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
Hello. Good afternoon and good morning to American participants. Welcome to the AUTO1 Group Fourth Quarter and Full Year 2022 Results Presentation. I'm Philip Reicherstorfer, Group Treasurer. As always, I'm joined today by Christian Bertermann, our Co-Founder and CEO; and our CFO, Markus Boser. We will start with the presentation followed by questions and answers. [Operator Instructions] Before I hand over, I must make you aware of the safe harbor provisions at the beginning of the presentation. This will apply to any forward-looking statements made by management today. And now over to Christian.
Good afternoon from Berlin, everyone, and welcome to the AUTO1 Group Q4 and Full Year 2022 Earnings Call. In August of last year, we celebrated our tenth anniversary. The first 10 years of our journey were exceptional. In many ways, we created a platform that has changed the way our market operates, providing our customer base of consumers and dealers with great value.
We launched products in our platform with a constant aim to create exceptional experiences. By today, they have served millions of customers. And we assembled the team of outstanding talent, propelling us from 0 to market leader by applying their smartness, dedication and determination for success every single day. I could not be prouder of what our team has achieved over the last decade. And I could not be more excited about the next 10 years of our journey.
Although we have been growing our business massively since 2012, it still feels like we just started. The strength and success of our platform has enabled us to take a unique market position. We are Europe's used car powerhouse. Our sell-only and remarketing offering generate the supply for the largest European used car wholesale platform, auto1.com. With Autohero, we have become the leading online retailer in Europe. And over the course of last year, fend-off competition successfully.
Under our financing arm, we built up a continuously growing consumer loan portfolio, which now has a total size of more than EUR 185 million. Because of the strength of our platform, we continued our growth path and landed a record market share of 2.5% of all used cars sold in Europe last year. We achieved this all-time high market share in a year when used car transactions came down significantly below their level of 2016. As a result, of the general economic climate on the one hand, high prices and high levels of uncertainty, there were 2.4 million less used cars traded in Europe than the year before. The fact that we grew our market share by 19% year-over-year in such a challenging market proves the resilience of our business model and the growing power of our platform under any macroeconomic circumstances.
For the group, last year was incredibly strong. We grew units, revenue and gross profit substantially and continued our long-term growth track record. We trade at more than 649,000 units last year, an increase of 53,000 units compared to '21. Group revenue increased to EUR 6.5 billion, representing an increase of EUR 1.7 billion compared to the previous year. Gross profit climbed to EUR 488 million for the full year, and that is EUR 57 million more than the year before, where we recorded gross profit of EUR 431 million. Not only do we have strong year-over-year increases in units, revenue and gross profit, '22 was also our highest unit and gross profit year since the foundation of the company. I am very proud of our teams achieving these company records, executing our ambitious strategy and delivering on our growth trajectory. Our talented team achieved all of this in the market where used car transactions were historically low in Europe, driven by a highly volatile price environment, combined with high levels of inflation.
Let's take a look at merchant segment performance now. Two noteworthy factors influenced the used car market in Q4. First, market prices for used cars dropped sharply last quarter. After substantial increases in used car prices for several quarters in a row, market prices reached a peak in Q3 of last year. What followed was a record decline in like-for-like market prices in the fourth quarter. Within just 5 weeks, prices dropped 15%. As mentioned several times, we expect the prices to decline, but in a slow and much more controlled manner. Instead, what we could observe was a fast and steep drop. This quick and record decline in prices put pressure on our inventory in Q4. The price drop had a negative impact on Q4 GPU and total GP. But taking so much decline in just 1 quarter also means limiting its impact to that quarter. Since the start of the year, we are observing a much more stable price environment comparable to Q4 prices.
The second factor influencing used car trading in Q4 was the tense situation in the transport market. As you know, we're not operating transport ourselves but assembled a network of several hundred carriers across Europe. The main reason for the logistics shortages are a current lack of overall capacity, in many cases, driven by a lack of truck drivers. Reduced new car volumes throughout the pandemic led to short-term capacity cuts by transport operators. A spike in new car deliveries in Q4 led to OEMs to buy up large capacities, additionally putting the transport markets across Europe under stress. The fact that overall transportation times in Europe were slower impacted merchant demand in Q4. While we expected the logistics market, where we expect the logistics market to return to normal capacities over the next couple of months, our strong and experienced team is navigating the short-term headwinds in the best possible way.
Our merchant business performed strongly last year despite the number of used cars traded being smaller than in 2016. We grew units by 31,000 compared to '21, and in total, sold 586,000 units in our merchant segment, despite significantly lower transaction volumes in the market. Out of this total number, we sold 89,000 units in our remarketing channel, representing an increase of 20% year-over-year.
Q4 added a total of 134,000 units sold to the group's annual results. Revenue in our merchant segment was strong with EUR 5.48 billion. That is 30% or EUR 1.3 billion more than in '21. Remarketing revenue increased by 35% year-over-year, generating EUR 879 million for last year. In Q4, merchant revenue contributed EUR 1.18 billion to the group's annual revenue.
Merchant GPU was EUR 714 for the full year, which is within our expected medium-term range. Merchant gross profit grew to EUR 418 million. Over the full year, we stayed within our merchant GPU range and delivered a solid level of merchant gross profit despite the strong market headwinds in Q4. In addition to the sharp market purchase price decline in Q4 and new to used car transactions in Europe, unexpected Tesla write-downs had an impact on our merchant GPU and pushed it towards the bottom end of our range in Q4. In a surprise move, Tesla cut their new car prices drastically. And as a consequence, the value and prices of used Teslas adjusted accordingly. If we exclude Tesla write-downs, merchant GPU would have been well within our merchant GPU corridor. Merchant gross profit came in at EUR 90 million for Q4. As we are now observing a much more stable price environment, we feel comfortable with merchant GPU transitioning back to former levels.
Let's switch to retail. Our retail segment is performing strongly. It was the best year in Q4, the best quarter in the history of Autohero. Throughout the last year, we strengthened our position as the #1 online used car dealer in the European Union. We continue to serve our customers excellent experiences. And we made a number of bold investments. We extended the leadership position of our Autohero brand across markets, and we were rewarded for our endurance by a number of competitors leaving our markets.
Autohero delivered a record number of 64,200 units in '22 compared to 41,400 vehicles in '21. That is a year-over-year increase of 55%. With this volume, Autohero hero belongs to the top 10 retailers in Europe out of some 200,000 merchants. In Q4, Autohero grew deliveries to 16,600 vehicles, which is a 19% increase compared to last year. Revenue hit the EUR 1 billion mark for the first time and climbed to EUR 1.06 billion in '22, which is 83% more than in '21. Q4 revenue increased to EUR 265 million, an increase of 27% compared to the previous year.
We achieved a record retail GPU of EUR 1,254 in Q4, representing an almost threefold increase compared to EUR 418 in Q4 of '21. The foundation we laid in '22 to improve retail GPU and unit costs in Autohero is continuing to pay off. Our team is very focused on every element of unit economics in retail and working hard to optimize each of them every single week.
Our historic growth track of GPU so far gives us confidence into reaching our long-term GPU target of EUR 3,000 as planned. And as a result of strong GPU improvement and high unit growth, retail gross profit for Autohero in total more than quadrupled to EUR 24.4 million in Q4, up from EUR 5.8 million in Q4 of '21. Retail gross profit is continuing to become a meaningful contributor of total gross profit for the group. It made up 21% of the total gross profit in Q4 and that is an increase of 40% compared to Q3.
Our in-house production centers are one of the main drivers for improving unit economics and continue to be one of our major areas of investment. Our production center in Brandenburg is our flagship facility, and it sets the bar for all other production centers in Europe. The production cost in Bradenburg has decreased by EUR 420 per unit from Q1 to Q4 of last year. This major success is based on 2 main factors: on the one hand, smart software is making our processes much more efficient. On the other hand, you benefit step-by-step from economies of scale. All other in-house production centers that have been in production for 3 months or longer are tracking very close to our center in So overall, there have been many great learnings by our production team throughout the year on how to do work more efficiently at a given quality, or how to move units faster through the whole process. In-housing our production volume, is a core pillar of our profitability strategy. We ramped up the in-house here from 6% of all units produced in January of last year to 54% in December. This was a massive ramp-up in just 12 months and I could not be proud of our teams to have achieved that target. Since the end of last year, more than half of Autohero vehicles are being refurbished in our own production centers. We aim to increase this share to 90% until the end of this year.
We also made outstanding progress on expanding our in-housing footprint across Europe. At the beginning of '22, our maximum production capacity stood at 90,000 cars. Throughout the last year, we opened 2 more production centers: 1 Italy 1 in Belgium and increase the capacity at our first center in Hema in Germany. In Q1 of this year, we announced the opening of a new center in Netherlands, which will go into production in the second quarter of this year. So if you take all production center capacity into account, our total maximum capacity now stands at 147,000 cars at full utilization, that is an increase of 64% compared to January of last year.
We also made significant progress on our efforts to improve marketing efficiency in retail. For the first time, marketing cost per car delivered were below EUR 1,000 at EUR 900 for Q4. This is a 68% decrease from the year before and EUR 200 less quarter-over-quarter. At the same time, brand awareness for Autohero in '22 increased 66% year-over-year to 20% of all Europeans in the 18 to 64 within our retail target markets. Experience, dedication and hard work of our team, combined with the outstanding Autohero product experience are the main drivers for these strong results.
Overall, last year was a record year for us. Reflecting on last year's results, I can proudly say that we successfully achieved many milestones and navigated macroeconomic headwinds swiftly. Our team stayed focused on our strategy, always keeping our customers in the front and center of everything we do. We moved with constant speed and the high energy level that characterizes us, and we will continue to do so in the years to come.
So with that, let's take a look at '23 this year and beyond.
As you know, since June, we began to successfully lay the foundation for long-term profitable growth. A number of our key drivers within our P&L were optimized into the right direction towards our targets, most of them having a direct influence on our P&L. With EUR 1,254 retail cars are now delivering a GPU that is EUR 836 higher than Q4 1 year ago. Autohero GPU has crossed the merchant GPU corridor over the course of last year, and is, therefore, become an important part of the group's total gross profit. Increasing the share of internally refurbished cars from 6 to 5 is very beneficial for unit cost in Autohero as we save several hundred of euros per car produced internally versus externally.
Marketing cost per unit for Autohero at just EUR 900 in Q4 means that we are now able to acquire Autohero customers and invest into the continued strength of the brand at lower cost than ever before. In addition, we increased NPS by 9 points to 70 year-over-year. Great NPS means that we continue to deliver best-in-class customer experiences, and that then turn helps us with lowering marketing cost and creating awareness for our brand.
And on a similar note, we improved the relation between cars sold and the amount of sessions needed strongly by 65% in Q4 year-over-year, also resulting in higher marketing efficiency. On top of that, we decreased the auto inventory days by 25% compared to Q4 '21 as we expect higher turns to lead to higher gross profit per unit realized. And on the merchant side, we clearly outperformed the decline in the used car market, the additional units in our profitable segment. They all have a direct positive effect on group GP and therefore, on the overall bottom line.
Lastly, on a group level, we were making very good progress on our OpEx efficiency targets and reduced OpEx by 14% in Q4 '22 compared to the year before. We have made great progress with analyzing our cost base in detail, separating smartly between not-needed costs and key investments. And continued progress in all of the above and that framework in all of those input factors, that will lead the way to our target of adjusted EBITDA breakeven in Q4 of this year.
In general, we could not be more about -- more excited about our business in '23. Our full team is extremely motivated and hard-working to make this year another success. We will concentrate on increasing our stock turns across the board as higher turns will lead to more growth and satisfied customers. We will start to expand our sell-only network with the aim to make our product even more convenient, and we will work hard to increase seller success in remarketing, transforming our offering into the marketplace with the best possible experience. We will continue our journey of reducing retail unit cost fueled by increasing adoption of our offering while delivering the best possible product experience.
We believe that by now, we have all the building blocks in place to transition the used car market to online sales at a fast pace, pulling more and more transactions onto our platform by providing our customers with constant great value. The launch, scaling and success of our retail offering under the Autohero brand was an important latest addition to our product strategy as it enables us to sell to end consumers directly. Together with our sell-only, remarketing and auto1.com wholesale offering, we believe we have now completed a powerful trade system that is able to generate strong network effects and high market share growth. Now over to Markus.
Thanks, Christian. 2022 continues to be a year of progress, which we made significant investments to further develop terrific experiences for our consumer and dealer customers, while also laying the financial foundations for sustainable profitability and growth in the years and decades to come. On a full year basis, we grew units by 8.9% to just under 650,000 units despite the volume declines in the European used car market. Revenue increased by 36.8% year-on-year to EUR 6.5 billion. The higher revenue growth consequence of the higher ASPs in the middle of the year as well as the substantial increase in the higher ASP retail units relative to merchant units over 2022.
In total, gross profit increased by 13% to 488 million in 2023, with Autohero unit growth and strong GPU expansion driving an increase in full year GPU improvements to 740 sets, helped by a doubling of Autohero financing GPU year-on-year on an internal and external relevant basis. Gross profit in this area where we expect to see continued progress as we improve GPU in both merchant and retail over the coming year. While our review on our purchase volumes in the current quarter is cautious by ongoing volatility, we already see positive GPU improvements in the current quarter. We ended the full year adjusted EBITDA with a loss of 165 million -- 166 million, reflecting, on the one hand, the significant investments in brand and brand marketing, technology and products as well as the in-house refurbishment. And on the other hand, a significant improved cost of discipline in the latter half of the year.
We were able to keep our adjusted EBITDA broadly flat quarter-on-quarter and significantly improved year-on-year. While the pricing and market volatility in Q4 led to a 40 million decline in merchant gross profit, the GPU improvements in retail, but our overall gross profit to a quarter-on-quarter decline of EUR 9 million. At the same time, improved discipline in payroll, marketing and OpEx, meant that the overall adjusted EBITDA remained broadly flat quarter-on-quarter. We believe that the changes we have made to pay all in OpEx are sustainable and will continue. Together with our improvements to GPU, we believe this positions our platform not only for near-term breakeven but also much greater profitability over the long term.
On the balance sheet side, '22 was a year where in the context of the volatile financing market, we made significant strides both in optimizing our capital structure, particularly on consumer loans, but also reaping the benefit of the robust inventory finance structure we set up in 2021. We ended the year with 542 million of cash as well as EUR 356 million of unused commitments under our unique EUR 1 billion rated nonrecourse inventory financing facility, providing us with a strong base to fund our ongoing investments through cash flow breakeven. In Q4, we also extended this nonrecourse rated ABS facility to 2025 of substantially the same conditions, reflecting the benefit of the sale this stable financing structure.
Gross inventory increased to EUR 618 million over the course of '23 primarily driven by the growth of our retail business as well as slowing sales in Q4, which still well below our peak inventory of 753 million in Q1. Over 2022, we increased our consumer loan portfolio to EUR 186 million. The result of building a unique consumer loan product in Germany and Austria that integrates our technology, product and financing capabilities to allow customers to receive an instantly approved and tailored view on their loan. We continue to invest in the consumer loan product as we believe it will allow us to create superior retail margins over time, particularly once the loan portfolio matures to the point of being able to access the public securitization markets.
In Q2 2022, we're able to securitize our consumer loans through an initial EUR 150 million nonrecourse structure and is the first step in generating substantially higher finance GPUs for consumer clients. And in Q4 of this year, we expanded that facility to EUR 250 million to accommodate the ongoing growth of this portfolio.
In terms of CapEx, we invested around EUR 30 million on CapEx in 2022, primarily in refurbishment centers and our glass delivery jobs. We expect to invest around EUR 50 million in 2023 and orally as we complete the investment in refurbishment centers as well as in expanding our purchased branch network. By the end of Q4 2023, we expect to have over 90% of our cars produced internally so that thereafter, CapEx should reduce significantly as the initial build-out will have been mostly completed.
On guidance, our overall focus in 2023 is to achieve breakeven on an adjusted EBITDA basis by Q4 of this year, while also investing in our products and services to capitalize on our market-leading position and set the foundation for long-term ongoing sustainable and profitable growth. Specifically, we expect to sell between 65,000 and 70,000 units in Autohero, driven by our focus on improving our unit economics in the retail segment. We further expect to sell around 590,000 units in merchant, but back-end loaded as selling customer expectations will take time to adjust to the new reality of used car prices.
In total, we expect to sell between 625,000 and 690,000 units for the full year 2023. Overall, we expect ASPs and Autohero to be broadly flat and slightly down in merchant as the price decline from Q4 appear to have stabilized. We expect gross profit of between 500 million to 550 million for the full year, which reflects the expected expansion to EUR 1,500 retail GPU by Q4 of this year. And the normalization of merchant GPUs that we are currently seeing in Q1. Lastly, we expect to generate an EBITDA loss of between EUR 60 million and EUR 90 million for the full year, with Q4 expected to be breakeven to positive.
With that, I'd like to open up for questions.
[Operator Instructions]
[Operator Instructions] And we will start with the question from Adam Berlin at UBS.
I've got two questions, if that's okay. The first thing, just sitting back for a minute. When you did the IPO, you were guiding for 4x Autohero units in 2023, versus where they were in 2021, and the new guidance is substantially below that original plan. Could you just summarize what you think are the main reasons why Autohero hasn't grown as fast as you originally planned. How much of this is a choice because you've changed your mind on what the right strategy is for the business? And how much of it has been forced on you by a tough end market and different capital constraints in equity markets more recently.
So that's the first question. And then a much more detailed question number two is, can you just update us on the right interest rates we should be modeling for the ABS financing and for the consumer loans? Both in terms of what the consumers pay and what you have to pay their loan provider now that the LIBOR rates have changed so much.
Thank you, Adam. So I think with regarding to Autohero growth, I think, firstly, I mean, we scaled it a lot until now. So we gotten to this group of the largest dealers out of 200,000 within -- how long is the IPO now like 2 years ago. So I think that's quite a growth track to get there. And I think this is also something that we are quite happy with.
Now in the context of the overall breakeven target, Arturo just needs to it needs to develop its unit economics and it's cost and it's the GPU track a little bit faster then I think if we were in another market on the one hand, public market environment, also, I think, geopolitical market environment. So in that sense, our overall target to have the group breakeven puts a, unit economics constrained on Autohero units. So we always said that we want to hit that Q4 breakeven target. That's also something that we gave at IPO as a target, 4-2-3 or adjusted EBITDA breakeven. And then if you look at those targets, like growth in units and kind of Q4 adjusted EBITDA breakeven, then the question is, what is the target that governs? And for us, at the moment, the target that government is the overall profitability. And because of that, we are putting more focus on the retail unit cost within Arthro and that means that optimizing cost and GPO at the same time, is priority above the sheer growth of auto hero. Nevertheless, after Autohero fulfills the requirements for our -- like from a unit economics perspective and also overall GP3 before headquarter profitability, we plan this to grow a lot because we see a lot of potential. But in general, also, you have to see Autohero online car sales, it's really like 999 right now. So it is 2020, it is 2023, but for online car sales, it's 999. So this will take a little bit until the adoption really pulls off in all of the different markets that we're running this. We're seeing great progress in this but this is also something that will come over time. And I think this is all the different color that I would add to how we position growth versus profitability in Autohero with regard to the overall group's profitability target.
I guess I'll take the second question, although interesting as you're looking at the IPO business plan, as you asked the question, I think we'd also plan the 700 GPU in 2020. So I think clearly, we made that decision following up, I think, from Christian's point, around in terms of what the expectation was, which was higher units, clearly, but I was also expecting at that point, a much lower GPU. And I think we significantly outperformed on that front. I think with respect to the consumer products, we actively manage the extra loan that we are -- the external rate that we're looking to get from consumers. And so I think the right way to think about it is that we're managing at around about 1.5% to 2.5% spread, relative to the external interest rate that we get versus the fully loaded expenses that we're paying, which would include both those expenses as well both the -- both the ABS expenses as well as any hedging costs.
Sorry. So on the ABS expenses, that's just -- what did you say, sorry, I missed what you said.
We are managing it to have a 1.5% to 2.5% spread between the loans that we offer to consumers on a blended basis and our expenses to refinance those loans. So the expenses are not just the ABS but also the hedging costs. So we're managing the spread against the entirety of those costs. the way to think about it is of 1.5%, 2.5% spread. .
And the stocking line?
And the stocking loan, we've never provided sort of the specifics, but it's a sub-200 basis point spread over.
With that, over to Andrew Ross from Barclays.
I have 4 questions, is a bit cheeky. So first one on merchant, which I think you told us a year ago did, EUR 125 million of EBITDA before central costs and it was free cash flow positive in its own right. What are those numbers for '23? Is it still a positive business in its own right? Second one is just to come back on the wholesale pricing decline in Q4. I'm not sure I fully understood why it declined so fast in those weeks. It would be helpful to get a bit more color there and perhaps anything you're seeing around kind of broader retail pricing. And the last one is a quick one. But just to go through the rest cash flow statement you've given us EBITDA and CapEx, but the rest of cash flow would be helpful. And then just to clarify a comment on merchants. Are we now back to low 700s of GPU in Q1?
Thank you Markus, I think you -- like the first question goes to you. Yes. So maybe I'll take questions. I think it was 1 or 3 and 4. Absolutely. Thank you. So we do that. So in terms of the disclosure that we provided in 2021 as at that point time we had said was a one-off disclosure, so we're not providing that for 2022. However, I can say, first of all, yes, this merchant business is a free cash flow positive business. I would say it is broadly on the same level as it was in 2021 with the exception of Q4, where we had the impacts that we had talked about. And yes, I think that addresses, I think, that question.
In terms of cash burn, obviously, we're still finalizing the final full P&L, so you should get that in early April when we release our full annual report. So I think the purpose of this call is to our key KPIs. Having said that and looking where if you want to say, you're kind of steady state relative to the 3 previous quarters, we would expect lease payments to be around $30 million for the full year. So that would come on top of the 166 plus an additional interest and local taxes of around 19 million. And then we talked about the CapEx of 30 million, which should bridge you down to the -- yes, so what that cash has been And then your final question on the GPU was related to merchants. So we are seeing a significant improvement on GPU through to kind of the data at the moment for the first 1.5 months, if you will, of performance of this year. And I think that kind of give us to Christian's point, which was obviously, we -- with EUR 6-7-3 of merchant GPU, but for the Tesla situation, I think we would have ended up well in the range between the 675 million and 800 million that we had always been given, work that we have given in the past, and I think we feel very comfortable that we are well in that range at the moment for Q1.
And with regards to your first and -- second question, actually, to the wholesale pricing, I mean, yes, we were surprised as well.
what our auto pricing technology or auto pricing algorithms have resulted in. So if you think about it in easy terms, then -- I mean we're tracking the marketplaces because we always try to match make between our purchase price on the one hand or a given unit and kind of the likelihood of fast sail of that unit than in merchant -- to our merchant base and then in retail to retail. But what you -- yes, what we saw was just this like overall week-by-week decline. So there was a top week, and that top week was also ahead of the week before. So it was really peaking and then it went down fast. So in other words, merchants only would accept the same cars at lower prices. And that means that the market price is falling. And this was very concentrated in a 5 to 6 weeks period, which is unpleasant to have in its entirety because if it's contrast so quickly, then also your potential for mitigating action. It's just limited because it's very fast. The good thing is also encapsulated then in that quarter to a larger extent. But yes, this is what our system has traded towards, so within those 5 weeks. Since then, it's pretty much stable. So there was like more normal seasonality around the Christmas and New Year's Eve. And in the same way, it has week by week, but on a slower pace as the data shown tracked up over almost a period of 1.5 years. In connection with that, maybe it's also noteworthy that we will soon publish also the AUTO1 market price indicator on a regular base. So that product is nearly done, so which will also deliver more insights across where used car market transaction prices, you are not listing prices, but transaction prices are heading towards. And that graph over the last 5, 6 years looks very similar to the U.S. just with like 1, 2 quarters of offset.
Thank you, Andrew. And we will then clearly of with two interesting questions from who are at Trinity Alps.
Yes, we can I think first question. I think the first question has been addressed. Maybe the second question, some of your token ChatGPT question, but really goes to the idea of leveraging some of these new technologies to drive down OpEx and drive conversion. I was just curious if you guys can put any thought into it, and especially on the customer-facing side, if there's an opportunity to explore those avenues?
Yes. I mean we're indeed experimenting with it in like different versions. For instance, we led it to -- like let it write a radio spot for our sell-only brands. And it was pretty good. I have to sell. So like, I mean, we trained at a couple of radio spots like but with like let's say, junior marketing manager. And yes, it's pretty fast. It was pretty impressive. Also, I think when it comes to translations, it's like a couple of notches better than we have been using so far. So I think there's a lot of interesting applications, but it would be too early to say that we could calculate an OpEx impact of that.
Do you think that something like that can ever replace some of the customer service functionalities that you have? So something more than sort of a typical standard chatbot I would think that it definitely could take over maybe first level support. But -- so like anything that is pretty standard, that would make sense in my view.
And we got some late final questions from Christopher at HSBC.
There we go, you should hear me hopefully. First question on CapEx. Is it fair to say -- I mean, you don't say specifically on this slide, but I would think that maybe 2023 could be peak CapEx. I mean if I'm thinking about all of the, let's say, the percentage of in-house refurbishing that you're targeting in 2023. Like is it -- like would it be crazy to assume that we could see a more material drop off let's say, beyond 2023. I know the year just started, and this is a bit of a mean thing to ask about '24 guidance essentially when you've just given '23. But I mean just to understand how you could think about the developing CapEx that would be interesting. And then second question on the development of Autohero marketing. I mean it's obviously gone down a lot, it's probably fair to say that the initial numbers on the sort of spend per car are not really relevant as to the ramp-up. But maybe there is a bit of color you can give as to how you expect it to develop in 2023.
I'll answer the first question. So yes, secondly, 2003 will be peak CapEx certainly for the next few years because right now, we're still in the catch-up process. So going from around the 50% in-house repurbishment, which represents the bulk of our CapEx expenditure, together with the Autohero trucks, which were another substantial part of our CapEx. And so on the one hand, having caught up by the end of this year, we expect to be at 90% of in-house refurbishment so that actually it will come down significantly thereafter. Obviously, there will be replacement CapEx, but it's very, very different than funding the full rollout of the new facility. So we expect that to come down. And similarly, on the trucks. I think right now, likewise, we feel that we have sufficient both right now and in the pipeline. So that after this year, we would expect that to go down again substantially down and really become more of a replacement kind of equal to depreciation type of CapEx spend thereafter. Maybe Christian, do you want to talk about the
Sure. Sure. Look, on -- on marketing, Christopher, it's definitely right that there was like a buildup phase where we -- if you calculate per unit delivered, then it doesn't make too much sense. Still -- I mean that's the number that we look at now. And this number has been continuously improving over the last couple of quarters, driven by, on the one hand, just smarter performance marketing spend, so similar to what we have been doing in our sell only business, optimize marketing by learning a lot on like the performance of different channels and just understanding the customer journey better and also, at the same time, bringing a lot of improvements on the platform itself. And then on the other hand, there was still is like a high amount of euros going into brand investment, making the brand known, making sure that the brand image is set into the -- into the right direction in the direction that we think the brand should be positioned off or should be positioned in. And that is something where we built up the awareness and the image now to a certain level, and we expect over the course of the year that we can keep the brand there with lower levels of investments, particularly reduced sponsoring activity and also overall just less euros in brand campaigns because we are there. So -- and we think that we can keep it keep it on that awareness level and develop it further from there as we go
And we got one question from Sherri Malek at Royal Bank of Canada.
I was wondering how you're thinking about the trade-off between an acceleration in growth after 2022, '23 versus profits and progressing towards free cash flow breakeven?
Yes. So I mean we are planning free cash flow breakeven in '24. But other than that, we are a growth company at heart. We think that the platform that we built can take a much, much higher market share than it's currently taking. So a multiplication of the market share in the future. And therefore, yes, will not sit there and kind of harvest the millions, but we will invest into making sure that we are transforming this market going forward and in the years to come to online sales as we have laid out since the beginning. That is our core vision. We think this market will transition to online sales, especially in retail, which is laying the growth foundation for that. But at the same time, we also don't want to be in a situation where we need to take external funds from time to time. We think we're now a grown-up company. We're having billions of euros in top line, and that should also enable us to run this company profitable. And this is how we think about it. So we want to hit that point, and then, of course, there will be a real acceleration of growth in the years to come.
And we also had a couple of questions submitted from Naiset Deutsche Bank. And maybe I'll just sweep them out for everybody. One, have the constraints related to logistics eased in 2023? And also, are there any further write-downs, I suspect in the likes of Tesla to come? And then consensus is currently at the upper end for total volume since 2023, in which scenario what AUTO1 reached the upper end of the guidance versus the lower end? And then three, how much is the consumer loan product contributing to Autohero GPU already? And finally, why do you expect flat auto volumes at the lower end of the 2023 outlook?
And you that was a bit quick, Philip. Can you start the first question again?
Have the constraints related to logistics,
Yes. I mean the transport market has gone a little bit better, but we're not yet back at the levels that we were used to be. So it's like that. So there's a little bit of relaxation. So it's not a situation like Q4 anymore, but it's also not this flawless service level that we would like to have. And it's still quite a lot of work to produce to service levels that we currently add. It needs a lot of focus because of the transport company still being under pressure, but it hasn't improved since like since -- in middle of November and especially December. So and then I think the second question was like under what condition we will hit the upper end of the full guidance?
Yes, I think the units guidance volumes.
Yes, the unit guidance. I think one thing is how fast will customers accept the new reality of lower prices for used cars because they have also been getting used to price increases now. I think that's one of the points. So that happens faster than this will be -- it will be easier for us to hit our upper end of the unit trajectory, and then also the general development of the market. I mean, we're seeing not yet a lot of market data until now because that typically takes some time to produce. So we've got the German market data, which is a little bit up January over January, but that's the only one that we got so far. And there's a big question mark on just how will the market develop. If you just look at the sheer volume of 2022, then yes, we should expect a rebound. But these days, it's really quite hard to say. But if markets develop positive, if customers accept the new reality of lower prices faster because at the moment, they're still driven by high listing prices for cars on the classifieds, high asking prices. Also, a lot of dealers did not yet reduced their prices. You can see that also kind of classifieds are quoting that. They have more inventory, but also longer standing times. That should never happen in retail. So if you have more inventory than your standing time needs to be the same. Otherwise, you were -- you're aging your inventory. If you aging your inventory, you will make less profits with those cars in the end. So this is something, I think, where the market is now shifting and that can be also a catalyst for more volume. I think that I would name those to I think -- and then, of course, there's a number of initiatives that we have also in our own hand with respect to conversion factors, kind of the demand pressure that we are able to generate on our wholesale platform so kind of the structure and the size of the network. So there is also a little bit of help expected there to reach the upper end. Yes, but I think all of this has to go into the right direction.
I think this was the question why do we think that at the lower end of guidance, Autohero will stay flat for the year.
Yes. So at the low end, I think the guidance -- the unit guidance not here is anyway quite narrow, right? So we believe that a level that we have been seeing in Q4 is something that we can keep. And that also, at the same time, gives us the opportunity to improve economics further. It could also be that if we are advancing with this, that we will be buying more units for Autohero euro, but I think at the moment, this feels like the right guidance.
And consumer loan, I think this goes to Markus.
Yes. So as you know, we have kind of two portions of where we get financing revenue in Autohero. One is our internal consumer financing and the other is external consumer financing. The internal is effectively the portfolio that we're building in Germany and Austria. And then external represents CapEx similar to those that every dealer gets. At the moment, we're around the EUR 250 per car range, which we've significantly -- I mean double release since over a year ago. So -- and see that continuing to improve. But that's on a blended basis.
So thank you very much, Marcus. And that concludes the earnings call for today. And we will all see you, I guess, latest in early May for the Q1 numbers.
Thank you very much, everyone. Thank you. Take care. Bye-bye.