HelloFresh SE (HFG) Earnings Call Transcript & Summary

Good day, and welcome to the HelloFresh SE Q2 2020 Results Conference Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Mr. Dominik Richter, CEO. Please go ahead, sir.

Good morning, everyone. I'm really excited to share with all of you our second quarter results as well as an update on the business and some of its underlying drivers. Without doubt, Q2 has been a very eventful quarter for us, a quarter in which we needed to react very quickly to the evolving corona crisis across all our markets to deal with really extraordinary circumstances that we faced. Here, a lot of our supplier relationships that we've built over the years, the software tools that power our supply chain and our customer analytics and also the operational excellence that we've put in place, all helped us cope quite well with the additional demand that we've been seeing in the second quarter. We're really proud that we've played such an instrumental part in these times for many families around the globe. Especially, the first half of the quarter was peak corona for us. We had to deal with huge week-over-week increases in demand, the implementation of strict social distancing and additional safety measures. Most of our manufacturing sites were running at pretty much max capacity. We recruited more than 1,000 new colleagues in the short span of 3 weeks, implementing extra shifts where possible and onboarded new sites to our manufacturing network so that we could ease that pressure within weeks and start welcoming new customers again, which further fueled our growth in the latter half of the quarter. Revenue growth then also continued well into the second half of the quarter, even as shelter-in-place orders and the lockdowns relieved -- eased across all our different markets globally. Now due to our customers continuing to spend more time at home, the strong habit formation that has been taking place, especially for new customers, and the absence of long vacations abroad in this summer season, these strong revenue growth trends have even continued into the third quarter. That is why we have once again upward adjusted our revenue and adjusted EBITDA guidance. Christian will spend more time on this in the latter half of this earnings presentation. For now, let me quickly review the highlights of the second quarter before covering in detail our TAM expansion strategy. First of all, the second quarter marked a quarter with very high revenue growth for us with 122% year-on-year growth in constant currency. Secondly, our revenue growth was in large parts driven by elevated order rates from our customers and higher average order value, a sign of the high trust customers have in HelloFresh and on -- and in our brands that we've built up over the last couple of years. Thirdly, we can report a record group adjusted EBITDA of almost EUR 154 million. That constitutes a 16% adjusted EBITDA margin despite a lot of additional investments in safety measures and social distancing, which really shows the profitability potential of our business model at scale and when relying predominantly on our tenured customers. This translated into a free cash flow of over EUR 132 million in Q2 alone, an indication of the strong cash-conversion mechanics that we inherently have in our business model. In addition, we put forward a strong commitment to sustainability, further advancing our lead over traditional supermarkets and food delivery companies by pledging to become fully carbon neutral. That makes us the first meal-kit company to do so and further cements our leadership position in sustainability. Finally, while we delivered a record number of meals in a highly challenging operating environment, we still continued our TAM expansion strategy. We expanded our TAM further by geographic expansion and launched in Denmark as the second market in the Nordics. And we also started internationalizing one of our specialist U.S. brands, our value brand, EveryPlate, which we launched in the Australian market towards the end of the second quarter after we successfully scaled that up over the last 2 years in the U.S. That's really a testament to the fact that we never think of our total addressable market as static, but rather continue to build out products for additional customer segments, internationalize our specialist brands and scale to more international markets to maintain our long-term growth momentum. Before I talk more about that TAM expansion strategy, let me briefly thank our frontline workers who have produced almost 150 million meals for our customers in a single quarter. In the span of a few short days, we ramped up our production capabilities massively while dealing with high demands week-over-week and with also high demands from health authorities and governmental bodies all around the world. Thanks to their herculean effort, we could maintain production at very high levels and let millions of customers and families benefit from HelloFresh during the time of crisis. We also engaged in giving back by increasing our donations to local food banks massively and by providing additional incentives paid to our frontline workers. Tactically, week-over-week, we also increased volumes where possible and engaged in closing more long-term facilities to include in our overall manufacturing network. Now let's talk about our TAM and how that has expanded, both as a result of COVID and our ongoing TAM expansion efforts. First of all, in normal times, a typical household eats about 4 out of 7 dinners at home by spending about 50% of his budget on food at home and 50% of his budget on food consumed out of home. With COVID setting in and increased share of people working from home, and obviously, not going out, customers are actually eating a lot more meals at home, often both dinner and lunch. So rather than 4 dinners at home, a lot of customers have been eating 6 to 7 dinners at home, plus lunches in addition. That just made our TAM expand naturally and made our TAM bigger by at least 30% if you look at food wallet spending, which, post COVID, is supposed to be around 2/3 at home and only 1/3 out of home. We think that is something that some of these trends will remain in the future. And so for example, one of the studies that we quote here is from Piper Sandler and actually sees an increase of 30% from 50% to about 2/3 of the food spend shifting to home. In addition, e-commerce adoption has also been pulled forward in a very short amount of time. What would have taken usually 3 years happened in 3 months and with very little customer acquisition costs. A lot of the consumers that have come to us for the first time have started to build habits, which would have taken a lot longer under normal circumstances and with more competing off-line options. One of the stats that we often look at is how long does it take for customers to cook their first 5 meals with us. And so just that time frame has compressed enormously during COVID, and that is something that we could also see for new customers with higher order rates that this type of habit formation has taken place. Both of these developments, more food being consumed at home and higher adoption in a short and compressed span of time, just means that our TAM has gotten significantly bigger, a great opportunity for a player like us who has the best product offering and the best service levels of all meal-kit players to benefit disproportionately. In order to deal with that massive TAM expansion we're going after and to relieve ourselves from some of the most acute capacity constraints that we had in some parts of the second quarter, we will open 2 new large sites in the U.S. and the U.K. in the remainder of 2020. This is on top of some of the tactical measures and short-term facilities that we have also onboarded already and should allow us to deal with the expected volume growth towards the beginning of 2021. Our U.S. site is in Newnan, Georgia. We've taken possession of that as of last week, and we are planning to ramp up that site towards the end of September. We're going to be providing about 750 new jobs, so onboarding a lot of colleagues in the process, and that hiring process has already started. The second large site that we will be onboarding towards the end of the year is another very successful market for us, which is the U.K. In the U.K., we've seen great market share gains over the last quarter again, and hence, now really want to build out capacity to pull away from all of the existing competition. That site is about the same size as our U.S. site. And also here, we're creating over 600 new jobs by the end of the year, making us one of the largest employers in that region. Now with capacity constraints eased at least by end of the year, we can really focus on our step-by-step TAM expansion again. Like I said, while dealing with record volumes in our existing markets in the second quarter, we still pushed forward for our incremental TAM expansion strategy. And so when we talk about TAM expansion, we usually look at geographic expansion, and we look at expanding to new customer segments and to new demographics. Our EveryPlate launch in Australia was an initiative to really increase our TAM in Australia to reach out to new customer segments with a much lower-priced product, strategy that has proven very successful in the U.S. market for us. The rationale behind it is that we can leverage our existing production setup and also utilize large parts of our existing supplier network. And hence, with 2 brands on the same production and supplier network really benefits by addressing different customer segments and taking advantage of the fixed cost assets that we've built up. The second TAM expansion initiative that we followed through in the second quarter was the launch of Denmark, our second market in the Nordics. It's a market that, like Sweden, exhibits all of the right traits for a successful market entry. It's an incremental TAM expansion. We're adding about 3 million to 4 million households in Denmark to our overall opportunity set. And those 3 million to 4 million households have very attractive customer demographics as we could also observe and monitor with our Sweden launch. What already exists in Denmark and in the Nordics more generally is the high product awareness for meal kits. And hence, there is little doubt that consumer adoption for a product like ours will be very, very high. So all in all, we remain very excited about the future of our growth strategy, especially as many of the different levers we have are reinforcing each other. With more meals consumed at home for a sustained period of time, plenty of international expansion opportunities and the internationalization of our specialist U.S. brands, there will be plenty of growth runway for the next years to come. With that, I'm going to hand over to Christian to lead you through our financials and main KPIs and then spend some time on our forward guidance.

Okay, Dominik. Thank you. So let's dive into our financials, starting with the development of our customer base. Overall, we have increased our customer base by 74% year-on-year. In international, we grew our customer base by more than 100% year-on-year and by more than 600,000 sequentially versus Q1. In our U.S. business, where we were more capacity constrained, we had to focus primarily on high-value existing customers to be able to fulfill demand. And it's important to see this in the context of our order growth, which we have on the next page. We've seen, over the last 3 to 4 months, a meaningful increase in our customer engagement. This primarily materializes in a higher average order rate per customer. Especially in the U.S., we have seen order rates increase by almost 1 full additional order per customer per quarter from 3.6 to 4.5 orders, and this drives our group order growth with 103% year-on-year. But we not only see a growth in our order rates, we also have noticed an uptick in our average order value. Average order value has increased from round about EUR 49 to, in Q2, EUR 54. This is driven by customers ordering more meals. So someone who would have taken 3 recipes before may take now 4 recipes per week, and on top of that, also orders more add-ons from us. On top of that, we also give less price incentives to new customers during that period, which, again, has increased our average order value. So when you put all of that together, a strong growth year-on-year in terms of increase in our customer base, each customer on average ordering more often and bigger bundles, and all of that, with less price incentives to newer customers, this results in very strong revenue growth, which you see on Page 11. In the second quarter, we've grown our revenues by 122% on a constant currency basis to EUR 972 million. And each of our 2 operating segments contributed strongly to that growth, as you can see on the next page. The U.S., despite the capacity constraints, with the revenue growth of 110% on a constant currency basis and our international business with a revenue growth of 138% on a constant currency basis. Now, next, let's talk a little bit about our profitability development, and let's start with our contribution margin. We were very focused on implementing early-on extensive social-distancing measures and extra safety procedures in our fulfillment centers to protect our frontline colleagues. This has a certain impact on productivity. In addition, we wanted to reward our direct labor colleagues for their critical contribution through temporary extra pay. As a consequence, our fulfillment expenses as a percentage of revenue are circa 2.9 percentage points higher than the same period in 2019, and our contribution margin is correspondingly lower with 26.2%. Some of these extra costs will likely stick with us throughout the year. So for Q3, you should expect the contribution margin broadly in line with what we delivered in the second quarter, which means, on a relative basis versus 2019, less negative differential given that Q3 is typically our lowest contribution margin quarter, but still, we will sit somewhat lower than we will do without COVID. Let me now discuss the development of our marketing expenses. Our marketing expenses as a percentage of revenue were down 12.6 percentage points to 8.5%. In fact, even though we more than doubled revenues, our absolute marketing spend decreased in the second quarter versus the same period last year. This is partly due to the fact that given our capacity constraints in certain markets, especially in the U.S., we dialed back on new customer acquisitions during part of Q2. But also, we realized very low customer acquisition costs in the second quarter, even meaningfully lower than the already very decent levels we achieved over the 4 preceding quarters. Let me also say a few sentences about the development of our G&A line. As you know, we've invested quite meaningfully during 2017 through 2019 into the further buildup of our centralized tech teams and into the buildup of our infrastructure. Now this year, you clearly see fixed cost leverage coming through. While we've grown G&A by only 11% year-on-year, we grew our revenues with 123% over the same period, and this means that we effectively halved G&A as a percentage of revenues to 3.3% in the second quarter. And all of this flows directly to the bottom line. We quadrupled our EBITDA margin to 15.8% in the second quarter 2020, and this while operating in an environment which was not easy and whilst more than doubling our business at the same time. Also, by the way, our adjusted EBIT margins, so after depreciation, amortization, is at a very strong 14.6% in the second quarter. Our very strong margin profile cuts across our entire group. Both our segments delivered an EBITDA margin of more than 15% in the second quarter. So to summarize, we have continued to generate both very strong growth and very healthy profitability across our entire business throughout Q2 but also for the full first half of the year. But not just that, we also have generated very strong cash flows. As you see on Page 18, we have generated, in Q2 alone, operating free cash flow of EUR 149 million and free cash flow of EUR 132 million, so even quite a bit above the already very decent level we delivered in the first quarter of this year with EUR 111 million. This means we have further improved already a strong balance sheet. In the first half alone, we added more than EUR 240 million of organically generated free cash flow to our cash position. This means we're not just the fastest-growing European Internet company at scale but also well on track to provide you with the most attractive free cash flow yield in Europe. On top of the organically generated free cash flow, we also took advantage of benign capital market in May and issued a small convertible bond of EUR 175 million. All of that together has increased our cash position at the end of the first half to EUR 612 million. This allows us to continue investing in our global business and taking advantage of growth opportunities as they arise, while we maintain capital discipline and a very strong balance sheet. Lastly, let me conclude by commenting on our updated full year 2020 guidance. And some of you may ask what actually has changed since we last updated our guidance 5 or 6 weeks ago. From our perspective, 3 crucial data points, really. The first one is summer seasonality. Given that we're now more than halfway through the summer holidays period in most of our markets, we have seen that there is, also this year, some summer seasonality, but that this seasonality pattern is definitely much less pronounced than what we would experience in a normal year for us. Secondly, development of the overall COVID situation. Infection rates in some of our markets are actually increasing again, and the lifting of lockdown measures and overall back to normalcy takes longer than we have originally anticipated. And thirdly, and Dominik had alluded to that already at the beginning of our presentation, the behavior of our customers, including the ones that we gained during the COVID period. Overall, customer behavior trends so far are somewhat better than what we had seen previously and that applies to both order rate as well as the retention profile. As a consequence of these 3 factors, we have upgraded our full year constant currency revenue guidance again from previously 50% to 70% to now 75% to 95%. By the way, if the U.S. dollar stays at a somewhat weaker level versus last year, so round about the level where it is right now, our euro reported growth may end up a couple of percentage points below the constant currency growth that we typically guide on. For Q3, with respect to our revenue growth, you should indicatively expect that we will grow close to around 100%. We've also updated our EBITDA margin guidance from previously 8% to 10% to now 9% to 11%. And for Q3 specifically, given that Q3 is typically our lowest-margin quarter, you should assume a single-digit EBITDA margin but possibly above 5%. With that, we are looking forward to your questions.

[Operator Instructions] We will now take our first question from Shaked Atia from Morgan Stanley.

Great. Two for me. First of all, on the capacity side. I'm just trying to understand in terms of revenue, how much additional capacity do you expect to have in the U.S. with the new fulfillment center? And on a group level, what is the maximum capacity you expect to have by year-end? And second of all, on the increased guidance. Just trying to understand how much of the guidance increase is driven by new customer additions versus existing customers? And what are your underlying assumptions for how order value and frequency should trend in the second half?

So the new site in the U.S. should -- is our fourth large site in the U.S. and should correspondingly give us the potential to increase revenues by about 25% from the current baseline. In addition to the fourth site in the U.S., we're looking at another large site in the U.S. that we want to conclude moving in at some point next year. And there is also a couple of smaller satellite sites that we have onboarded, which should also give us a couple of extra percentage growth. But the 4 sites that I talked about previously in the presentation should give us about 25% revenue growth potential alone.

Shaked, on your second question, in terms of customer growth, how we see that pan out during the rest of the year, for Q3 specifically, I would assume a slight increase in our overall customer number and that applies to both of our segments, and then a further uptick in Q4. In terms of average order value, you should expect that to sit above the level where it has been in the pre-COVID period. So let's say, if EUR 48 or so AOV is the baseline, we will sit above that, but somewhat below what you've seen in Q2, given that also as we move into the back-to-school period in September, we will increasingly work with price incentives again. So from an AOV perspective, ballpark, round about EUR 50 is, I think, a good planning assumption for Q3, certainly. And then in terms of order frequency, despite Q3 typically being the quarter with our lowest order frequency, given the more favorable summer seasonality that we see this year, you should assume that Q3, definitely, this year is above in terms of order frequency compared to last year and also above previous trend line. So round about 4 orders per customer is, indicatively, I think, a good ballpark guideline.

We will now move to our next question from Robert Berg from Berenberg.

Three questions from me, if I can. First, you mentioned a higher retention from Q2 to acquire customers, one of the reasons for the guidance upgrade. How does this impact your marketing plans going forward? Are you still seeing a reduction of the customer acquisition costs as well in Q3 so far now that you've started marketing again more heavily perhaps in some of the new regions? And with the higher retention, do you expect to see the time to breakeven brought forward from where it has been pre-COVID and how that impacts your marketing plans? The second question is a follow-up to the previous one on the new fulfillment center. Just to check, by 25% increase in revenues, I assume you're referring to the Q2 kind of maximum capacity levels as the baseline. And how should we think about the impact of this new site in terms of contribution margin before it fully ramps up through the coming months? And the third on Denmark. Anything you've learned from the launch of a new country kind of in a post-COVID launch? Are you finding it easier or more difficult to ramp up in recent months?

Rob. So in terms of higher retention rates and what impact that has on our marketing plans, first of all, as you rightly concluded, with very low customer acquisition costs and higher retention rates, obviously, our payback periods come down massively. So if you look at some of the customer acquisition costs that we've seen in Q2 and now also in the beginning of Q3 and pair that with the higher order rate, then payback periods are very, very short at the moment, and very often, we're breaking even within a few months -- within a few weeks to months. So that's definitely historically been a very, very benign environment. For our marketing plans, I think, overall, we want to look at it opportunistically. On the one hand side, obviously, with low customer acquisition costs, it makes a lot of sense to get a lot of customers through the door. On the other hand, I think we have a lot of great behavior from existing customers, and we want to make sure that we can serve them with the best products. So I think my short answer is we look at it opportunistically compared to Q2. As a percentage of revenue, it will certainly likely go up again, but it should trade meaningfully below the levels that we've seen pre-COVID. With regards to your second question, yes, so capacity for that new site is somewhere around $500 million per year. That's about the revenue that we can do from that new site. In addition, we'll pursue a staged ramp-up. That means you don't go from 0% to 100% utilization right in the first week, but you onboard new labor, you train new labor, and then over a couple of weeks and months, you actually then reach maximum production capacity. So for us, that's probably going to be over the course of Q4, so that latest by the end of the year, we should be able to have that site fully up and running at 100%. And like I indicated before, there are additional satellite sites that we can spin up in shorter periods of time, like we have already done in Q2 and which we can continue to rely on also in the foreseeable future.

And then, Rob, on your last question on what have we learned from our most recent launches and what the impact there was on -- from COVID on these launches. So our most recent ones, Sweden and Denmark, both of these, there, we were effectively meeting a market environment where the market has been somewhat educated, somewhat educated about our products already by our local competitors, and that's helped us to ramp up our operations and gain demand relatively quickly. And COVID -- the whole COVID situation probably has further aided that very quick ramp-up that we've seen in both of these geographies.

And just one point that was missed on the contribution margin impact of the new fulfillment centers. Is there any material or noticeable drop until you ramp up those sites?

So there will be some impact. However, versus the Q2 baseline, that should be manageable, i.e., if you assume for Q3 a similar contribution margin for the group as in Q2, that's a good assumption. That's baking in already initial ramp-up costs and inefficiencies.

We will now take our next question from Andrew Gwynn from Exane.

Just 2 questions for me. So one, we talked a bit about capacity for this year. I'm just wondering what the plans are for next year? It does strike me that you're going to be using this very, very quickly. And then the other question, just coming back to the marketing initiatives and the return that you've seen so far. Anything that you carry into next year? Any sort of permanent change that you think you could make?

So thanks, Andrew. I mean I alluded to some of our plans that were kind of like hit in full force in 2021 when I talked about TAM expansion. So I definitely think that our growth strategy has a number of legs and doesn't just rely on one -- we're not a one-trick pony. It's not just acquiring new customers. It's definitely also broadening our assortment, internationalizing our specialist brands, reaching out to new customer segments, working on service levels, et cetera. We want to dedicate, actually, our Capital Markets Day, which is going to be towards end of November, beginning of December, to our sort of like longer-term growth strategy. But rest assured that I think thinking about how to put together that growth strategy and which legs to invest in at which point in time is something that is front and center for us, has always been, and I think we're pretty bullish on our long-term opportunity. In terms of return on marketing, I think, number one, a lot of the adoption that has been pulled forward, that was obviously like a gift to us. What we have nevertheless seen is that a lot of the customer analytics, software tools, et cetera, that we have developed have really allowed us to manage that situation in a very granular way. So focusing on the best customers that we have, acquiring only the best customers that we can find, switching on and off of certain channels by the hour to actually meet kind of like overall capacity targets, all of those things have played out fairly nice and, I think, have been a testament to the fact that when it comes to growing our customer base and understanding and dabbling in the different customer acquisition channels, that we've done a good job in building those up to the level that they are at the moment. And so I would assume that some of the higher order rates are sticky because they're due to habit formation. I also would assume that a lot of the data that we feed our internal tools with will allow us to acquire customers cheaper going forward. And so I don't think that the payback that we've seen in Q2 will be the new normal. Not at all. But I also do think that we've made a lot of learnings, and some of those learnings will carry forward into the next year and will allow us to basically have a well-oiled growth engine going forward as well.

We will now move to our next question from Caron, Fabienne from Kepler.

Three questions from my side. The first one, can you comment on the different pattern of the fulfillment cost international regarding the U.S. for the quarter? And the second one, behind your guidance, could you give us a bit of color between U.S. and international? And the last question from my side is that due to COVID, we've seen as well an increase in online grocery for some Southern Europe countries, like Italy and Spain. Does this mean or do you believe that midterm, those are countries that you may be ready to look at because the grocery online penetration is coming maybe to a level which would make the entry to those country interesting for you? Given the size of the country, I guess, it may be interesting target.

Yes. Fabienne, on your first question, so different in terms of our fulfillment expense development across our 2 segments. What I would say there is that in the U.S., we have been also going into the whole COVID situation, somewhat more capacity constraint, i.e., the measures that we had to take in terms of social distancing, by taking out in certain production lines had a more disadvantageous effect certainly on our U.S. business overall than on most of our international markets. The international markets, you see that negative effect quasi offset by overall higher capacity utilization so that our international contribution margin actually on a net-net basis hasn't suffered a lot versus on the U.S. side. You really just see the -- this increased cost impact and lower productivity impact. In terms of our guidance for our 2 segments. So for Q3, certainly, you should assume the, let's say, high 90s to 100% top line growth rate that we put out indicatively for Q3 that applies to both of our segments and, similarly, in terms of the EBITDA guidance that I've given for Q3.

With respect to some of the Southern European markets, I think they're definitely on our long list of opportunities, and we definitely think that it is a supportive development if online groceries and overall e-commerce penetration increases in the aftermath of COVID in countries like Spain and Italy. I think what we have alluded to in the past is that we always look at our long list of opportunities, and then force rank them by attractiveness and sort of like confidence level that we have because we can only pick a couple on that long list of ideas that we have to really ensure best execution possible. And so I think both of the countries that you mentioned as well as others are on the long list of opportunities where we think we can build up a strong business, but they're probably not countries that you will see us go to in the next 12 months. So in short, definitely part of the long list. We want to be there at some point in time. But given that we force rank the opportunities, we'll probably -- you won't see us do that in the next 12 months.

We will now move to our next question from Ali Birkby from Citi.

Three, if I may. Firstly, please, could you outline the operational objectives you have in place to improve productivity and raising optimum utilization rate in your fulfillment centers beyond pre-COVID levels? And secondly, within the mix of new customers, please, could you specify the penetration of reactivated customers in the second quarter and provide the split in conjunction with first-time and referred customers. And then finally, after a stronger working capital benefit in the first quarter, could you please outline the building blocks of working capital in the second?

So when it comes to productivity in our fulfillment center, we had to take a hit given that we had to implement social-distancing measurements and, in the beginning, had to comply with the strictest rules, given that we got different guidances from different health authorities and governmental bodies. We still have social-distancing measures, obviously, in place but have now much, much clearer guidance from the authorities what we need to do and how we should address that. And so going forward, I think the main levers to get better at fulfillment costs, especially when it comes to productivity is, number one, having a higher share of trained staff. When we have week-over-week increases, that means that you have an influx of people that don't know your business very well. And so in the first 3 to 6 weeks, their productivity is lower. So that has already changed. Secondly, some of the additional incentive pay for basically working extra hours and working over a certain minimum hour limit, that is something that usually costs us and where we have already started walking some of that back as we have onboarded and trained up more new labor. And then thirdly, from the social-distancing measures, we basically have walked back a little bit but still have them by maturity in place. And so overall, I think you should see sort of like productivity increasing. At the same time, utilization was very high in the second quarter. So that's a slightly offsetting effect if we're not running at 90% or 100% utilization, that some of that then translates into slightly lower productivity per square feet. So those are some of the dynamics that we actually see in our production setup.

Ali, on your second question, the share of reactivation in terms of our customer gains in Q2. I think that's -- it helped us to take a step back. Before COVID, as you probably know, reactivations have made up round about high 20s percent of our conversions, so customers who joined us. During large part of Q2, given the capacity constraints we saw in some of our markets, we, on purpose, put quite a few of our CRM and reactivation activities on hold. So we kept that in our back pocket. We can effectively, to a large degree, spin our reactivation numbers quite a bit in terms of how actively we reach out to former customers, and basically, ask them whether they would like to try the service again. So we, on purpose, dialed back on that somewhat in the second quarter and kept it in our back pocket. And to a similar extent, that also applies to some of our referral activities, so encouraging existing customers to recommend us to their friends. On your last question on working capital trends. Working capital has been a quite meaningful source of cash inflow in the first half. This is due to the fact that our working capital overall is negative, i.e., when we grow sequentially, we typically see an inflow of working capital. That means for the second half or for Q3 specifically, you should assume something like a modest cash outflow from working capital. And then that's largely offset in Q4 then again by a certain cash inflow, depending a little bit what exactly our volumes towards the end of December are. So for H2, you should roughly plan with neutral -- a neutral effect from working capital flows.

We will now move to our next question from Linda Pasquini from Reuters News.

I just -- I was just wondering, you mentioned there are already clear indicators on how customers have started to form new habits, and I was wondering if you could give us more details into that from what you've seen in this weeks of July and the first week of August.

So I think, overall, habit formation usually happens if you do a thing over and over again in a short amount of time. Now what we typically see is that when customers cook something like 5 to 10 meals in their first 30 to 45 days, that's when habits become sticky, and that's when they become really good, long-term customers. So that behavior usually takes longer in non-COVID times. And so with people actually stuck at home, working from home and just spending a lot more time cooking meals at home, that time line for habit formation has actually compressed, and a much larger share of our customers have reached some of those critical milestones where habit formation usually takes place. So that's something that, I think, is a qualitative explanation for habit formation. And if we then look at the quantitative side of things, we could clearly see that some of those new customers have actually higher order rates than new customers previously had, basically confirming the hypothesis around habit formation taking place.

We will now move to our next question from Nizla Naizer from Deutsche Bank.

Great. I have a couple of questions from my end. Firstly, on the international margins of nearly 19%. Within this, could you give us some color on how your developed international markets did? And how much of the margin was sort of reinvested in your growth initiatives such as the EveryPlate launch in Australia and even moving into Denmark? Some color there would be great. Secondly, on competition. I mean we're seeing a lot of the online, I guess, grocery retailers piping up and saying that they're starting their own meal-kit product, given how well the category is doing. Any color you can give us on how you view these new products and why you think HelloFresh may be a better option in this scenario? And lastly, on 2021, I understand that you'll give us a lot more color on the Capital Markets Day. But from the impressive scale that you would reach in 2020, how could we, at this point, look at growth in 2021? Could it still be in the mid-teens sort of range that we see consensus is at the moment? Some color on that would be fantastic.

Thanks, Nizla. On your first point on the development of our margin in our developed business -- businesses within international, we don't break that out, but it's, for some markets, somewhat above the close to 19% where we came out in Q2. For the total international segment, having said that, we are very pleased with really the margin profile across the group, across all of the -- substantially all of the entities that we have in our international business. Now how much profit did we reinvest in our new geographic launches, namely Denmark and our Swedish business that is, for Q2, somewhat below 50 basis points of our total group margin, not just international but total group margin. In terms of -- sorry, on your last point, investment into the EveryPlate ramp-up in Australia, given that this was really just launched, it's -- there's hardly any impact of that on our Q2 numbers.

With regards to additional competition, competition from supermarkets, I think we had seen different efforts of supermarkets over the last couple of years as early as 2013-'14, then with more focus in 2015 and '16. And I think that the overall -- the general situation hasn't changed. What we run is a sort of like fully integrated food manufacturing business, where it's about the design of recipes, understanding customers and the whole food manufacturing parts, includes -- including integration with a large diversified pool of suppliers, doing a lot of value-added work ourselves to cut down to exactly the same sizes that you need to cook a meal. All of those abilities, supermarkets still don't really have. And so when it comes to the quality of the offering, I think it's just nowhere near than what a specialist meal-kit company can do, whether that's ourselves or even some of our smaller competitors. It's just very, very different from running an inventory-bearing retail business rather than a fully integrated brand and manufacturing business. That's just a very, very different ball game. And that's also why I have not -- definitely been not blown away by some of the offerings that we have seen in the market. Finally, with regards to your question around sort of like a growth rate, growth momentum that we foresee for 2021, I think what you quoted mid-teens is certainly something that we're very comfortable with. Like I said, more details when we do our Capital Markets Day. But I think where consensus is at the moment, mid- to high teens, that is something that we feel very comfortable with.

We will now move to our next question from Marcus Diebel from JPMorgan.

Two questions from my side, one for Dominik, one for Christian. For Christian, I think you highlighted a slight increase in active customers in your outlook for the -- your outlook statement. Was that meant for just Q3 or just for H2, so for both quarters? And could you maybe give us maybe an indication also the split between U.S. and international? I mean not exactly, but in terms of the trend, do you expect a slight increase in both divisions? Or would they be different? The other question for Dominik, more strategically, in the past, we talked about ready-made meals, and you highlighted that in Australia, this is growing really well. What's the strategy here? Is that something you will roll out also in your bigger markets? It will be quite interesting if you could talk a bit about the recent success in that business. Yes, those are my 2 questions.

Okay. Thanks, Marcus. So on your first point, in terms of the indicative guidance of the modest increase in active customers, this applies to both Q3 and the total of the second half and also to both of our segments, both U.S. and international. Again, this is indicative if, for example, the whole COVID situation worsens in the U.S. and we would have to -- would be -- given our capacity constraints, we'll have to focus again in terms of where we allocate that capacity, that may tweak the picture. But from where we stand right now, this applies to both of our segments.

With regards to some of the, let's say, quick meal opportunities, so you specifically mentioned ready-made meals. But I would put that into the bucket of basically quick meal solutions. That is an area that, overall, we're very optimistic about. But it definitely has taken a back seat in the last couple of months because that is something where we actually need capacity in our manufacturing facilities to basically prepare those type of meals. So whilst we have seen very good initial trends and a good welcoming of those type of meals in some of our markets when we put them on the menu, we haven't really fully invested into that over the last couple of months as we had to deal with corona, et cetera. So I think in the overall context, that is certainly another avenue that we think will allow us to reach out to new customer segments and will also allow existing customers to order more often when they can also order meals where they don't have to spend 0.5 hour or 45 minutes in the kitchen but meals that they can do in 5 to 15 minutes. That is certainly one of the segments and one of the avenues where we think we could -- we can further broaden our appeal but something that we'll probably now sort of like -- has been delayed by 6 to 12 months.

We will now move to our final question from Christoph Bast from Bankhaus Lampe.

So 2 questions from my side, please. Firstly, a couple of days ago, Marley Spoon announced the launch of their low-budget product, Dinnerly, in Germany. And I think the price point is extremely low at roughly EUR 3 per meal. Could you tell us whether you have any plans to react to this move? And how we should think about the impact on the competitive landscape in Germany here? And secondly, it seems like your Canadian competitor, Goodfood, has successfully expanded their product offering through groceries. And also, yes, Delivery Hero is tapping into this business. Could this also be an option for you as it would increase the basket size, order frequency and client interaction? I think Thomas was recently mentioning something like this in his latest FAZ interview. That's it.

So with regard to your first question on the German market and potential emerging competition, I think, on the German market, our market share is very, very high, among the highest in basically all of our markets. I think the new offering that one of our competitors is now trying to push into the German market, in terms of the price point, that's at about the same price point that actually our core product is, while our core product has much, much better meals, more ingredients, higher-quality ingredients. So I think a side-by-side comparison actually makes us look, with HelloFresh, very, very good. Nonetheless, we obviously always look at what the competition is doing, both in Germany as well as in all international markets. And so we obviously did a similar thing, launched, EveryPlate, in Australia and do think that there are certain segments of customers that we can reach out to with a cheaper product. I think, right now, on the German market, that's probably not the first market that I would have looked at to go even cheaper in terms of the price of meals, given where we already trade with HelloFresh. The other thing that you mentioned, in Canada, our competitor, Goodfood, I think we've had great market share gains during COVID as we kind of like pushed very aggressively our core products. Nonetheless, I think very interesting, what they're doing, and we're definitely monitoring that. I think I would refer to the same answer that I gave before. That is definitely one thing on the long list of opportunities that we're constantly looking at and scoring against each other. We do think there are some opportunities that we have even higher confidence in and that are even more attractive for us. But if you think about a 2-, 3-, 4-, 5-year time horizon, then certainly, better monetization of your existing customer base, which means offering them more products to choose from, and hence, getting a higher share of their food wallet, that is definitely also a very, very attractive opportunity, and we'll be monitoring that player as well as all other players that engage in similar activities very closely to learn as much as possible ourselves.

This concludes today's...

All right. Please go ahead.

Apologies. I would like to hand back to the speakers for any additional or closing remarks.

Yes. Thanks for the extensive set of questions. Looks like there's a lot of interest, especially in our future growth strategy. Thank you all for attending, and we'll be back in a couple of months with the update on the third quarter, and then as indicated with the Capital Markets Day towards the end of the year, most likely beginning of December, to really outline in more detail where we see 2021 and beyond trading and what the major opportunities are that we want to fully invest in. Thank you, everyone. Bye-bye.

Bye-bye.

Ladies and gentlemen, this concludes today's call. Thank you for your participation. You may now disconnect.