Morning, ladies and gentlemen. Today marks not only the day we present our third quarter results for 2021, but it's also the four-year anniversary of our IPO, which we successfully completed back on November 2nd, 2017. We'll also be celebrating 10 years of HelloFresh in November, which makes us still a relatively young business. November at HelloFresh, definitely always a month where a lot of things happen. If you actually go back to our IPO prospectus or to any of our previous capital markets day presentations, you'll find that we've always been remarkably consistent in how we think about our long-term growth strategy.
While we've operated in very different competitive environments since we started, and especially in the last five quarters or 18 months, and a very difficult operating environment, we've never lost sight of how we think about our long-term opportunity and how we can make our growth flywheel actually spin faster. That coherence over time is really rooted in our mission that we wanna change the way people eat forever. Seven out of 10 dinners globally and 5 in 10 dinners in the U.S. are cooked and consumed at home. This really is one of the largest consumer spend categories. We believe that in the last 10 years, we've already fundamentally changed, for millions of our consumers globally, how they actually think about getting a great home-cooked meal on the table, allowing them to eat better, cheaper and more delicious meals than with traditional alternatives.
What we also firmly believe is that what you're currently seeing is the unbundling of the traditional consumer food budget, which, for decades has been going primarily to grocers and restaurants with little change over time. We believe we can play a major role when food budgets are being reallocated in the next couple of years away from traditional grocers. That's also reflected in our vision that says we wanna become the world's leading food solutions group. While we're very bullish on the growth outlook and penetration upside for the meal kit business itself in our existing as well as in yet to be launched new geographies, we've also broadened our product suite quite massively, and the accessible TAM has increased significantly as a result of that.
With the acquisition of Factor late last year and now Youfoodz in Australia in early Q3 or early Q4, we have firmly committed to also becoming a leading player in the ready-to-eat space. We really remain laser focused on the opportunity ahead of us and are committed to play an important role as traditional food budgets get unbundled and reallocated by consumers. Now with that, let's focus more on the near past and I'll walk you through the highlights of our recent third quarter 2021. First of all, we grew revenues 45% to EUR 1.4 billion in the third quarter. These growth rates in a post-COVID environment are higher than almost any other e-commerce player in the U.S. or Europe that we know of and have allowed us to gain further market share.
They're also really a great testament to both our strong repeat purchase business model as well as operational excellence. Secondly, we've seen very positive underlying trends in all of our revenue driving KPIs. Our order rate is actually up 14% compared to pre-pandemic levels and 3% compared to the same quarter last year. Our average order value is up 6% compared to pre-pandemic levels, and again 3% compared to the same period last year, and it now stands at over EUR 51. We've also been very disciplined around our marketing investments, which are fully under control despite the massive revenue outperformance we've seen on top line. Thirdly, we achieved an EBITDA margin of about 5.6% in the seasonally weakest quarter of the year we have. That amounted to about EUR 80 million in EBITDA.
This level of EBITDA, combined with strong cash conversion, has allowed us to remain free cash flow positive despite Q3 having been a 1/4 of significant investments in capacity, in people, and in technology. Finally, we have continued to broaden our set of exciting future growth opportunities with the launch of Italy, the second new geography in 2021, and the completion of our acquisition of Youfoodz. Both of them follow a tried and tested playbook, and while they won't have an immediate huge impact on our group numbers right away, they are investments into capturing our long-term growth opportunity to its fullest. Now speaking about those, TAM expansion, avenues that we've gone down. Italy provides us with an additional 26 million households, so massively increasing our overall total addressable market.
COVID has overall been a huge accelerator for e-commerce in Italy and for online food in particular. We've launched with a tailored Italian cuisine offering a couple of weeks ago with a strong focus on sustainability from day one, and the market has been launched on time in full, and the first customers seem to be really happy and showing great customer satisfaction levels. We're operating that country out of a fulfillment center site near Milan and with a corporate head office in Milan. Now Youfoodz is really something where we double down on the ready meals business. We've told you about the very encouraging results from our Factor75 acquisition in the U.S. last year, which has broadened our U.S. total addressable market, and hence that success has made it an easy decision to actually take over Youfoodz.
We will try to introduce the HelloFresh playbook next year and really see this as a multi-year grower in our overall portfolio. The ready-to-eat space really remains a large area of investments and future growth for us. It's now live in two of our geographies with dedicated standalone offerings, and we have planned over the next couple of years to also bring that to more geographies globally. Now let's come to our revenue build and start at the top with active customers. Active customers increased 39% year-over-year to reach 6.94 million in Q3. That customer growth was driven almost equally in International and the U.S. segment, with both increased at roughly the same rate, 38% and 40%, respectively.
As in previous pre-pandemic years, we've dialed back on customer acquisition during the peak summer months in July and August before scaling advertising and resulting active customers again in September for the back-to-school season, which really marks the start then also for the fourth quarter, which is seasonally a much stronger quarter for us usually. Most of our quantitative and qualitative data points now show us that consumers have resumed a more normal routine after summer, and we believe that the isolated COVID behavior is now mostly behind us, with customers having settled into new routines and picked up learned behaviors again.
If you look over the whole year at our 2021 active customer development across the different quarters, then it really broadly follows the same seasonality pattern you've known from us in the pre-pandemic years, with a big jump in the first quarter, slightly down in the third quarter compared to the first quarter. In Q4, we expect active customers to pick up again. The one exception has really been Q2, which is more technical, where we have carried over a lot of customers from the first quarter to actually have a very high Q2 customer number that we booked in our Q2 results. Our year over two-year active customer growth rate is 164%. Like I said, for Q4, we expect active customers to pick up again.
Now coming to order rates. We've seen strong fundamentals in the third quarter, with order rates increasing both 3% against strong COVID comps last year when nobody had been taking holidays or vacationing, and by even 14% compared to pre-pandemic levels. It now stands at exactly four orders per quarter per active customer in our seasonally weakest quarter. I think that's a great achievement and shows that our investment in the customer value proposition, such as a larger menu, better service levels, and more competitive pricing, have actually been paying off. Average order value has also increased in Q3, even though consumers worldwide have been eating more outside and usually cook less meals at home during summer times. It's been up over 3% year-on-year and 6% compared to pre-pandemic levels to now EUR 51.3 per order.
The main drivers here were increased take up of our HelloFresh Market items and the tendency to take more meals on average as we've broadened the choice for consumers and now have larger menus in basically all of our geographies. Now looking at these continuous incremental improvements in aggregate over a longer time horizon, you can see how powerful these actually have become. At the time of IPO, so four years ago, exactly to the day today, we retained about 20% of net revenue of a cohort after two years and a little less than that after four years.
All of the aforementioned improvements in AOV and order rates now allow us to retain about 40% of net revenue after two years for more recent cohorts, and we project to retain closer to 30% of net revenue after four years, driving an almost 50% improvement in CLV to CAC over the past four years since we IPOed the business. What's even more remarkable is the long-term stability of our cohorts in the outer quarters. Once a customer finds value in our product, it really becomes a key part of their home cooking routine. Such strong and highly predictable revenue retention is really a key feature of our business model and makes it so much more attractive than general e-commerce.
This has also translated into very strong year-over-year net revenue growth of 45%, which puts us significantly ahead in our trajectory of our initial net revenue guidance for the year, which stood at 20%-25%. Both segments contributed to that growth strong growth momentum. With the U.S. growing 51% year-over-year, and our international businesses growing 38% year-over-year. On a two-year comparison, it marks a revenue growth of 221%. That basically means we have more than tripled orders, the number of meals going through our supply chain and ultimately revenues over that two-year period. I think that's no small feat given the complex supply chains we run and had to scale in such difficult operating environments during COVID.
With that, I'll hand over to Christian, who's gonna lead you through the remainder of the earnings call.
Thank you, Dominik. Let me kick off with the discussion of our contribution margin. Our contribution margin in Q3 was 22.5%, down year-on-year by 4.2 percentage points. This is a result of two factors. One, high ingredient expenses by around about 30 basis points. Secondly, higher fulfillment expenses of circa 4 percentage points. Now let's unpack those fulfillment expenses development a bit further. Firstly, and most importantly, our ongoing capacity expansion and ramp up of our recently launched fulfillment centers has a corresponding impact on upfront costs and productivity. This is something that we have discussed over the last couple of months already, and this will stick with us for the next couple of quarters. Impact of that, as discussed in the past, is round about versus last year's benchmark, round about 2 percentage points of margin.
By the way, in line with market, we have also started to increase production wages in certain geographies and are planning to do more of that in Q4 as well as going into Q1 2022. Secondly, we're also back to more normal seasonality, lower fixed cost utilization during peak summer months, July and August, and from a sequential perspective, higher packaging expenses. Thirdly, compared to last year, we also had to cope with a certain level of price inflation in logistics, including through higher fuel surcharges that makes up about 1 percentage point of margin. Lastly, as quite a few of you are aware, there were some special or one-off effects that we encountered in Q3, which we do not single out as special items. These were namely two.
One, there was a storm damage, hurricane damage to our main Factor facility, which impacted business there for two weeks at the beginning of the quarter. Then secondly, there was COVID-related quarantining for a number of colleagues in production in Australia and New Zealand that impacted both costs and production volume for a number of weeks in Q3. When you take both of these events together, they had an EBITDA impact of somewhere mid-teens euro million, round about 1 percentage points margin impact for the quarter. Now for Q4, without those one-off effects and with a somewhat more favorable seasonality, we target getting back closer again towards the 24%-25% contribution margin. Okay, let me now turn to marketing.
We've been very disciplined in the way we deploy our marketing spend, and with that, continue to achieve very strong revenue growth year-over-year in the quarter. We therefore managed to maintain marketing as percentage of revenue at 14.8%, which is meaningfully below the 15%-17% we had guided for at the beginning of the year. As you will recall, it also quite a bit below where we've been sitting prior to the COVID period. Going into 2019, marketing expenses were at 20% and above of revenues, and now for this quarter, we're at 14.8%. This is despite us investing robustly into new customer acquisition in September when our back-to-school period kicked off again. With that, let's have a look now at EBITDA. We delivered an EBITDA of EUR 80 million in the quarter.
This corresponds to an EBITDA margin of 5.6 percentage points, where Q3 seasonally represents our lowest margin quarter. That compression versus last year is driven by the effects that we just discussed, so around about 4 points from contribution margin and 2 points from a normalization in marketing. When you break that down across our two operating segments, you see that both of our segments have contributed pretty equally to that EBITDA, so each of them generated around about EUR 50 million of EBITDA in the quarter. Let me now turn to our cash flows. As previously flagged, we have further stepped up our growth investments this quarter to close to EUR 60 million in Q3 alone in terms of growth CapEx.
This means we are on track for our north of EUR 200 million CapEx investment for the full year. It also means we're well on track with our capacity expansion, where our plan is to more than double by H2 2022 versus the Q3 2020 baseline, as we have discussed in the past. Now, despite this meaningful expansion CapEx, we have again generated positive free cash flow and therefore further increased our cash position to EUR 955 million at quarter end. Just FYI, after quarter end, concurrent with the successful closing of the Youfoodz acquisition in Australia, we have paid AUD 125 million for that acquisition, so around about EUR 82 million that came out of our cash, out of our bank account after quarter end. Let me now conclude with our full year outlook. Most of you have seen that probably yesterday evening already.
Based on our continued strong growth to date through the end of October, we have decided to further increase our constant currency revenue guidance for the year to 57%-62% from previously 45%-55%. This includes roundabout 0.4% growth contribution from the acquisition of Youfoodz that I just alluded to. Given that the transaction closed on October 27, it will be consolidated in our financials for November and December. Our EBITDA margin guidance remains unchanged. After the first nine months, we sit at a 9% EBITDA margin, and for the full year it's reasonable to assume that we end up somewhere in that zone. With that, we're looking forward to your questions.
Ladies and gentlemen, we will now begin our question and answer session. If you have a question for our speakers, please dial zero and one on your telephone keypad now to enter the queue. Once your name has been announced, you can ask a question. If you find your question is answered before it is your turn to speak, you can dial zero and two to cancel your question. If you are using speaker equipment today, please lift the handset before making your selection. One moment please for the first question. The first question is from Miriam [Sheridan] at Morgan Stanley. Your line is now open. Please go ahead.
Oh, great. Morning, everyone, and thanks for taking my questions. Firstly, just on the contribution margin, I'm just wondering if you could just share more color on what makes you confident to get back to that 24%-25% level in Q4, given some of the headwinds that you mentioned that happened during the quarter, and you sort of talked about wage inflation increasing since then as well. If you could just sort of give any more color on that would be great. Also sort of any early expectations for what you're expecting next year. I think at the Q2 results, you sort of talked about margins expecting to improve in the second half, and clawing back some of that 200 basis points of fulfillment costs.
Would you still expect that to be the case, or do you expect some of that pressure to remain in the second half of next year? Finally, if you could just talk a bit about the contribution from the add-ons from Factor and from HelloFresh Market to revenue growth and any sort of color on what you're doing with HelloFresh Market at the moment, and whether you're still on track for the 500 SKUs by the end of the year. Thanks.
Miriam, thank you. It's Christian here. First, on your first question, what makes us comfortable to get towards the 24%-25% contribution margin in Q4, it's really two things. One, as I'd alluded to, there's around about a 1-point margin impact in those Q3 results really from one-off special factors, so that I had alluded to. Hopefully we'll not see the same in Q4. That basically will lift margin hopefully accordingly. Secondly, we typically see in Q4 a somewhat more favorable seasonality. What I had discussed about in terms of lower fixed costs utilization during really two out of the three months in Q3, we typically don't have that in Q4.
Also from packaging perspective, given the temperatures are lower as we enter the winter in most of our markets, that saves us a couple of basis points as well. If you take all of that together, that makes me comfortable that we will get towards that zone. On your next question, with respect to outlook into 2022, we would want to ask you for your patience effectively. That's something we would like to work through with you in a bit more detail at our Capital Markets Day in December. That's really the first time when we want to discuss how we look at 2022 in a bit more detail.
On the contribution of the different growth vectors that we have in the business, I think you alluded to Factor. Factor in the most recent quarter contributed at mid-teens growth to our U.S. segment and mid-single-digit to our overall growth of the group. I think very well on track. We've had two weeks where it didn't contribute fully because we've had some roof damages and had to stop production in some parts. Overall, I think a very, very strong development of Factor, which has also given us the confidence to go ahead with the Youfoodz deal. HelloFresh marketplace, I think is also well on track. The pilot in Benelux has now been extended to about 500 SKUs that we offer to our consumers.
In the U.S., we're now delivering out of four of our fulfillment centers to about 75% of the U.S. population, an assortment of about 100 products. Hopefully as we go into 2022 or by mid-2022, we can also launch that offering in some of our other international geographies. I think overall it's something that we definitely believe will play a big part in our future growth strategy, especially if you look at customer uptake numbers and customer satisfaction numbers. I think it's something where we can deliver value and where we become more relevant to customers to take on a larger part of the overall consumer's food budget.
Great. That's very helpful. Thank you.
The next question is from Fabienne Caron, Kepler. Your line is now open. Please go ahead.
Good morning, everyone. Two quick questions from my side. Can you share with us the weight of customer reactivation in the quarter, in the U.S. and internationally? The second question would be, some of your competitors have been talking about increasing costs for acquiring customers. I was wondering if you could share your views on this point. Thank you.
Thanks, Fabienne. Christian here. On reactivation percentage of total new customers joining us in that quarter, it is at around about the mid-20% that we had discussed in the past. In terms of development of our customer acquisition costs, there we don't see a massive inflation that you had alluded to. We've seen a certain normalization versus the depressed levels in 2020. Overall, we see CACs broadly in line with what we've seen towards the back end of 2019, which we considered as very healthy CACs for us as well. Since then, we've increased average order value, which obviously then lifts our customer lifetime value. We also have a very strong re-trend in both order rates and retention as Dominik has taken us through.
In terms of overall customer profitability, and that includes CACs in that equation, we are actually very upbeat with what we're seeing in our trends.
Okay, thanks a lot.
The next question is from Andrew Gwynn, Exane BNP Paribas. Your line is now open. Please go ahead.
Hi there. Yeah, also, two fairly quick questions. Just firstly, again, back on the operating environment. Obviously we hear a lot about sort of transportation wages and so forth stepping up significantly. You flagged that a bit. I'm just wondering how far through the curve you are on that. Maybe a sort of appendix to that question, which is about unionization we're hearing a lot of in the U.S., whether or not that could be something for us to have in mind as we do our models. Then the second point, you've kind of touched on it a little bit, but just on the improving order size, it obviously has seen pretty significant step up. You know, as you mentioned, things like COVID behavior is somewhat behind us now. Just wondering, what's happening there.
Is that sort of marketplace or something else going on? Thank you very much.
Thanks, Andrew. These were actually four questions if I counted correctly. Let me take the
I'm greedy. Sorry. Yeah.
The first one. In terms of specifically to logistics costs, including wages there, we think the current baseline is on par with what we would expect in the future. Where we expect more to come is both in terms of production wages as I had mentioned, and then also in terms of ingredient price inflation effects that we expect some more to come.
On unionization efforts in the U.S., we obviously respect the choice of our employees whether they want to choose or refuse union membership. I think as a business we obviously prefer an open and direct dialogue with our employees. I think we have a very strong employer value proposition with fairly high satisfaction rates with you know very good insurance levels in the U.S. for our employees with very good salary levels in the U.S. I think we actually have a strong employer value proposition and we prefer to have a direct dialogue with our employees but obviously respect whatever they choose or refuse to do. I'll also take the next one on order rate.
You know, in the short run, order rates can be driven by things like price incentives, by things like sort of like some digital features, making it easier to choose the right meals or to filter meals, et cetera. Those are kind of like small things that can drive up order rates, but really over longer periods of time. Two years, three years, four years. What really structurally moves order rate is if you improve the assortments that customers can choose from. Over that two-year period, we've quite significantly expanded the number of meals that customers can choose from. We've also quite significantly improved the service levels, so the time from order to delivery. Plus we've, on top, you know, launched our marketplace items in many different countries.
I do think that if you compare just one to one, the customer value proposition that we so often talk about as it was two years ago and as it is today, then all of those improvements actually compound and have an impact on the higher order rates of the 14% higher order rates we see today compared to two years ago.
Okay, thanks all.
The next question is from Marcus Diebel of J.P. Morgan. Your line is now open. Please go ahead.
Hi, everyone. Three questions also from my side. The first one is on Slide 9, the revenue retention. Could you just help me to understand how it's built? The 60% retention after Q3, does that include or is it, does it clear up customers who dropped out before? Or does it really say, okay, you have a 60% revenue retention of the entire cohort? Just to be clear on this. Secondly, there's a lot of noise, news flow around ultra-fast delivery networks, Gorillas and Flink and others. Some of them are offering new kits. So far we see very different proposition to what you have.
Do you feel you have to partner with them in the future and would like to use this, I would say, a quickly growing distribution channel as well? Or do you think you still want to run your business more on a kind of like standalone base rather than partnering? The third question, maybe also for Christian. Given the reopening, I mean, there's a lot of obviously speculation what will happen at the beginning of the year. Could you maybe tell us maybe conceptually what your tools are to drive your very confident view on also Q1, Q2 next year? I appreciate you don't want to give guidance, but more conceptually, what is the data that you look at today that makes you comfortable on the H1 performance next year?
That would be quite interesting in terms of active customers. Thank you.
Marcus, attention in slide nine. This is the same view that we had shown back at IPO, and I think in one of the capital markets days, which basically looks at how much revenue do we generate with a starting cohort in the first quarter, and how does that develop over future quarters. That hence includes everyone who has churned, and it also includes reactivations, so customers coming back, even if they haven't been a customer in the quarter prior to measuring it. With regard to your second question, some meal kits in the fast grocery space. In the end, to me, that's the same as the meal kits in, you know, offline grocery or online grocery.
If you look at other players, right, an Ocado, a Kroger in the U.S., a Waitrose, I think in the U.K., Lidl here in Germany. I think meal kits, different grocers have tried offering meal kits in all different sorts of formats and at different price points. I think this is a more comparable offer to what you have traditionally found at grocers. Hence we look at it pretty much in the same way that we looked at competing offers from offline or online supermarkets, which means that I'm sure there is like a portion of customers that actually likes one-off purchases that is happy and content with the small variety, small number of meals to choose from and very little rotation.
I think there is the vast majority of consumers who rather trade much better quality, much better variety and much better affordability actually, for having to wait for three days rather than getting it in 10 minutes. That's really the trade-off. We think the vast majority of customers, because home cooking is something that they usually plan out over many days in advance, will actually prefer, you know, the solution that we have. What you see in fast grocery is pretty much, the same or the equivalent of what you have seen in supermarkets for many, many of the past years.
Okay. Marcus, on your last question, given that you assigned it to me, let me have a go at that. What makes us comfortable about starting 2022, you know, on a strong footing is really a number of points. Number one, the consistently strong behavior we see by our existing customers. Therefore also when you compare the, let's say, the number of customers with which we enter 2022 versus what we had at the end of 2020 makes us quite comfortable that we will start the year strongly. On top of that, Q1 obviously seasonally is always a very favorable quarter to bring in new customers to the service.
In addition, we see very strong reactivation rates as well of former customers who then want to come back to the service. We don't have any indications that would suddenly change from going from Q4 into Q1. More underlying all the points that Dominik had alluded to also in his presentation. I'm constantly improving offering in terms of selection type of recipes and all of that at very attractive price points, especially considering
The price measures that some others have undertaken already.
Perfect. Thanks. Thanks for the answers. Thanks.
The next question is from Adrien de Saint Hilaire, Bank of America. Your line is now open. Please go ahead.
Yes. Good morning, everyone, Dominik, Christian, and the whole team. A few questions for me, please. First of all, how much of the, if I may say, the lost users in Q3 do you think you can regain in Q4? Or perhaps asking it differently, what is your outlook for active consumers for Q4? And perhaps you can explain to me why marketing spending increased so much in Q3 and yet active customers were down. Is it a question of, let's say, seasonality on the investments? Secondly, on the supply chain disruption risks, can you explain to us how this is changing your approach to the business? I've seen that you've made additional investments in last mile delivery in Germany, for example.
Is it something that you're planning to do at a larger scale? Third question, you talked a lot about inflation in your prepared remarks. What do you want to do to offset this, I would say? Do you think you have some room to increase prices for your meal kits? Perhaps if you can discuss the sensitivity that you expect from these potential price increases. Thank you so much for the answers.
Let me start off on active customers and give a bit more context to what I alluded before in the prepared remarks. I think, if you look at seasonality patterns of active customers pre-pandemic, and you always saw a big jump in Q1, sort of like, about the same in Q2, slightly down in Q3, modestly up again in Q4, before another big jump in Q1. I think, this is the same seasonality pattern that we expect for next year, and this is also the seasonality pattern that we have seen this year, with the exception of the second quarter, where we just carried over a lot of customers from late Q1. Where I think, you all know how our business model works.
In the beginning, we have a lot of trial customers, and so a lot of customers that tried the product towards the end of Q1, beginning of Q2. I think this was the sort of like one exceptional thing about the active customer pattern throughout the course of this year. To come back to your question, in Q4, we'll see that following the normal seasonality pattern, so modestly up from the Q3 levels where we have been right now, and then a big step up in the first quarter again as we enter the New Year.
On marketing costs development in Q3. So, a s mentioned, we are very happy with that. Q3 typically is basically a 1/4 where we initially during July, August, dial back somewhat on our marketing activities because of the holiday season and then in September, step it up quite robustly and acquire new customers. Quite a few of them then only get their first delivery in October. Against that backdrop, if you think about the 14.8% of revenues where we ended up, we feel very well. If you put that into a longer term development versus Q3 2019, for example, we are very happy with where we kept our marketing expenses as a percentage of revenues.
Again, that was sitting below the low point of the guidance that we had originally provided, that in the quarter where we typically have higher than full-year average marketing costs as percentage of revenues. Let me just quickly take your other question as well on potential more mid-term development of our supply chain. Again, also keeping in mind basically certain cost developments. There are one, as you had alluded to, we're certainly looking at where it makes sense to further expand our own logistics operations. We have successfully introduced that in Germany, where we're doing almost 20% of our volume already through our own fleet.
We are also testing that in certain markets in our U.S. business, and those tests look quite good to us. On top of that, as you know, we already are doing that through our own fully owned logistic operations in our Benelux markets, as well as in our Australian markets. Outside of logistics, the other area where we, from a, let's say, more midterm perspective, certainly have a lever to mitigate any potential wage pressure is within production. As more we introduce further automation to our fulfillment centers over time, where we have certain plans which we can discuss at some point in the future that will basically expose us less to labor volatility on that front as well.
I think the last question was around the pricing. I think high level, right? We always wanna be very attractive compared to all the other alternatives that consumers have. At the moment it feels like a lot of other companies also still struggling to understand what they want to do. I think with the absolute levels that we see in terms of pricing, I think we offer great affordability to consumers. I think with the payback we have at the margins that we generate on our customer acquisition spend on the customer lifetime value, how much that is being generated with the margins that we have. We feel very content with that. I think overarching, a consumer's choice is always versus alternatives.
We're obviously tracking and monitoring quite accurately what are all the alternatives, so offline supermarkets, restaurants, delivery platforms, et cetera, doing. This is what we benchmark ourselves against and always try to find that price point that really maximizes long-term customer lifetime value rather than short-term protecting margins or boosting margins or decreasing margins. I think we always want to optimize for the long run, and we always want to do what is most attractive to consumers and drives long-term customer lifetime value.
Thank you so much, guys.
The next question is from Clément Genelot, Bryan, Garnier. Your line is now open. Please go ahead.
Good morning, everyone. I've got two questions from my side, if I may. The first one is on the U.S. Just to make sure I understand how we continue to see a stronger performance in the U.S. Would you say that it is because overall larger corporates are still not yet fully reopened, their offices and their employees are still working from home? Or is it just because more globally speaking, U.S. consumers are now starting to really shift from out of home to towards at-home food consumption? My second question is whether on the top line guidance and these centers.
Given the new and higher top line guidance, do you think that you will have enough centers opened to, let's say, fulfill all the orders, maybe upcoming months, I mean, quarters? Or will you need to, let's say, open or extend shifts in some centers? Thanks a lot.
Okay. Let me take a stab at the first one, the U.S. performance. I think the performance differential between our U.S. business and international business is mostly explained by the additional factor. Factor75, which we took over towards the end of last year and has been contributing to this year's numbers. I think outside of Factor75, both international business as well as the U.S. business actually performing mostly in line and at fairly the same growth rates. I definitely do think that a more persistent work from home environment, both in the U.S. as well as in international, is something that is a net positive for us because whenever consumers eat more meals from home, that obviously increases the TAM that you can access and for which you can be relevant.
If work from home routines persist at the same level as they do today, then certainly that is going to be a net positive for us.
Clément, on your question on capacity, whether that suffices for our Q4 guidance, answer is yes. Our available capacity is commensurate with where we guide it to for Q4. Having said that, keep in mind that for the U.S., there is one more big fulfillment center that we will start to ramp up towards the end, at least of Q4, our Phoenix fulfillment center that will go live. There are certain pockets where we are still capacity constrained.
There are constraints, at least two markets within international where we are quasi at full capacity and we'll have to expand further until we can further crank up volume from there. Also in the U.S., for our Ready-to-Eat business, given that very strong growth that we had alluded to earlier, we are quite filled up as well, but we have on our capacity expansion roadmap basically plan to address that in the near future as well.
Thanks.
The next question is from Emily Johnson, Barclays. Your line is now open. Please go ahead.
Hi. I've got three questions, please. The first one is, given we're now in November, why have you maintained such a wide range for your EBITDA margin guidance at this point in the year, particularly for Q4? Can you explain the biggest swing factors within that and quantify some of those cost headwinds that you're facing? Secondly, the average order value increased in the U.S. in Q3 but declined internationally. Can you explain those two contrasting trends? How much of that was attributable to the mix of new versus existing customers in each market, for example? Third question is, can you please provide an update on some of the recently launched and planned geographical launches? How are you progressing in Japan, for example? Thank you.
Emily, thank you. It's Christian here. Firstly on your first question on EBITDA guidance, why have we kept that stable? It's not necessarily that we
think we will end up at either bookends, but we're comfortable to end up within that guidance. That's why we didn't tinker with that. As I had alluded to earlier, for the first nine months we stand at around 9% EBITDA margin and all told, that's the zone where we should expect to sit for the full year as well. Secondly, on the trends in average order value, U.S. versus international. U.S. basically, all that Dominik had said earlier applies, a very strong development in terms of overall basket size and so forth, and that basically helps further our average order value in the U.S.
In international, just keep in mind that this is really a composite of effectively 15 different markets, and therefore there are some mixed effects in here as well. On a like-for-like basis, you also see actually a positive trend there on average order value in most geographies as well.
On the geography launch updates, we've done Norway in June, July. We've now just launched Italy a couple of weeks ago, and we hope that we can launch in Japan and get the first orders to Japanese customers towards mid-December. If not mid-December, then early January.
It's fairly imminent and it's definitely challenging because with the corona in Japan and visa restrictions, et cetera, a lot of our country launch teams have had a hard time actually getting into Japan. It's been a quasi-remote launch. We're nonetheless excited to be able to get some live customer feedback in either December or January for the first time. From then on refine our product to really find product market fit in Japan.
The next question is from William Woods, Bernstein. Your line is now open. Please go ahead.
Hi there. Thanks very much for taking my questions. I've got three. Just on the new facilities, could you give us a view of how many have launched and how many are set to launch? Are you able to quantify how well they are ramping? The second one is discounting in Q3. How did that trend, and how much does that weigh on your contribution margin over the last quarter? The final one, just building on the question on U.S. AOV growth. I think you've just added over a quarter over a euro quarter-on-quarter. How much of that is driven by you raising delivery fees in the U.S. and the mix of Factor within that versus kind of actual organic AOV growth? Thanks.
William, it's Christian here. On your first question in terms of new facility fulfillment center launches, we actually launched in international three alone in that quarter and are still in the process of ramping up the previously launched quite quickly as well, and actually faster than what we had originally planned. With respect to your second question on price incentives, there basically we basically didn't change our tactic versus the previous quarters. We also see in our average order value, so which is up sequentially as well as year-on-year. You see that there's also a big shift in terms of our price incentive policy. Otherwise, you would see that impact on average order value as well.
Then William, sorry, I forgot your third question.
Sorry. The final question was just on U.S. AOV growth. You've added, I think, just over a euro quarter-on-quarter. How much of that is driven by raising delivery fees in the U.S., the mix factor versus actual kind of organic AOV growth?
The vast majority of that is organic AOV growth, i.e. more market items and bigger basket sizes.
Thanks.
The next question is from Sarah Simon, Berenberg. Your line is now open. Please go ahead.
Yes, morning. Most of my questions have been answered, but I've got a couple more. One was on Factor. You obviously talked about kind of getting it in before you started to really integrate the business. Have you started trying to cross-sell Factor into your sort of traditional meal kit customer base in the U.S. yet? Second one, just on prepared food. You obviously did the U.S. deal and the Australian deal. Do you think you've now got the expertise of how to do this so that when it comes to other markets, you can do it organically, or is it more likely you'd look to buy local brands?
Then the third question, just following on from the comments around frequency and basket size, is it reasonable to think that a lot of the customers that drop off in Q3 are, let's call them the very light users? 'Cause that would be another reason for frequency to go up. Because normally the holiday season, even with the strong customer base, you'd expect people to probably dial back a bit. Is some of the, let's say, the sequential improvement more to do with the mix and just losing people who are only ordering one box a quarter or something and just didn't order this quarter? Anything you can say around how that's evolved will be helpful. Thanks.
Sarah, on Factor, we have not had the opportunity to actually cross-sell that into the HelloFresh customer base. As a matter of fact, at Factor, we're right now capacity constrained and we've been scaling that business stand alone. We would love to start doing that and sell that either via our menu or via the HelloFresh Marketplace through our large customer base of HelloFresh and Green Chef in the U.S. But we haven't been able to do that, we just don't have the capacity and that's something that is probably another 6 to 12 months out. Yeah, that's basically what I can say to that. With regard to prepared foods more generally, I think we're building up that expertise at the moment, right?
I think with every launch of a new DC that we're undertaking, so at the moment, right, we're building the second DC for Factor, second large DC for Factor, brand new one, in Colorado. So this is something where now our team is really from end to end involved in understanding that fully and understanding the manufacturing process, et cetera. So that we actually insource more and more of that knowledge and just generate more of that knowledge in-house, so that as we look towards going into other international markets, we always have a choice of either doing it ourselves or working with local competitors. I think the only other thing I would say to that is that sometimes looking at established competitors is just a matter of time to market.
I think when we look at Youfoodz, for example, what we also really like is the fact that it already has an established customer base, has an established manufacturing setup. That just means if we had done that from scratch, would have probably taken us two to three years to get to the stage where Youfoodz is today, at least. Growing it from there. For example, taking over business like Youfoodz, we can now leverage some of the playbooks that we have and rather help us to, you know, 3x revenues from where they are, rather than taking three to five years to get to the revenue where they are today.
With regard to your last questions on frequency and order rates, I think the biggest impact by far is seasonality that you see compared to Q2 compared to what you will see in Q4. That we had like you know worse customers kind of like not ordering that often, anything like that if at all. Small impacts, I think the much bigger impact is the seasonal factor.
Great. Thanks.
The last question for today is from Nizla Naizer, Deutsche Bank. Your line is now open. Please go ahead.
Thanks. I'll keep it brief and only two more questions from my end. Firstly, thank you for sharing the retention update on Slide 9. A question there on how the cohorts since the pandemic began have sort of trended, and if you place them on that chart that you've shared with us, would they sit above the Q4 2019 line? Some update that would be great because we've heard one of your smaller competitors talk about volatility in the customer behavior, you know, in Q3, for example. Just curious to understand if your customers have been a bit more stable in terms of their behavior. Some color there would be great. The second question is on Q3 international growth.
Clearly, it's a mix of multiple countries, but have there been any regions that have grown faster, more, sort of attractively and others that were more challenging? Some color on, you know, who contributed to that growth or which country contributed to that growth would be fantastic. Thank you.
Nicola, on your first question, obviously those cohorts haven't matured here. This is really long-term retention and retention after two years, three years, four years. If you look at the most recent data points and how we project that to proceed as actually those cohorts mature, then it should sit pretty much where you see the cohorts of 2018 and 2019 sitting. This is sort of like where I think it's now the sort of like more long running average that you'll see from us.
Nizla, sorry, just to make sure I got your second question properly. This was whether any of our markets grew exceptionally strong in Q3 or. Sorry, do you mind to repeat that question?
Yeah, Christian, just trying to understand, the mix effect within the international growth, as in were there markets that grew much faster than the average and others that grew slower and why that might have been the case. Just to understand the regional sort of development.
Yeah. I would say we're not talking about massive differences. Across most of our geographies, we've seen very healthy growth. Even if there is a certain mix shift left or right, overall, we basically saw most of our markets in international actually expanding very nicely. There's a bit of, I'd say different seasonality, Australia and New Zealand, for example, on a different seasonality versus other markets. We have a little bit that effect in the mix as well. Otherwise, if you abstract from that, we are actually quite happy and have seen robust growth across most of our geographies within international.
Great. Thank you.
To sum up, I think we've seen some very strong underlying trends in our customer base. I think a lot of what we have done to improve customer value proposition over the last two, three years is really paying off. We are entering or will be entering 2022 on a much higher customer number than we entered 2020, and hence, I think have a positive outlook, especially driven by those strong underlying fundamentals. I think, despite the fact that we obviously go into some periods of heightened uncertainty, I think overall, we're extremely confident with all of the investments that we're making in people, in technology, in capacity, because that's really the basis for us to continue to grow and compound over the next couple of years.
I think, overall, we have a very positive outlook on our trajectory towards the EUR 10 billion mark that we have announced on our last capital markets day. With that, I hope to see you all at our upcoming capital markets day in December. Until then, I hope you have a great day.