean for the financial model? If that channel in the future looks like it has in the past, it's higher selling priced products at a lower take rate with higher gross profit dollars per transaction. So it is additive to the overall profitability story given the inherent nature of the mix of products that tend to come through that channel.

Got it. I got a couple of e-mails from investors that are listening in and just wanted to ask generally maybe this is a broader industry point, given you gave an update last week. But generally, how are you feeling about the consumer and the luxury consumer in Q4? How are they behaving? How different is it this holiday period maybe than prior holiday periods? And then we'll get back to things more detailed around The RealReal.

I don't know that we're necessarily the best barometer of overall luxury consumer behavior. But if we were a barometer to the consumer is very healthy. The demand for luxury is quite strong. But the mix of what people want right now is different, right? It is definitely skewed more to the high-value products, right? So people weren't so much concerned with refreshing their wardrobes. The fact is they're not going to events. They're not even going to work, right? So we saw ready-to-wear mix down. I think it's starting to recover. I don't know exactly why that is, maybe it's just fatigue of seeing the same stuff in their closet or optimism that post-COVID is around the corner. So I think we're starting to see that mix up fortunately because we've got plenty of that supply. So I think luxury over the years has proved to be very resilient. It tends to be impacted less in just general recessionary times and come back faster, right? So I think that's likely to be the case here. People always built an inherent demand for the products from the top luxury brands, and that has not changed.

Got it. One of the most interesting things over the last year or so has been the company has announced a lot of partnerships with brands in the luxury industry. Can you just give us -- take a few minutes just to help us understand how some of those partnerships came about? What you think those partnerships do in terms of amplifying the efforts that against the long-term addressable market that the company is going after?

Sure. Yes. So the most recent partnership we announced was, I guess, a couple of months ago in October was with Gucci. And there's a couple of things that are unique to the Gucci partnership and I'll sort of do it in sequence. So we have existing partnerships with Stella McCartney and Burberry. And the basic idea here is and there's -- important to understand, there's no financial exchange in any of our partnerships. Really as these brands coming out on the side of the circular economy in saying that retail is healthy to the luxury ecosystem, and they want to be part of encouraging that. When you're done with your products, please keep them in circulation, resell them to somebody else and then please, in turn, come back and buy more new stuff from me, right? So that's the basic spirit of partnerships where we create an incentive to close the loop and encourage more activity in the primary market. Gucci also has that. But where they're different is they also came out on the side of advocating that people buy preowned as well. Their thinking is if more people experience their brand and the accessibility provided by resale and the value they can get, it can like lead consumers to fall in love with their brands and develop a very long relationship. And they're so genuinely sincere about their sustainability that they are really fine. Their products are built to last. They want to see them continuing to recirculate. So that's Gucci. I'll say that -- there were some other things that we had in mind. The partnership was basically locked and loaded right as COVID hit. So we weren't able to execute on all the elements of the partnership. But we hope that it's what we're experiencing right now, which is Gucci is doing very well relative to the overall business, but that will form the basis of a longer partnership that we consider to build on, both with Gucci specifically and then ally within other brands within the Kering portfolio and beyond. So I think as luxury brands look at resale and increasingly come down the side, it's here to stay, they need to find a way to embed that into their business and their strategy. And they survey the landscape, they're real -- like we are the only logical partner worldwide. We are the only company who goes through the painstaking process of authenticating everything and really doing the best possible job that we can to keep fakes off the marketplace. So long as that's also very important to the brands, we become the partner -- the logical partner of choice.

Got it. Okay. That's super clear. 2 of the biggest initiatives within the company. And two we get asked a lot about by investors are automation and authentication. You touched upon pieces of those in a couple of your answers so far today. Can you give us a broad update on both of those 2 initiatives within the company against sort of where you want to be a couple of years down the road?

Yes. So both -- in both of these cases, and I'll hearken back to a comment made earlier, the objective is to increase quality and consignor or buyer experience and simultaneously drive scalability and cost efficiencies in the business. So we'll take them one at a time. Automation. We started with a few different automation work streams. First one was pricing, automating -- we price all the products to resell, automating parts of the copy, description, the title, et cetera, and then the retouching of photos. At the beginning of this year, we thought we could get to about 75% item penetration in those 3 areas. We got there at the end of the third quarter, and we're continuing to find ways to continue penetrating into it. We're now into the 80-plus percent range, and we don't frankly know where the limit is. Each additional bidder penetration requires a different type of unlock because every subcategory is a little bit different nuance. We'll never get to 100%, but that's another good story during COVID. We executed on our strategy, didn't have setbacks and now we kind of are where we want to be. So the next step is driving the efficiency gains that come from that, right? So scale -- driving scales when we start to realize all the benefits of the investments that we've made. But by having a lot of this stuff algorithmically generated also gives a level of consistency in pricing. So if you think it's like 25 of essentially the same item came in, in the old world that had 25 different people pricing it, you could get 25 different prices, right? So the algorithm is always going to pick the same answer, and we can tune it to get to the highest part of the range and then watch the data to sell. So we can actually improve the consignor experience by extracting more ASP out of a given product and improve the buyer experience because they're seeing -- like they see the same price for the same item. Because like we see the same thing for 2 different prices, like, well, is one better than the other? I don't know. So these things go hand-in-hand, quality, scalability, cost. So we're in a good place there. We're going to keep working at more and more things, but that's all tracking very well. Authentication is very similar. So as you well know, I mean, like we've invested a lot for a long period of time to be very good at authentication, and we continue to do so. And as we get bigger and see more items, we scale our operations, we get more granular and better at that in a few different ways. We've always had some element of risk scoring items. So a Chanel handbag is a higher risk to be a fake than a low -- a kid's dress, right? So we build algorithms that factor in a lot of different variables to assess with -- the risk of an item being a fake and then putting that to the right level of expertise in our operations. Second, we do the same thing on the consignor side, right? So you may -- your item may not be all that risky. But if you just created your e-mail address that's in Russia and you live in Montreal and you've sent us fakes before, like then you're going to get a higher level...

Red flags, all kinds of red flags.

Yes, yes, so you've got an item and a consignor risk score that we keep optimizing it, adding more data, getting more data to work with, making better and better. And then when there's a person to look at it, that's the other thing that we've done as we've grown, there's increasing specialization. So if you went back 5 years, you have this team of experts and there were sort of generalist authenticators. Obviously, gemology was always going to be separate, but you could have somebody doing handbags and shoes and clothing, et cetera. Now we've created a tiered system level of expertise that's category specific, right? So you can have an individual train very, very highly in something that's very specific with appropriate risk scoring so that we continue to like do a better job of identifying potential fakes and keep them out of the marketplace. So we've -- our success rate has always been very, very high. And we're never going to get to exactly 100% effectiveness, but we're going to continue getting better and getting more efficient and scalable and driving cost out of it while also increasing the quality.

Great. Okay. There was a lot in there. But we only have a few minutes left. I want to leave on one other question. We talked a lot about where growth can come from the opportunity. How you continue, whichever form the platform in terms of how you source supply and some of the big efforts around automation and authentication. Can you roll it all up because, I think, the other piece away from growth we get is the path to profitability from investors. So maybe just in the last minute or 2 left we have, take us into your longer-term goals of what the platform will evolve to from a profitability standpoint? And what we should be watching for signpost along the way?

1 minute to forecast profitability?

You can go over. That's -- I'm trying to keep you on schedule.

Okay. There's a few markers. So the building blocks of getting to breakeven. Of course, that's not the end state. Gross profit per order, variable expense leverage and fixed expense leverage. We need to see gross profit per average transaction of about $100. We were at $90 last year. And the final piece in the puzzle was renegotiating our shipping contract with UBS. That's done. But then COVID hit and average order values [indiscernible]. So once we kind of get back to a post-COVID world, presumably, that's going to kind of normalize and that part of the puzzle is there. Then last year's $90 of gross profit, then you have to add another $70 of variable expenses that include marketing, operations, sales and our store operating cost. That was $70 and left us with a contribution of $20 million. To get to profitability, we need to shave about $10 of that $70 out. About half of it is going to come from marketing efficiency, the other half from automation-driven operational benefits. If there's something on top of that for sales efficiencies, great. Either we get there faster or we get there -- we just added additional margin. And then the final piece of the puzzle is fixed cost. So that was a little over $50 last year. So there's nothing magical about leveraging fixed cost. But the one thing that I'll point out is the incremental step-up since fixed costs become less and less significant the more we scale. So when we added a warehouse a year ago, we doubled our fixed costs overnight. We're about to add one in Arizona next year, and that's going to be a 50% increase in our -- our warehouse rent fixed cost, and the next ones are 30% and so on and so forth. So you can see the path that it doesn't contribute all that much between here and breakeven, but it starts to really pull a lot more weight from breakeven to forward. We think over the long term, this is a 20% to 25% EBITDA business. That's 2 minute.

That's great, Matt. Thank you so much. Matt, Paul, please, I want to say thank you for, again, being part of the UBS TMT Conference this year. Hopefully, we're going to do it in person next year. I wish -- my team and I wish you and your families a great set of holidays. Be safe. Be well. And thanks to all the investors for tuning in today for the fireside with Matt. Thanks.

Thanks, Eric.

Thanks, Eric.

Bye guys. Be well.