AUTO1 Group SE
XETRA:AG1
If you don’t study any companies, you have the same success buying stocks as you do in a poker game if you bet without looking at your cards.
Utilize notes to systematically review your investment decisions. By reflecting on past outcomes, you can discern effective strategies and identify those that underperformed. This continuous feedback loop enables you to adapt and refine your approach, optimizing for future success.
Each note serves as a learning point, offering insights into your decision-making processes. Over time, you'll accumulate a personalized database of knowledge, enhancing your ability to make informed decisions quickly and effectively.
With a comprehensive record of your investment history at your fingertips, you can compare current opportunities against past experiences. This not only bolsters your confidence but also ensures that each decision is grounded in a well-documented rationale.
Do you really want to delete this note?
This action cannot be undone.
| 52 Week Range |
3.35
11.16
|
| Price Target |
|
We'll email you a reminder when the closing price reaches EUR.
Choose the stock you wish to monitor with a price alert.
This alert will be permanently deleted.
AUTO1 Group SE
You don't have any saved screeners yet
[Audio Gap]to ask your questions. Before we start, I would like to draw your attention to the important notice and safe harbor statement at the beginning of this presentation. This presentation is also available for download on our IR web page, and we will also host the recording of this call on the web page.And with that, over to you, Christian, after a brief example of our current French marketing campaign. Christian?
Did the video already play, Philip?
We only have the picture, sorry.
Okay. We only have the picture. All right. Good afternoon, everyone, from Berlin, and welcome to the AUTO1 Q3 earnings call. We started AUTO1 on the firm belief that the traditional car buying and selling model leads to poor customer experiences. For nearly a decade, we have invested extraordinary, talented teams and superior technology to build the best possible way to buy a car. Our value propositions for both consumers and partners are unparalleled in speed, simplicity, comfort and trust across the markets we operate.For years, our teams have been executing tirelessly against our mission: building the best way to buy and sell a car online. Being a public company today, we feel it's the right time to add a corresponding financial target to that mission statement. It is our firm conviction that the value proposition with the highest level of customer satisfaction will win the market in the long term. Following that logic, our financial goal is to be the largest and most profitable car dealer in Europe. We have a world-class position to realize that target.AUTO1 is the industry leader in this massive EUR 700 billion market, combining all elements that we need for further success. Autohero is the fastest-growing online seller in Europe, seeing enormous interest from customers to buy their next car online. AUTO1 is the largest buyer of cars from private consumers in Europe, creating an unmatched set of inventory in size, attractiveness and GPU potential. We are aggregating the used car trade across 30 markets in the European Union. And this way, seeing an ever-increasing amount of trade flows across our markets. We own the largest amount of transactional data in the market we operate in, generating an extremely hard-to-replicate competitive moat. We are integrating used car production as a skill set deeply into our DNA, rapidly building our own production capacities.AUTO1 owns a EUR 485 million ABS facility backed by global top-tier banks, ready to scale quickly for future growth. And most importantly, AUTO1 is run by an exceptionally experienced and proven team of industry experts, who are extremely motivated for the next level of AUTO1 growth and profitability. Ladies and gentlemen, Q3 was a record quarter for AUTO1 on all metrics. We increased group revenue by 64% year-over-year to EUR 1.26 billion. In parallel, we increased group gross profit to EUR 116 million, growing 34% year-over-year. These results are driven by the exceptional experiences that we deliver to our customers and the superior market position that we have built over the last decade. Our Q3 achievements make us more than confident that we are on the right path towards our financial goal of becoming the largest and most profitable car dealer in Europe. Furthermore, our Q3 results continue our impressive long-term revenue and gross profit track record. Based on the firm belief that there is a better way of buying and selling cars, overcoming traditional car buying experiences, we have built a company which is truly customer-focused. This has led us from strength to strength with Q3 being our 26th quarter of exceptional revenue growth. In the last quarter, our products resonated stronger than ever before with consumers in all of our markets. Over all these years, we were able to hire the best talent in the market, scale all of our operations across Europe and win the hearts of our customers for the long term. Q3 was by far our strongest quarter in the history of AUTO1. With over 160,000 cars purchased, we bought more cars than ever before. Propelled by our market-leading sell-from-home solution, we made it, again, easier and more convenient for our customers to sell their vehicle, leading to a new record level of 140,000 happy sellers. Our purchasing power allowed us to successfully increase our retail cars purchased for Autohero by 4x year-over-year to over 16,000 cars in Q3. Our average purchase price climbed from EUR 5,709 to EUR 7,655 year-over-year. We attribute the increase in average purchase price predominantly to the higher share of more expensive retail cars purchased. Throughout Q3, we created an unmatched set of inventory in size and attractiveness, enabling us to offer the greatest variety of cars at fair prices. In Q3, Autohero delivered a record quarter on units, revenue and gross profit. We delivered 11,300 cars in Q3, which is an increase of more than 364% year-over-year.We increased revenue to EUR 160 million for the quarter, representing an increase of 413% year-over-year. We grew retail gross profit to EUR 4.1 million for the quarter, growing more than 533% year-over-year. In Q3, we experienced stunning demand from customers across all of our retail markets, who are fully embracing the better way to buy their next car. Our world-class team of over 5,400 car enthusiasts worked tirelessly throughout Q3 to meet the exceptionally high level of demand. Throughout Q3, we successfully ramped our retail operations Europe-wide and laid the groundwork to become the largest and most profitable car dealer in Europe. Following that goal, we have developed a clear roadmap to our long-term GPU target of EUR 3,000.We increased our retail GPU to EUR 365 in Q3, and we'll now start to realize the pathway towards a long-term GPU target. We identified several key long-term drivers and their potential impact on our GPU target. We estimate reducing average days to sale and cross-border selling to have a potential GPU impact of EUR 500 to EUR 700. As the leading digital automotive platform for buying and selling used cars online, we are in the pole position to effectively capitalize on used car price differences across countries and regions. We estimate smart pricing for purchasing cars to have a potential impact of EUR 400 to EUR 600. We already own the largest amount of transactional data across the markets we operate and are continuously investing into our pricing algorithms. We estimate cost of sales efficiencies to increase our GPU by EUR 50 to EUR 100. With the ongoing investment into our Autohero e-commerce platform, we will realize this GPU bucket step-by-step. We estimate that in-housing used car production has a potential impact of EUR 500 to EUR 700. We already made great progress with our in-house production strategy and see significant opportunities ahead of us. Furthermore, we estimate in-house financing to increase our GPU by approximately EUR 800 to EUR 1,000. Finally, we estimate increased conversion of additional products and services on Autohero transactions will likely increase our GPU by EUR 100 to EUR 200. Over the course of Q3, we further built out our Autohero brand, which resonated strongly with our customers. We were able to rapidly increase our brand awareness across Europe. In Germany, we achieved a brand awareness of 24%, just 5 quarters after launching our brand-building activities. In France, we even exceeded the strong results for Germany with the start of brand-building. Based on the learnings from our home market, we accelerated the brand awareness in France to 14% after just 3 quarters of launching our brand-building initiatives. Our branding and marketing investments generate an overwhelming level of consumer demand. Our web sessions increased by over 70% in Germany quarter-over-quarter to 3.9 million throughout Q3. While our customers are absolutely delighted by our brand and value proposition, we believe that making the Autohero brand known quickly throughout Europe will enable us to progress faster towards our long-term GPU target. Despite our strong growth, we were also able to increase our industry-leading NPS and our excellent Trustpilot ratings in Q3. Our group-wide Autohero NPS grew to a new record high of 65 in September, a clear reflection of our outstanding car quality, our superb service and, of course, the best car buying experience. Our externally measured Trustpilot scores reached an excellent rating across all of our markets in Q3. Germany has achieved an outstanding growth in Trustpilot scores over the course of last year driven by our relentless focus on customer experience. Throughout Q3, we made remarkable progress in scaling our in-house production facilities. We achieved our '21 used car production capacity target of 50,000 units already earlier this week and are on track for further rollout. This capacity consists of 2 centers in Germany, having output for 15,000 and 35,000 units, respectively. Our center in Hemau is already operational today, and our new Berlin, Brandenburg-based center will start ramping up at the beginning of next year. To deeply integrate used car production into the AUTO1 DNA, we by today employ 552 full-time employees across Europe to carry out key steps within the production process. These employees consist of skilled industry experts that carry out floor planning, operational, quality management-related, photography-related, mechanical- and repair-related tasks within the production process. The combined knowledge and skill set of these employees is the base for in-housing a growing volume of internally refurbished cars quickly. In-housing production will allow us to fully control each step of the process, leading to a faster and more efficient production and thus will be one of the key drivers to our long-term GPU target. Our Merchant business showed amazing progress throughout Q3. For the first time ever, our Merchant business alone achieved more than EUR 1 billion in revenue and more than EUR 100 million in gross profit in 1 quarter. C2B revenue grew strongly to EUR 933 million, while remarketing revenue grew in parallel to EUR 167 million, leading to a total revenue of EUR 1.1 billion. This is by far the strongest result we ever had. Merchant gross profit also hit a new record high with EUR 112 million for the Merchant business alone. Our Merchant average sales price climbed to EUR 7,575, setting the stage for a new record level of Merchant GPU at EUR 772. Overall, we achieved stunning results in all segments over the course of Q3 and made excellent progress towards our long-term goals. Today, we feel stronger positioned than ever before and view the opportunity disrupting the massive EUR 700 billion used car market further as greater than ever. Let me now hand over to Markus for an update on our financial performance during Q3.
Thanks, Christian. Our third quarter represented a landmark quarter in all respects. We achieved record quarterly revenue of almost EUR 1.3 billion in Q3, representing 64% year-on-year growth on a group-wide level. In Autohero, we sold 11,300 retail cars, representing a quarterly growth of 34%, reflecting the overwhelming demand from consumers for our product. GPU has continued to grow across both our Merchant and Retail businesses, reflecting -- resulting in a record gross profit of EUR 116 million on a group-wide level. Our adjusted EBITDA was a loss of EUR 24.7 million, representing a slight decrease of circa EUR 2 million quarter-on-quarter in absolute amounts but an improvement in the margin as the increasing investments in brand marketing and refurbishment was offset by the higher growth in revenue and gross margin.Our adjusted EBITDA result demonstrates the power of our model and our ability to invest significantly more in marketing. We invested around EUR 11 million more quarter-on-quarter, almost completely in Autohero while improving our EBITDA margin and keeping the overall loss broadly flat.Our ability to grow the Autohero brand in sales, while at the same time, significantly increasing our purchases proves out our model and our ability to very efficiently purchase the cars we seek, further demonstrating the synergies between the sales channels. As we turn to inventory, we significantly added to our inventory despite an environment where most car dealers are finding it incredibly difficult to source cars. As Christian highlighted at the beginning of the presentation, we're building our Autohero inventory in particular as we continue to grow the Retail business. While we added over EUR 100 million in inventory quarter-on-quarter, mainly in Autohero, our days of sales increased only by 2 days to 27 days, reflecting our ongoing ability to manage inventory very efficiently as a result of having Europe's largest wholesale business embedded within our Retail business. Our balance sheet continues to be incredibly strong, a result of the equity received at IPO, a profitable Merchant business and our existing rated EUR 485 million ABS facility with which we can finance our inventory on a pan-European and nonrecourse basis with an extremely low cost of capital.We ended the quarter with around EUR 767 million of total cash and liquid investments. We also ended the quarter with EUR 384 million of inventory against which we drew EUR 95 million over the course of the quarter through our ABS facility for a total draw of EUR 115 million. On the basis of circa 90% loan-to-value on our ABS, we have a further circa EUR 231 million of available facility headroom, leading to almost EUR 1 billion of total liquidity. Also on the asset side, we ended the quarter with around EUR 25 million of consumer finance receivables. We are in the process of putting together a warehouse facility, which we believe should be completed by Q1, which will finance the overwhelming majority of these receivables, giving us even more liquidity on a pro forma basis. As a result, we believe we have the available investment capacity to build Europe's largest and most profitable car dealer. As you heard in Q2, and we reiterated today, we have pulled forward our investments in production centers as well as our unique Autohero glass truck delivery fleet to be able to meet the overwhelming consumer demand that we are seeing for the Autohero product. With respect to production CapEx, as you heard earlier in the presentation, we have already secured 50,000 of production capacity in 2 centers to build our in-house production capabilities and are deep in discussions with several more and believe we now have a much more detailed and clear basis to provide guidance in the necessary investments, specifically over the next 9 quarters. On the basis of these ongoing discussions, we expected the majority of the production centers we are looking at over this time period to be long-term rentals of mainly brownfield sites. Some of the production sites might be acquisitions, and some are greenfield sites. But on the basis of what we are currently looking at, a significant proportion will be rented. As a result, we will invest circa EUR 120 million in CapEx for our production centers, which assumes that we will have capacity equal to around 400,000 annually refurbished vehicles. On the basis that we have signed 400,000 unit capacity by end 2023, we expect the quarterly rental cost to be around EUR 2 million. In addition, as we are further advanced in rolling out our Autohero delivery fleet, we expect to spend circa EUR 50 million on the Autohero delivery fleet as a result of faster investments and the positive consumer experience that it yields. Nonetheless, we believe that we can and will achieve increased delivery and truck cost efficiencies over time. Finally, the EUR 30 million of other CapEx over the next 9 quarters primarily represents previously planned investments in maintaining branch network and general fixed asset CapEx. Now to guidance. We are increasing our 2021 revenue guidance to EUR 4.5 billion to EUR 4.6 billion for the year, up from EUR 4 billion to EUR 4.4 billion as a result of first improvement in the average quality of the car sourced, a direct consequence of sourcing for Autohero, thereby yielding higher-quality cars also in merchant; second, fewer lower-end cars sourced being the ongoing consequence of our all-digital sell-from-home product; and lastly, the current market environment. We are increasing our gross profit guidance to EUR 415 million to EUR 425 million, up from EUR 310 (sic) [ 380 ] million to EUR 410 million. We believe our adjusted EBITDA will be a loss of approximately 2.5% for the year. Group units should be circa 600,000 units, merchant units around 560,000 units and Autohero units between 40,000 and 42,000 units for the year. We are very proud of our results and believe this represents ongoing proof that we deliver on our commitments. When we look back at our ambitions for this year at IPO, we are proud to have achieved and indeed exceeded all of our financial goals. With that, I'd like to hand over to Q&A.
[Operator Instructions]
Thank you, Rafael. And with that, we'll start with Catherine O'Neill from Citi.
So I have a few questions. Maybe I'll take it one by one to make it easier for you. Firstly, on used car pricing obviously remains elevated due to market conditions. I just wondered if you could give us a bit more detail on how much of your GPU or EBITDA be related to this versus some of the other factors you talked about and how sticky you think the RPU level -- RPU levels will be? And in relation to that, I guess, how you think about the supply challenges easing going forward and how that then feeds through to pricing, volume and gross profit for you, especially on the merchant side.
Yes. Thank you, Catherine. So I think to answer that question, it's important to look at the base car and its price development over the last 12 months. So if we take a comparable car of a certain average cluster across Europe, then we estimate the 12-month price increase to be around 5% to 6%. This is where we see the market being at, at the moment. This means if you translate this into what is market favorable effects that we've seen in the data and what is the -- what are the effects that are here to stay, then our belief is that around about 75% to 80% of what we're seeing today is something that will stick with us that is driven by other developments that drive up ASP and GPU. And then I will come to a couple of explanations to this in a second. And I would say 20% to 25% is market made. The effects -- one of the predominant effects that we see generating higher ASPs for the Merchant business is that we are -- have been very, very active in pricing retail cars. So when we are pricing retail cars for Autohero, then we are looking at a very broad range of costs. And the more data we're seeing, the more confident our algorithms get at pricing retail cars. So we're very much at the beginning of this, but we're starting to see positive effects. Why would this have -- this high amount of retail cars priced have an effect on merchant cars? So at the moment, it's still possible for our merchants to overbid us in cars that we price for retail. So if you were selling your car for EUR 10,000, and we would like to sell that in Autohero, and if another merchant puts another EUR 500 in terms of -- as a bit on top of that car, that car will go to that merchant. So we ensure competitiveness of Autohero within our own platform. So what we're seeing that there's a lot more merchant cars that are now appearing in the retail type of clusters of cars. This is one of the strongest factor of driving ASPs up in the Merchant business. There's also -- of course, if you look at the average price level in sourcing as a whole, the effect of the Autohero cars purchased, which are much higher. If I look forward, I would say, the Autohero gross profit -- the Autohero ASP that we're at the moment seeing, we're pretty comfortable with the level that we're seeing right now. And we also think that this will continue throughout the next year. On the other hand, the merchant ASP, I think we can come back to what I just explained. So we think that 80% of that effect is there to stay. 20% of that effect might be seasonal. But yes, we are having, I think, as the full market a hard time and creating an expectation of when new car deliveries will hit the market stronger. At the moment, I think the market assumes that the earlier state could kind of be yes, late Q2, Q3 next year. But overall, we are very happy with the numbers that we have achieved. And we also don't see any material effect in the numbers that will -- yes, that will drastically change the KPIs that we are seeing due to the explanations I've just given. I hope this helps.
Yes, that's really helpful. I mean, actually, on the Merchant business, I just wondered if you had any view about volumes because we are yet to get back to the 2019 levels. Do you think that's something that could occur in 2022?
I think that could occur, absolutely. I think we have to compare apples-to-apples. So due to the introduction of sell-from-home, we're seeing a lot less cars below EUR 500 and a little bit less of cars below EUR 1,000. That is also good because those types of cars are not really matching our gross profit requirements. I think overall, we are quite bullish when it comes to units next year, but maybe, Markus, you have some more detail on that.
Yes. I think we're going to provide an update in terms of how we see next year when we report our annual numbers in March of next year. But I think we're quite comfortable with our units for next year and the expectations out there.
Great. And I just wanted to come back to the -- your long-term retail GPU target of EUR 3,000 and the drivers to get there. Could you talk about what you need to do in terms of where you are now and what you need to do to get there and maybe the time frame you think about to deliver on some of those key drivers? I mean, is there a particular sort of unit volume you need in Autohero to achieve this? And then specifically on financing, that seems to be the biggest component or single -- biggest single component around that expansion target. So if you could maybe provide us an update on the progress on in-housing, what the attach rate is, et cetera, that would be great.
Yes. So in short, we need to do a lot of things to make that happen. So as you can see on this Slide 10, there's completely different buckets that we are addressing to a certain extent at the moment but with -- to a much larger extent over the coming quarters. A lot of those buckets are driven by scale of Autohero -- by scale of Autohero and unit and also by the customers that we have on the platform. So reducing average days to sale is a reflection of the demand that we're seeing on the Autohero platform, is to a good extent an output factor of the brand awareness and the traffic that we are generating on the Autohero platform. Cross-border selling is also a unit game. So it doesn't pay off to try to realize a cross-border margin if you're using like a single car transport and send one car between 2 markets. So we need scale. We need to use large transporters to create efficient unit costs, to have logistics costed up in favor of the overall cross-border margin potential. We're talking about smart pricing for purchasing. This is a pure data game. You can see in the elevated merchant ASP levels that there's a lot of retail cars that are making it through to the merchant side already. And this is a reflection of the continued increase and the additional data points that we are generating through the much elevated transaction levels that we're seeing in Autohero today.But we are at the beginning of this, yes. So we want to scale Autohero in all of the 9 markets that we are operating in parallel much more. And the faster we're getting the local markets scale in terms of transactions, the more of that smart pricing effect we will see.Cost of sales efficiencies are related to a lot of smaller variable costs, which also will be optimized through scale. Increase in our production, we already spend a lot of time on that. But this is a clear driver because on the one hand, there's the margin that our partners are making. On the other hand, there is also the specialization of tasks that we can do much better targeted towards the type of cost that we have in our funnel when we're taking this over ourselves. And then in-house financing markets, it's probably the most long term of this because we need to build up a book with a historic track record, where we can show that we are able to deliver decent loss rates and decent performance of the book. But maybe Markus, you can add here. Maybe I'll finish with the last point, additional products and services. This is anything that you can measure as an attach rate to the actual Autohero sale that is not financed. It can be insurance, it can be tires and a lot of other small things. It can be warranty service and so on. We estimate this to be EUR 100 to EUR 200. It's a -- on all units, measured in all units. And it's a question of conversion of those products on each of the transaction in each of our markets. But maybe to in-house finance, Markus?
Yes. So on in-house financing, we continue to be exactly on track from where we are. So we are currently providing financing, internal financing in both Germany and Austria. Attach rates continue to be at or above the point-of-sale market rates in those countries. I think in the past, we had talked about expanding in Q1 to other markets, and so we are still on track to do that. As I mentioned in my scripted piece, we're in the middle of working on a warehouse facility, which we expect to be finished by Q1. And again, on track to do that. I think that is the first step to eventually be able to then sell the loans and get the, if you want, the gross profit of that in the same quarter that it's generated. So I think all the elements are exactly on track for what we are planning to do.
Thank you, Markus, and thank you, Catherine. And with that, over to Sherri Malek from RBC.
I have 1.5, given one of my questions have been answered already. First one, should we assume that all of the improvement in Autohero's GPU this year has been on the trade margin? Or how much improvement has there been, if any, on the refurbishment cost per unit and financing income? And just part of the reason for that question is to understand the starting point for those 3 components of the margins so that we can apply those long-term GPU drivers, which you've hopefully shown on Slide 10, correctly?
So I would agree with you, Sherri, that a lot of the effects that we have seen this year are coming from trade margin, specifically through the introduction of branding. So you get a little bit of the first EUR 100 of the first bucket. I think this is what we have seen. There's now cross-border selling in there. There is not really any in-house production effects in there. And also when it comes to the other buckets, I think I would attribute as you were saying, like most of it to reduce average days to sale because of a very successful introduction of the Autohero brand across all of our markets. And the half question, sorry, what was it again?
You've answered both in one. Just maybe a follow-up question, which has come to mind, around marketing. How should we think about marketing investments, given kind of the priority really to drive brand awareness? Originally, it was EUR 200 million over 3 years, but it looks like it might be more than that. So if any color you could give on that would be very helpful.
Yes. So I think we said in the past that we were going to increase that EUR 200 million over the next 3 years. So that will be higher, although we haven't given specific marketing guidance. In Q2, I think we said we'd spend around EUR 100 million in the second half in total, of which the majority of that increase, of course, is coming from Autohero. We're running slightly ahead of that. And so that's going to continue to be an area of investment.
Thank you, Sherri. And with that, over to Will Packer from BNP.
Three from me, please. Firstly, in Q2, you talked to reconditioning and last-mile delivery as key chokepoints dragging on Autohero quarter-on-quarter unit growth. Do you think we're through the worst of those issues and you can sustain the improved quarter-on-quarter growth as we look forward? That's the first question. Secondly, you talked very clearly about your accelerated CapEx investments, reflecting your plans to expand reconditioning and last-mile delivery footprint. Does that translate into faster unit and GPU growth for FY 2022 at Autohero? Is that the right way to think about it? So could you kind of quantify that? Or is it just more CapEx with a theoretical longer-term benefit? And then finally, you talked to reconditioning units of around 200,000 by the end of 2022 if we assume all reconditioned capacity is utilized. Now I understand it takes a bit of time to get to. So what's a realistic level of actual utilization by the end of 2022?
Maybe I can start with the first question on the quarter-on-quarter growth. So an easy explanation of what happened in Q2 over Q1 is simply that we sold out. Yes, so we sold out in between Q1 and Q2 because we were surprised by the strong level of demand that we generated to the initial introduction of the Autohero brand and the ongoing marketing. By now, I think we have very well understood how the cohorts from having a car purchased until it finally delivered work much better throughout our different markets. So I think we have now reached a point where quarter-on-quarter growth rates should be seeing, I would say, similar tracks than we have seen in the last quarter. But obviously, always based on the overall unit expectation that we have guided to. On the CapEx, you think, Markus, maybe this one is yours.
Yes. Happy to do that. Will, thanks for the question. On CapEx, we wanted to provide, as noted, very specific CapEx guidance because we're now, if you want, deep in the weeds with conversations with a number of different players. In terms of our 2022 guidance, we will be providing that in March when we report our annual results. Right now, we are very comfortable with the existing consensus forecast of 88,000 Autohero units for 2022. And so I think we stick to that. I think likewise, we had always said that we were looking for EUR 1,000 per car GPU in 2023. We are extremely comfortable with that. I think the fact that, Christian, we are now presenting our roadmap to go beyond that to EUR 3,000 per car, I think shows that this is now going to become an increasing focus for us to really drive that GPU. But we are not improving our guidance on either those points today.And I think you then had a final question around how to think about the ramp-up of the reconditioning units. So the way we think about that in terms of ramp-up is particularly since the majority -- the vast majority of the conditioning or reconditioning or production sites, I should say, that we are investing in are generally brownfield sites in the sense that they are existing buildings and parking lots, and they have all the essentials there. It generally takes us around 6 months to achieve one shift, so kind of half of the full utilization. So the capacity that we're giving is the capacity on the back of 2 shifts, and then it takes a further 12 to 18 months to ramp that up to the full capacity.
So if we were to think about the percentage of your Autohero units which will be reconditioned internally at the end of 2022, any update on how we should think about that or no change?
The percentage of Autohero units conditioned internally, I think we're not giving some specific guidance on that. We're working as quickly as possible to have as much as possible be reconditioned internally over the course of 2022.
I think what we said was that we were -- in December refurbished 20% of our December deliveries, measured in deliveries in-house. That has obviously nothing to do with the December deliveries. So it's just a way of measuring it so any type of refurbishment cost impact you would then see in Q1 as those units would get sold. Obviously, it's our target to grow that share step by step quarter-over-quarter because otherwise, it wouldn't make sense to invest all of that money into refurbishment. But I think it's not yet the time until we can give a completely clear guideline on how the percentage will look like quarter-by-quarter. Having said that, also with our external refurbishment partners, we are increasingly seeing cost efficiencies that we can realize step by step.
Thank you, Bill. And that's over to Adam Berlin from UBS.
Three questions from me if I could. The first thing is you said you bought 16,000 retail cars in Q3, up 4x year-on-year. How much higher do you think this can go in the next 2 to 3 years on a quarterly basis? The second question I had was, I don't think you gave the CapEx number for Q3, at least I couldn't find it anywhere. So can you just help us with the -- how much of that's already been spent? And how much more CapEx will be in Q4 and maybe try and break down the EUR 200 million for us as much as you can between Q4 next year and 2023? If you can give any more detail on that, that would be very helpful. And the third question was just a technical question on the asset-backed financing withdrawal process you have in place. Can you draw 85% of the value of the entire inventory of AUTO1, including the B2B cars, for merchant use? Or is it just on the value of the inventory for Autohero?
Maybe I'll start with the first one, and then the second and third one is for Markus. So 16,000 units purchased, yes, absolutely right for Autohero in Q3. How much higher can that go? Much higher. I think we will, yes, in the medium and also in the long run, continue to source a large amount of our overall available Autohero units as sourced from private customers.So those are the vehicles that are most interesting, create the largest set of variety and has also -- have also the highest gross profit per unit potential in the long run. So they are a substantial part of our EUR 3,000 GPU game plan. While overall, we estimate that long term, the share between externally sourced and internally sourced will be 75 to 25. But we feel very confident. And I think the Q3 results show again how powerful the infrastructure is we have built over the last 10 years. And therefore, we're very positive when it comes to that part of the business.
Yes. On the CapEx, I think if you look at our Q3 report, you can find that number for the first 9 months of 2021. It was around EUR 15 million of -- EUR 15.8 million of CapEx. In terms of the quarterly breakdown going forward, not comfortable to give that on a quarterly basis. Clearly, that depends on the speed and the timing of the negotiations and the conversations we're having right now around the refurbishment centers. And so I think that's why we provided the framework that we did. With respect to the ABS facility, it's around 90% loan-to-value, and it's on the total inventory. So inventory across both merchants and Autohero. It's a single facility across all of our inventory.
Okay. So can you say anything about how much of the CapEx will be this year? I mean, it's already nearly the end of the year. You must have some idea where Q4 is going to land.
I mean, I would say, given that we are looking at it from a kind of 9-quarter basis, I would say it's somewhat -- it's a little bit -- it will be a little bit more than 1/9 of the total amount.
Thank you, Adam. And with that's a new voice, Stephanie [ Placencia ] from Goldman Sachs.
It's actually Lisa Yang from Goldman Sachs. The first question is on your in-house reconditioning. So as your -- you just acquired 2 facilities. I'm just wondering if you could maybe give us some color in terms of how that transition to in-house is going? Like can you talk about any potential challenges that you're facing that you're doing things in-house, also things are going pretty smoothly? Any sort of initial feedback would be great. The second question is on the remarketing business, which seems to be a bit weaker than expected in terms of units sold. Can you talk about the trends you're seeing there in Q4 and next year? And do you still feel confident you can meet your mid-term targets for marketing more specifically? And third question is on your Q4 outlook. So you have sort of tightened the range a little bit for the Autohero units sold and lower the upper end of the range from 43,000 to 42,000. So just if you can maybe share a bit of color in terms of what are the risks you might see into Q4, especially given the limitations we might see in terms of delivery or logistics? Any color you could give us on that would be great.
So on the in-house reconditioning, I think the primary or primary challenge is around hiring fast enough and finding enough talent. So for instance, here in the Berlin, Brandenburg region, we have also Tesla to a certain extent, looking for the same type of people. But yes, we're more than confident that we are on a very good track here. We already employ 552 full-time employees across all of our centers, including at our external partners that are having exactly that type of profile. But this is a clear challenge. And this is something that we are very much focused on because of that. But yes, other than that, I think the center is there. It's close to our logistics provider. So we are making sure this way that we don't need to move the car too much and overall equipment and kind of small in-house changes that we need to do with those facilities, they are, I think, less of a challenge. Yes. With regards to remarketing, the overall B2B channels are facing, I would say, the strongest effect when it comes to the effect that new car reduced deliveries half on the market. We are still growing substantially in remarketing volume year-over-year. Most of the players are actually shrinking. This -- the strategy that we are running behind remarketing is to push the amount of active partners. We have shared some data when it comes to the total amount of active partners in Q2. And we're seeing continuous positive trend there because we know that once the market is back, we'll also get the benefit. But having said that, remarketing has also -- is not a stand-alone operation within the AUTO1 Group. Remarketing has also a clear target to source externally interesting cars for Autohero, and we will measure it step by step more by that, reach of goal there as well. And yes, we absolutely think that it can contribute nicely to interesting remarketing -- to interesting retail supply while it can continue to grow over the next couple of years also on the units that it makes available to our merchant base. And then on Q4, I think, Markus, that will be your question.
In terms of -- I think the question, if I recall, was the challenges around Autohero units for Q4. Was that the question?
Yes. So you just tightened your guidance for units sold in Q4, but you lowered the upper end of the previous range. So just wondering what drove that change?
Yes. I think we're very comfortable. We want to make sure that we hit precisely our targets. I think we want to keep the record that we have had at least -- we want to keep the record we have had for this year in terms of meeting, if not beating our targets.And I think there's -- this is the first time in December that we are really ramping up such a huge number of units. So we wanted to make sure that we're aligned, not just internally with ourselves but also with our partners to be able to do that. And I think that's what led to the tightening of that Autohero unit range.
Thank you, Lisa. And with that, we've actually come to the end of the questions from the analyst community. So with 5 minutes to run, I think that was very efficient. So thank you very much, everybody, and we will hopefully talk latest with the Q4 numbers, if not before then. Thank you.
Thank you, everyone, and have a great day.
Thank you. Bye-bye.
Bye.