Deliveroo PLC

LSE:ROO

Earnings Call Transcript 2021-Q3

[William Shu, Founder, CEO & Director]

Hello, everyone. Good morning. Thank you for joining us. I'm Will Shu. I'm the Founder and CEO of Deliveroo.

[Adam Miller, CFO & Director]

And I'm Adam Miller. I'm the CFO here at Deliveroo.

[William Shu, Founder, CEO & Director]

It's great to have you all online for this morning's presentation of our Q3 trading update. As you might expect, this presentation will be pretty short, and we'll move straight into Q&A. So let's get started. In terms of key takeaways from these results, I wanted to highlight a couple of points on performance and then some of the new initiatives we've launched in the quarter. In the third quarter, orders were up 64% year-on-year, and GTV increased by 58% on a constant currency basis. Consumer engagement remained robust despite the easing of lockdown restrictions in many markets. Adam will walk through some of this in the coming slides.Also worth pointing out, there are clearly a number of macroeconomic disruptions at the moment impacting both restaurants and grocers, particularly in the U.K., such as labor shortages and supply chain issues. We're watching these very, very closely. Although we're mindful of these uncertainties, we expect our strong performance to continue for the remainder of the year, and as such, we're increasing our GTV growth guidance to 60% to 70% for the year. In terms of new initiatives, the Deliveroo Plus partnership with the Amazon Prime and Deliveroo Hop are 2 of the initiatives we've launched in the quarter that we will take you through today. We are excited about the potential of both. Let's start with Plus. Towards the end of the quarter, we launched this new partnership with Amazon Prime, allowing all U.K. and Ireland Amazon Prime members to sign up for 1 year of free membership for Deliveroo Plus Silver. Plus Silver gives members access to unlimited free delivery on orders over GBP 25 or EUR 25. We've seen excellent initial traction since this launch with a number of Deliveroo Plus subscribers in the U.K. and Ireland more than doubling in a month. I'll give more color on this later on. At the end of September, we also announced the launch of Deliveroo Hop in partnership with Morrisons. This is our new rapid grocery delivery service, which enables deliveries in as little as 10 minutes from delivery-only sites and sits alongside our existing on-demand grocery service. We are very excited about this, and we'll talk more about this later as well.Now let's turn to our marketplace. I'm sure some of you will recognize the slide from previous presentations. I think it's a really important one to highlight because we are operating a complex 3-sided on-demand marketplace. And we think about each side, our riders, restaurants and grocers and end consumers, all as customers of Deliveroo. And we need to balance the interest of all 3 sides and Deliveroo. But ultimately, the strength of our performance is driven by how well our proposition is working for each of the 3 groups. And throughout this quarter, we have seen continued engagement from each side of the marketplace. Our average monthly active consumers have reduced slightly from 7.8 million in Q2 to 7.5 million in Q3. This reflects typical seasonality we would expect for this time of year. In line with this, our year-on-year growth for monthly active consumers has remained healthy, up 56% on Q3 2020.Rider satisfaction has remained strong at 84% across our global network of over 150,000 riders, and we continue to see encouraging application pipelines and rider retention rates. That being said, we are monitoring the U.K. labor market closely, and we do not expect to be completely immune to labor shortage pressures in Q4. Overall though, we are seeing people continue to choose to work with Deliveroo even with a record number of vacancies across the hospitality and retail industries in the U.K., which I think is testament to the strength of our rider proposition and the flexibility of on-demand work.In Q3, we have further strengthened our proposition to riders. Since 2018, we have provided all riders with free accident and injury insurance from their very first order. In Q3, we went further. We've extended our offer to include earnings support for riders who are unable to work due to illness across several markets. We've also launched a one-off lump sum payment to support riders when they expand their families, enabling riders to take time off following the birth or adoption of a child. And we're continuing to explore extending these enhanced protections to further markets.On the restaurant side, we further increased our great restaurant selection with more than 143,000 partner sites now live on the platform globally. And for grocers, we have continued to rapidly expand the rollout as well now with over 10,000 partner sites live. I do want to just say it's impossible to ignore that our partners are facing a difficult time. They're coming out of lockdowns with the full return of dine-in customers. This is alongside an elevated demand of delivery and against a backdrop of supply chain and labor issues in certain markets. I just want to take this opportunity to thank our restaurant and grocer partners for their continued partnership with Deliveroo. Overall, we are pleased in Q3 to see robust engagement from each side of our marketplace. This led in turn to strong operational performance and continued positive momentum despite reopening effects and tougher year-on-year comps.So with that, I'll hand over to Adam to walk through some of the operational performance slides.

[Adam Miller, CFO & Director]

Thanks, Will, and good morning, everyone. As you've already heard from Will, we delivered a strong operating performance in Q3. Growth in orders remained healthy, up 64% year-on-year, while GTV increased to nearly GBP 1.6 billion, up 58% on a constant currency basis.Given that 2020 was a pretty extraordinary year, we have provided some growth numbers on a 2-year basis. Looking at this 2-year view, orders in Q3 2021 increased by 171%, very similar to the 172% 2-year order growth achieved in H1 2021. This demonstrates that growth momentum has been maintained despite the easing of lockdown restrictions in many markets.Similarly to Q2, the order growth outpaced GTV growth in Q3, with GTV per order decreasing by 4% year-on-year in constant currency. This is primarily a result of average order values continuing to revert towards pre-COVID levels, which we'll delve into more in the segment-specific slides.The sequential reduction in both orders and GTV reflects typical seasonality. Q3 is traditionally our slowest quarter, particularly in our U.K. and Ireland European markets, with people taking summer vacations and spending more time outdoors. Again, I'll touch on this more in the segment-specific slides.Let's now look at consumer engagement. This is comprised of average monthly active consumers on the platform and average monthly order frequency during Q3. Although the number of average monthly active consumers was lower than Q2, this again reflects typical seasonality. Year-on-year growth in average monthly active consumers continue to be healthy, up 56% versus Q3 2020. This increase in monthly active consumers on the platform relative to last year continues to be the biggest driver of our 2021 growth.Despite most markets operating free from lockdown restrictions in Q3, monthly average order frequency has remained stable at 3.3x per month. As you can see, this order frequency has remained flat, essentially since the beginning of COVID. However, as a reminder from the half year presentation, the overall frequency of 3.3x is the blended average of frequency across all consumer cohorts.What is interesting about this is that we see that consumers become more engaged with the platform over time. So the more mature the cohort, the higher the monthly order frequency. And we showed you some examples of this at the half year. So it's very encouraging that average monthly order frequency has remained flat throughout a period where we continue to add so many new consumers.Now let's move on to our 2 segments. First, the U.K. and Ireland. The U.K. and Ireland has maintained its growth momentum with consumer engagement remaining robust despite the reopening. At the half year, we said that we had not seen any material impact on consumer engagement post reopening. And this remains the case with order growth remaining very resilient despite lockdowns receding, up 59% year-on-year. We have seen a small sequential reduction in orders as a result of typical seasonality. Alongside order growth, GTV increased by 56% in Q3. The GTV growth is just below order growth, mainly due to a slight decrease in average order value. This is primarily due to party sizes reducing as expected as lockdowns recede.Again, for a more normalized view, we have shown the 2-year growth rates compared to the respective quarters in 2019, which was pre-COVID. When looking at order growth, it was 182% in Q3 this year compared with 184% in Q2 and 170% in Q1, demonstrating that growth momentum has been maintained despite the easing of lockdown restrictions.We have continued to drive consistent growth across all geographic regions, London and the London suburbs, other major U.K. cities, other parts of the U.K. and Ireland, during this period. Our geographic expansion has continued, and we're now serving every U.K. town with a population over 50,000 people alongside our broader portfolio of smaller growth markets.Besides the launch of the Amazon Prime partnership and Deliveroo Hop, in August, we also launched a trial with Boots, one of the U.K.'s best-loved pharmacy brands, offering consumers access to over 400 Boots products on demand. We're really excited about all these initiatives and expect them to help drive growth and strengthen our value proposition for consumers, restaurant and grocery partners and riders. So overall, continued good momentum in the U.K. and Ireland in Q3.Now let's move on to look at our International segment. As we said at the half year, our International segment has been characterized by a number of different lockdown and easing scenarios. Here's a high-level reminder of these dynamics. In Asia and the Middle East, those markets went into lockdowns first and came out first. We've seen very limited impact of reopenings on our businesses there. In Continental Europe, restrictions were initially introduced later and were eased later. We saw a slight to moderate impact from reopenings with the easing coinciding with summer seasonality, which we see each year. And in some markets like Australia and Singapore, they've remained in lockdown much longer, and we're still seeing very strong growth.Now back to reopening impact. In August, we said that reopenings had caused some slowdown in order growth in both France and Italy. At the time, we said it was difficult to disaggregate the reopening impact from seasonality, which traditionally has a greater impact on our European markets. Encouragingly, now that we've moved out of summer, our data shows that the slowdown was largely a result of seasonality rather than easing of restrictions.In Italy, we've seen order volumes revert back towards peak COVID levels. In France, the recovery has been a bit slower with a warmer-than-normal September. We're still moving in the right direction, and we're seeing a gradual recovery there post easing of restrictions and post summer seasonality. This seasonality impact was less pronounced in 2020 due to lockdown restrictions in many of our markets, meaning people were less able to take vacations.In spite of this slowdown, we've continued to see strong year-on-year growth in Q3, with orders up 70% and GTV up 60%. Like in the U.K. and Ireland, the GTV growth is below order growth, primarily due to average order value reverting back towards pre-COVID levels. When looking at the 2-year view, you can see that growth momentum has been maintained with order growth of 162% in Q3 this year compared with 163% in Q2 and 173% in Q1.Outside of the operational performance, the International segment has seen exciting developments in a number of areas. We've strengthened our position across major cities in key markets such as Marseille and Rome, where local consumer value proposition improvements are driving market share gains. We continued the rollout of our on-demand grocery proposition, including major partnerships such as Park N Shop in Hong Kong. We're continuing to roll out our Plus proposition outside of the U.K. and Ireland with penetration as a percentage of overall monthly active users increasing in the markets where we've rolled it out. So overall, we're continuing to see good performance internationally and continuing to develop our CVP to drive future growth. And with that, I'll hand you back to Will to take you through 2 of the exciting new initiatives that we launched in the U.K. and Ireland in Q3.

[William Shu, Founder, CEO & Director]

Thanks, Adam. We wanted to give a bit more color on a couple of our new initiatives. The first one I'll talk about is the launch of our Deliveroo Plus partnership with Amazon Prime. Deliveroo Plus is an important part of our consumer value proposition. It's our consumer subscription program that unlocks unlimited free delivery for a fixed monthly fee. It removes delivery fees as a barrier to ordering. It increases order frequency, and it improves retention.We have 2 tiers of Plus subscription. First one is Plus Gold, which in the U.K. is GBP 7.99 a month, offering free delivery for orders over GBP 10. And then there's Plus Silver, which is in the U.K. GBP 3.49 a month, offering free delivery for orders over GBP 25. We launched Plus Silver earlier this year because we think it represents a great option for families who typically place orders with a higher average order value.On the 15th of September, we launched this new partnership with Amazon Prime allowing all U.K. and Ireland Amazon Prime members to sign up for free Deliveroo Plus Silver membership for a year. Now this offer is available to Amazon's entire U.K. and Ireland Prime base, which opens up a pool of millions of potential new consumers for Deliveroo. And we're very pleased with the initial sign-ups.To give a bit of color on that, so prior to the launch, so let's say the end of August 21, Deliveroo Plus subscribers represented approximately 15% of all UKI monthly active consumers. And as of today, the U.K. Plus subscriber base has more than doubled in size since the launch, so about a month since the launch. And so far, what we're seeing is that existing Deliveroo consumers who use Prime are signing up to Plus as well as attracting an entirely new consumer base who have never used Deliveroo before. And very importantly, we have seen a very, very small percentage of sign-ups coming from existing paying Plus subscribers. So to summarize this, the initial traction has been excellent, and we are excited to continue developing this partnership.All right. So the second of our new initiatives relates to our on-demand grocery business. We have an existing and fast-growing on-demand grocery service. We went into detail about that in our H1 slides. As I also mentioned, we deliver from over 10,000 partner grocery sites globally. And at the end of September, we launched a new rapid grocery delivery service. We call this Deliveroo Hop, and this was done in partnerships with Morrisons.Deliveroo Hop will operate from delivery-only grocery sites. This is great for customers because it enables, a, greater stock accuracy, really eliminates the need for substitutions; b, you can get deliveries in as little as 10 minutes; and then c, there's also access to private label, own label products that customers really, really want.All right. Let's talk about how the model works. So Deliveroo, we're responsible for things like finding the sites, obtaining planning permission, the fit-out of the stores as well as the delivery and the demand gen. We have a lot of relevant experience in many of these factors with our delivery kitchen network additions. Now Morrisons, they will act as a wholesaler to Deliveroo. This allows us to benefit from their private label products but also their established supply chain, which, in my opinion, is the most complex part of this dark store model.Now we launched this less than 3 weeks ago. It's early days, and we'll update more -- we'll update you more on these plans in due course. But what I can say is this. We are super excited about the early days of this. We're super excited about the model. We see it as another tool for us to drive future growth and to strengthen the value prop for consumers, grocery partners and riders.So now I'm going to quickly hand you back to Adam. He's going to finish off with a slide on guidance.

[Adam Miller, CFO & Director]

Thanks, Will. So to conclude, we're going to update you on our guidance for the rest of the year. Following the growth we reported at the first half, we've continued to trade strongly in Q3. While we are mindful of current and potential macroeconomic disruptions and uncertainties, we expect further strong performance for the remainder of the year, which means we are increasing our full year GTV growth guidance up to 60% to 70%, up from the previous guidance of 50% to 60%. We are not making any changes to our gross profit margin guidance.

[William Shu, Founder, CEO & Director]

Thanks, Adam, and thank you again for joining us this morning. Also thank you to the Deliveroo team for executing in a great quarter. Now let's open it up to questions.

[Operator]

[Operator Instructions] Your first question is from the line of Giles Thorne from Jefferies.

[Giles Thorne]

My first question is on Amazon Prime. Presumably, there's a sharing of economics between yourself and Amazon as each platform funnels new subscribers to the other platform. Your changes in guidance today with GTV growth upgraded for no change in unit economics suggest that those economics with Amazon currently represent a cheaper overall customer acquisition cost than other tactics or other channels you've been using. Is that a fair appraisal? And if it is, then when can we expect the next country launch for the partnership?Second question is on Hop. As we all consider the possible strategies for you to compete in the ever more crowded on-demand grocery segment, it feels that your overwhelming advantage here in the U.K. is a lower overall cost to serve. So I'd be interested to hear how you will be bringing that to bear.And then the final one is on the question of rider regulation. The European Commission is due to publish its recommendations for regulation of platform workers in, I think, less than 2 months now. The European Parliament's table, what it wants to see in the directives and the ask from Parliament looks to be an iteration of the current model, not an eradication of the current model. I appreciate most of your business is outside the EU, so less of a consideration, but this is going to frame the debate in Europe around rider regulation perhaps for forever. So Will, I'd be interested to hear what you think the commission will come forward with. That was it.

[William Shu, Founder, CEO & Director]

All right. Thank you, Giles. So let me, I guess, start with the first question, which is our partnership with Amazon Prime. And I want to just say it is actually a -- this is a one-way partnership. So it -- so -- sorry, Amazon Prime customers get free Deliveroo Plus Silver for a year. So it's not a 2-way sort of partnership. I mean I'll just say a few things here. So we launched this on September 15. So it's been about a month. It's -- we've seen excellent traction, as you may have heard on the discussion earlier. So what we said was 15% of UKI monthly active users where Plus subscribers before we launched the program. Since then, we've more than doubled our subscriber base. And what's been, I think, very interesting for us is the vast majority of new Prime subscribers are coming from existing Deliveroo consumers who are new to Plus as well as bringing new customers to Deliveroo for the first time.So to answer your question, we do view it as quite a robust customer acquisition channel. So really excited about it. Also worth pointing out, we've seen very limited impact from existing Plus subscribers canceling and switching to the Prime -- free Prime version, this Plus Silver program. And then to your question, we're not going to get into sort of the commercial arrangement with Amazon, just like we probably wouldn't talk about that with any of our partners. But I will say this. The Plus/Prime unit economics in terms of this partnership, it is very similar to our pay-as-you-go unit economics. And so we feel great about the customer acquisition channel. We feel great about the unit economics, and we are incredibly pleased with the amount of traction we've seen so far on that.Onto your second question, Hop. I think, look, at this point, what really we're trying to figure out is how does this fit into our overall ecosystem with regards to food. We have obviously a very big store pick model, a very big restaurant business. And for us, I think the attraction really is can we work with partners to bring private label products to customers, can we take advantage of a supply chain and do we think that we can operate a dark store model efficiently. I think we're trying to figure all of that stuff out right now. I think I am personally very excited about what we've seen so far from our trial site. But we're now, I think -- Adam, how far in are we now? It's been, what, 4 weeks.

[Adam Miller, CFO & Director]

Yes. It's a little under a month. Yes.

[William Shu, Founder, CEO & Director]

A little under a month, right? And I think what we've seen from a user experience standpoint in terms of retention and frequency has been very encouraging. But I would just highlight this is in one neighborhood with one partner. I wouldn't draw too many conclusions to it. But it is early days, and we're excited.I guess to your point, why do we think we're well positioned? Well, one, we already have partnerships with many, many grocers. And so this is a model that we think over time will be appealing to different grocers around the world. I think secondly, we have a well-established consumer base that associates us with food. And I think that combining these 2 things will drive a lot of value to all sides of our marketplace. So really, really, really excited about that.And then I think on your final question, which is around the European Commission. So in September, the European Parliament voted for platform workers to be considered as employees unless platforms can demonstrate that they're self-employed. Now as I understand it, this is an opinion from the Parliament. It's not formal policy. The European Commission will bring proposals by the end of '21. And then these proposals in turn will be considered by the Parliament and the council to be finalized in late '22. And then the way it works, I think -- sorry, it's getting complicated. But the way it works after that is even if European Commission proposals do not become law, then they have to be considered and implemented by member states subject to their local legislative procedures.So I guess in short, we're at an early stage of this process. We are engaging constructively with EU decision makers. I personally met with decision makers in the commission recently to discuss how self-employed workers can be given more protection without compromising flexible work. Our position is that we want to protect that flexibility that I've always talked about that comes with self-employed work whilst making sure riders can have more protections at the same time. And so the commission is considering a broad range of measures. The majority would have no bearing on our model, and we have been encouraged by their recognition that the positive role that platforms play in local economies and labor markets.

[Operator]

The next question is from the line of Rob Joyce from Goldman Sachs.

[Robert Joyce, Equity Analyst]

Three from me as well. So firstly, just on the guidance. So for the fourth quarter, I think the range is pretty wide, sort of 15% to 45% growth in the fourth quarter, top and bottom of the range. Can you talk us through where you're tracking quarter-to-date and just give us an idea of the major swing factors between hitting the top and bottom of that guide? Second one is just on the Amazon Prime. The numbers you gave, I think, back of the envelope, you've added at least sort of 0.5 million incremental subscribers in the last month in the U.K. Just how much of -- how many orders you're factoring in from those customers in the fourth quarter? Or is that really sort of upside to come beyond 2021?And then the third one, just -- I think it was a question Giles actually asked, but what's the potential of rolling this sort of Amazon partnership out into markets beyond the U.K.?

[Adam Miller, CFO & Director]

Rob, thanks for the questions. Let me take the first question on guidance, and then I'll let Will take the second 2 you on Prime. So look, I think the way we thought about guidance going forward and really Q4, there are still -- although we printed, we think, a really strong Q3, there are still some uncertainties coming out of lockdown. We're not fully out of lockdown in all of our markets yet. I think we noted in the opening remarks some supply chain issues in some of our markets, particularly in the U.K. And then the continuation of sequential AOV declines as we're moving away from -- sorry, we're moving away from kind of COVID lockdowns and really the correlation between orders for larger party sizes. As that reverts we've seen expected AOV declines. But all that said, we are raising guidance again to 60% to 70% full year based on the strong execution we've seen throughout the year from the team, and we do expect that strong execution to continue into Q4.

[Robert Joyce, Equity Analyst]

Are you tracking broadly in line with the midpoint of that, Adam? Is that fair to say?

[Adam Miller, CFO & Director]

So look, I think we're very happy, I think, with the traction we saw in Q3. And as I said, we expect strong execution in Q4. Try to get it out of them.

[Robert Joyce, Equity Analyst]

More to come.

[William Shu, Founder, CEO & Director]

There'll more to come. All right. So let me -- such a public forum, too. All right. Let me get onto the second question. So the Amazon Prime/Plus base, so you did some back-of-the-envelope analysis, and yes, I mean, we had obviously massively increased our subscriber base here. In terms of what we're seeing -- look, we launched the program about a month ago, right? And these sign-ups are continuing to happen, and so it's a little hard for me to know exactly what this will imply for order volume or GTV growth going forward. But what I can say is historically, when we look at the behavior of these consumers, it has certainly been positive for retention and frequency. Now this is a Plus Silver program, where I think the characteristics to Plus Gold are a little bit different given the higher minimum order value, which is why we're really targeting families. So I guess the way I'd answer it is this is a new program. Super excited about the traction so far. But I think this is something where we certainly can see over the long term being quite impactful to our business. And then sorry, Giles, I forgot to answer your question. So Rob, thanks for pointing it out. In terms of more countries for this program, I think -- I don't want to speak for Amazon, but certainly, we were very happy with what we've seen so far, both in terms of sign-ups but also engagement. I think in terms of other countries, we will have those discussions. We think this is a really good benefit for Prime members, and we think it's a great acquisition channel for us.

[Operator]

The next question is from the line of Georgios Pilakoutas from Numis.

[Georgios Alexandre Bela Pilakoutas, Analyst]

I hope you can hear me all right. The first one is you mentioned you've introduced sick payment and parental leave for riders in some of your markets. I guess just interested to hear, what gave you the comfort to do this? There's always kind of been a bit of a balance of offering too many rights and then the kind of risk being classified as employed. So what has kind of allowed you to do this now?Second one, kind of asking the Prime partnership one in a slightly different way. Why was September the right time to launch that in the U.K.? And is that anything to do with your penetration in terms of kind of population coverage? And so therefore, when we're looking at other markets, we should be thinking, okay, once deliveries reached a certain level of scale and penetration that will then allow it to launch these kinds of initiatives.And then a couple for Adam on the numbers. One, just on U.K. growth. I guess can you talk a little bit about what the contribution is from new towns and cities within that? And then a final one on numbers is International AOVs. Just wondering if you could provide a little bit more color in terms of the mix impact from different regions. So kind of is Australia kind of a bit of a tailwind but then [ the strength in ] the Middle East a bit of a headwind? Where are we kind of trending now versus kind of pre-COVID AOVs?

[William Shu, Founder, CEO & Director]

All right. Thank you very much. Let me take the first 2. Adam, you can take the next 2. So I guess the way to explain this is philosophically, we've always wanted to be in a position where we can offer maximum flexibility with enhanced benefits over time. And so in 2018, we actually provided free and automatic accident and injury cover and third-party liability insurance, and we did that for every rider even on their first order. And in the third quarter, we extended this insurance to provide a few different things. So it's not just sickness. It's -- so if you're more -- if you're unwell for more than 7 days, you can't work, you can be paid up to 15 days. We're also, of course, offering one-off payments for riders with new children, either birth or adoption. And I think your question is a very good one, which is does this sort of put us in a different position. How did we get comfortable offering these benefits? I mean there's no simple answer. It's just a lot of discussion with different governmental bodies, different discussions with rider groups. We felt that this was something that riders really wanted. And we thought that this would not put our risk -- our model at risk at all. So we feel very good about it and it's live in the U.K., Ireland, Belgium and Australia. We're starting in France at the end of the year as well. So we intend to roll this out to all markets in the future. But of course, it's really important, as you point out, to tailor this insurance to be compatible with the self-employed model we operate in each market. So we're working with the insurers to develop these new policies. In terms of the Plus/Prime question, so was this -- why September, why the right time to do this. Well, I think there are a few factors, right? Number one, this is a long-term sales cycle, right? This is a long-term partnership cycle. This is not something you can sign up in a few weeks. We have been discussing this with our partners at Amazon for quite a long time, and I think we're both excited about the potential mutual benefits it brings. And these things just take time, right? In terms of the geo coverage side, yes, we do think that at this point, offerings, a program like this, especially when we're targeting families who may live outside the inner cities, this, I think, is a good thing. I don't know that we were hyper deliberate in planning this out in terms of when can we get to maximum geo coverage. But I do think coincidentally, it is quite a big benefit. And so we are even more excited about that given our geo coverage now.And Adam, do you mind taking the last 2, please?

[Adam Miller, CFO & Director]

Yes, happy to. So I think, George, on your first question on U.K. growth and really how much of that is coming from the new towns and cities, I would just reiterate again maybe to start that really pleased with the execution from the U.K. team in expanding population coverage so far this year. As we said, we're now covering all towns and cities with population greater than 50,000. Very similar to what we said in the end of H1, we really view this as planting the seeds for future growth. The vast majority of our U.K. contribution comes -- and GTV comes from towns and cities where we've been operating for more than a year. But this is really kind of, again, setting the stage for us being able to expand into these cities and have them contribute more over time. So very excited about the progress we've made now, which we expect we'll see the benefit of in future periods. I think on your question on AOVs, you talked about International in particular. Let me just kind of set a bit of context on that. Then I'll answer the question specifically. We've reported obviously year-on-year numbers. I think the way we think about it internally is just trying to understand kind of what's been happening broadly and then what we're seeing sequentially. So if we think about the 2 kind of peak COVID lockdown periods, we had kind of Q1 and Q2 last year depending on the market and also Q4 and Q1 this year. And what we've seen since Q1 this year is sequential declines from Q2 to Q1, same thing in Q3 to Q2, which were expected. And as we talked about, the biggest impact we saw from COVID was an increase in the orders from larger party sizes and as lockdowns and restrictions eased and receded and that percentage decline, we've seen an attended decline in AOV, which is expected. I think in International, as you noted, it's a mix of markets in different stages. You have the UAE and Hong Kong who are the furthest away from lockdown restrictions, and sort of they're the furthest along the path of kind of moving back towards pre-COVID levels. I think if you think about Continental Europe, that's -- the lockdown restrictions really were easing tail end of Q2. And so we're not as far past that point. And then in Australia and Singapore, those are markets that are still in various stages of lockdown. But that is the general trend that we're seeing across the business, that sequentially as lockdown ease and those orders for larger party sizes start to recede, we're seeing AOVs come down. Now we haven't seen either of those things revert all the way back yet, either in terms of the percentage of orders of larger party sizes or AOVs revert back to pre-COVID levels. But we are seeing that trend sequentially.

[Operator]

Next question is from the line of Andrew Gwynn from Exane PNB Paribas.

[Andrew Philip Gwynn, Senior Food Researcher & Analyst of Food Retail]

Two questions, if I can. So the first just on the guidance again. Apologies to come back to this. But I think at the half year results, you spoke to 100 million step-up in the marketing and overhead line sequentially. So I'm just wondering if that was still part of the thinking. And by extension, I think consensus is around about 115 million EBITDA loss. So just whether or not you're happy with that.And then the second question, again coming back to the Plus launch. It seems to be sort of relatively low key. I haven't seen much advertisement from Amazon directly. So just wondering if there's a plan going forward to kind of materially step up how vocal you are about this potential by the partnership.

[William Shu, Founder, CEO & Director]

Adam, do you want to take the first question on the guidance? And I'll take the second one.

[Adam Miller, CFO & Director]

Yes, happy to. So I think you're right. I think when we talked about in the half year call, we talked about opportunities for continued growth investments in the second half, and that is unchanged at this point. I think we're excited about those opportunities and haven't changed our thinking on that.Look, on EBITDA, we haven't provided any formal guidance, but I would just say that you have kind of the view on GTV and gross profit margin. And as I said, I think we still continue to see an unchanged opportunity for accelerated growth investments in the second half. Hence, the discussion and color last quarter around the sequential kind of increase in marketing and overhead spend.

[William Shu, Founder, CEO & Director]

Great. Thanks, Adam. On the Plus/Prime question, I think your comment was it feels a bit low key. I don't know that I would agree with that. For about 2 weeks, if you just went to the amazon.co.uk homepage, it's very prominently just sort of up there. We've been the beneficiary of a number of push notifications as well if you have the Amazon app. I would just say this though. This is a long-term partnership we have. And so there'll be a lot more customer acquisition initiatives coming down the pike. I think what we've seen so far with good engagement from both parties is obviously our Plus subscriber base more than doubling. And I think that we'll see more in the fourth quarter of how we can activate this even more. But so far, given the resources that both parties have put in, I would say we're very, very pleased so far. And there's definitely more to come.

[Andrew Philip Gwynn, Senior Food Researcher & Analyst of Food Retail]

So could you just wind back just -- I think earlier, you mentioned the penetration figure on the customer base, the Deliveroo Plus. I just wanted to make sure I heard it right. So is it possible to repeat it?

[William Shu, Founder, CEO & Director]

Yes, absolutely. So as of August 15.

[Adam Miller, CFO & Director]

August.

[William Shu, Founder, CEO & Director]

August 15. That's what I just said. Yes. So as of August 15, our Plus penetration in the U.K. as a percentage of our monthly active consumer base was 15%. And since then, we have more than doubled the Plus subscriber base.

[Operator]

The next question is from the line of Monique Pollard from Citi.

[Monique Pollard, Director]

Three questions from me, if I can. The first question was just whether you could give us the breakdown for 3Q of the U.K., Ireland and International customer numbers rather than just the group numbers, please.The second question was coming back to that question of AOVs in the International segment. Obviously, the decline we saw in the AOV in the International segment in 3Q, much bigger than what we saw in the U.K. I was just trying to understand, is it that in the U.K., partially the sort of return to AOVs pre-lockdown is being offset by grocery initiatives like Hop or maybe the Prime tie-up, focusing more on families and higher AOVs is offsetting some of the headwinds in U.K. and Ireland?And then the final question, just coming back to regulatory developments and rider costs. I know that the historic model in Italy is under appeal. But can you give us an understanding of the cost you might have to pay for the change in the rider definition there historically?

[William Shu, Founder, CEO & Director]

So Monique, it was a little hard to hear the third question. Did you say -- was this in regards to a particular country?

[Monique Pollard, Director]

Yes. That was in regards to Italy, if you could give us some understanding of the costs you may have to pay for that challenge to the historic model there?

[William Shu, Founder, CEO & Director]

Okay. Great. Thanks for your questions. Adam, I think these 3 are probably for you.

[Adam Miller, CFO & Director]

Yes, happy to take them. So yes, I think the first question was just around have we broken out monthly active users beyond just the consolidated number. And we haven't done that at this point in time, so just reporting on the consolidated numbers as of now. I think on the question on AOV, I think the similar question that I think George asked. But this, to me, is really just the difference between looking at year-over-year versus sequential and also some of the mix impacts that we've seen for the portfolio of markets across the International kind of segment. So again, if we think about the year-over-year changes relative to sequential changes, if you just think about the effects that we saw kind of in Q4, Q1 last year and this year and then what's happened in terms of lockdown restrictions easing, the sequential changes in AOV are expected and are things that we plan for as we start to see the number of orders for larger party sizes decrease. And then the international markets year-over-year, again, is just kind of an impact -- or a reflection of where different markets were at different points in time. And we talked about UAE and Hong Kong. We talked about Continental Europe, and then we talked about Australia and Singapore, which just are in different stages of restrictions. But the sequential declines that we're seeing are consistent as things ease out of lockdown in various markets.

[William Shu, Founder, CEO & Director]

Yes. I think just to reiterate the point you were making, there are -- I guess, overall, obviously, we've seen different peaks of COVID in different markets. But in many markets, we have seen, I call it, twin peaks during the height of lockdowns, right? And so if you take Europe and the U.K., really that sort of Q2 2020 and again in Q1 '21. And so as Adam said, when we look at it sequentially, it kind of makes more sense. When you look at it year-on-year, it just gets a little more confusing because we've had AOVs increase twice basically, right, due to lockdowns.

[Adam Miller, CFO & Director]

Yes. And then I think the third question on the -- which is on the Italy labor inspection. So there is a labor inspection in February of this year that reclassified all platform workers in this quasi employee status. This is focused on a historical period from 2016 to 2020. As we've said before, we have a collective bargaining agreement with a labor union that's on a historical model only. We disagree with the judgment, are appealing it, but it is still at an early stage in the process. So these are conversations and discussions that take place over months and years. We're at too early a stage to give specific costs. But we have provisioned accordingly for this based on where we are in the process and what we know, but we can't break this out more specifically at this point in time.

[Monique Pollard, Director]

Sorry, just coming back to that question on the AOVs. What I'm also trying to understand though is, is the reason why the U.K. AOV isn't declining as much partly to do with things like the grocery initiatives and how they've been rolled out and the Prime tie-up focusing on family potentially with larger basket sizes?

[William Shu, Founder, CEO & Director]

So I think there might be some of that, Monique, but I think the bigger factor is the timing of the peaks, right? So if you look on it on a sequential basis, you'll see that it increases, drops, increases and drops. And when you compare that year-on-year, it just looks a little bit funny. But I think that's the majority of the explanation.

[Operator]

The next question is from the line of William Woods from Bernstein.

[William Woods, Analyst]

The first one is just on kind of seasonality. Obviously, we've seen growth hold up on a 2-year basis, but customer numbers are down by about 300,000. Why have you been losing those customers? And with the summer seasonality, wouldn't you expect customers to remain but order less frequently?And then the second question is on order economics in the U.K., particularly thinking about the launch into those smaller towns and cities. You mentioned you cover all the towns and cities above 50,000 people. But I've seen kind of moving into smaller towns of maybe 20,000 to 30,000 people. How are you seeing the demand density in those towns? And I suppose linked to the AOV question, are you seeing a reduction in the premiumization of the offer as you move into those smaller towns and cities?

[William Shu, Founder, CEO & Director]

Okay. Sorry, I'm just writing everything down. Thanks for the questions. I think the first question was really on the sequential decline in monthly active users and why that's happened. So I guess maybe worth just bringing up in general, seasonality at Deliveroo. So Q3 is historically our weakest quarter. It has been since I started this business. Many of our largest markets are impacted by European summer seasonality. So that's the U.K., France, Italy, Spain. And I think also when you think about 2020, it was a different time, right? International travel was more restricted. People found it more difficult to vacation. And so if we kind of think about MACs but also orders, I mean, it's not that dissimilar. So if we look at orders in the third quarter, down 4%. If you look at the same number in 2019, orders were also down 4%, right? So it's exactly the same trend that we've always seen. This is completely expected. Now I can't obviously speak for competitors, but at least that's how we see our business operating. And in terms of the monthly active user trend, we always see a seasonal decline in Q3 as well. And just -- we just -- I guess this is, I guess, a pretty obvious point, but just worth clarifying our definition, which is monthly active users, our users active in the last month, right? So not active in the last 12 months. What we also see on the monthly active user side is as we -- as people come back from holiday and they're in the cities again, we generally see a significant increase towards the tail end of Q3 as people come in. And this year is no different than before. So really, when we compare our business to years past outside of 2020, this is exactly what we've always seen, and this is exactly what we anticipated. Then I think you had 2 other questions -- maybe it's one question, which is small towns, kind of towns less than 50,000, some towns 20,000, 30,000. And sort of what's been the traction there? And also, how does AOV look in those towns? So I would say a few different things. So one is, yes, we're excited about the launch of a number of these markets. I think the number we really focus on is population density. And in the U.K., 50,000, 40,000 people towns can have sufficient population density to really make this model work, the logistics model. So we're excited about this. I think we're also excited because we have a lot of grocery partnerships that have units in these smaller towns, and this is just something that people couldn't get before. So that, I think, is an area that we're very focused on. In terms of what AOV trends are like in these small towns, I don't know off the top of my head. To be honest, I don't know exactly how the smallest towns trend towards versus, say, the bigger cities. But what we do see in smaller towns and suburbs, and I'm generalizing that because these aren't necessarily the newest towns we launched, this is more of a general point, is party sizes tend to be bigger in those areas. And so although maybe London food prices are higher, you actually have more single people in London ordering than you do perhaps families. And as you step outside, let's say, outside of the M25 and outside of the Southeast, you actually see AOV be quite resilient in these small towns and suburban areas, where it's mostly families ordering. I hope that's helpful.

[Operator]

The next question is from the line of Adrien de Saint from Bank of America.

[Adrien de Saint Hilaire, VP & Head of Media Research]

Yes. I've got 2 short-term questions and maybe one long-term question. So Will, you mentioned that in 2019, orders were down 4% Q3 versus Q2, but then they rebounded about, I think, 21% in Q4. Should we assume that '21 will be similar to 2019 in that respect? Secondly, Adam, when you discussed AOV trends, I wasn't quite sure whether you gave any sort of indication about Q4. Could you clarify whether you expect further sequential decline in AOV as markets further reopen? Or would you expect that it stabilizes and perhaps there's a difference here between the U.K. and International? And then thirdly, more of a longer-term question. In August, you've signed a trial partnership with Boots. Can you discuss the opportunity that you see to deliver other products than just restaurants or grocery? Perhaps what's the addressable market there? And what's the potential for you to unlock that in the near term?

[William Shu, Founder, CEO & Director]

Thanks for the questions. Adam, do you mind taking the first 2? And I'll chime in on the third.

[Adam Miller, CFO & Director]

Yes. So thanks, Adrien. I think the first question was just on what happens typically in Q4. And I think you're correct that as we accelerate out of Q3, which is, as Will said, traditionally kind of our softest quarter, Q4, sequentially, we see an increase in terms of overall kind of volume on the platform as we move out of the pronounced summer seasonality periods. And our expectation this year in Q4 is certainly that will continue. I think on the AOV trends, I think you're right. I mean it's the same thing that we've seen, again, kind of sequentially Q2 versus Q1 this year as well as Q2 -- Q3 on Q2 as essentially restrictions ease, that leads to orders for larger party sizes starting to revert back towards pre-COVID levels. And at least what we've modeled in, in our expectations is that, that reversion will continue in Q4. And so that's not typically what you see in terms of the AOV trends from Q3 to Q4. You're not seeing a decline, but this is obviously a unique time coming out of COVID, and that's what we're expecting going forward. Will, do you want to take the Boots question?

[William Shu, Founder, CEO & Director]

Sure. So thanks for the question on that. We launched a partnership in August with Boots, the U.K.'s largest pharmacy, offering about 300 to 500 SKUs, depending on the unit. And there's been -- I mean the traction has been very, very good, I guess, is the short of it. But it is a small trial. We want to expand the range because 300 to 500 SKUs is on the low end of any convenience offering. We want to expand sites to see what traction looks like in more areas.I don't know if I would read too much into -- is this a completely different vertical for us. We think about Boots, as I'd call it, food adjacent, right? There's obviously a lot of convenience items in there, but it's also a vendor that's known for food. In fact, I ordered one of the sandwiches for lunch yesterday, which was very good. It was a Chicken Caesar Wrap, actually. So I think that we're trying to figure this out. It's been a very good opportunity, I think, for both parties. And there are things we need to work out on the op side, I think, on both sides. But overall, it's been something that we want to definitely expand the range and get into more sites.

[Operator]

The last question is from the line of Andrew Porteous from ASBC (sic) [ HSBC ].

[Andrew Ian Porteous, Analyst, European Retail]

A couple for me on the grocery side of things. The color on Hop was really interesting. But clearly, if Morrisons is selling to you on a wholesale basis, do you then need to think about effectively taking control of that inventory and retailing? And just wondering what you're thinking about in terms of the gross margins you'll need to sort of support the unit economics and whether you're thinking about any sort of different delivery charges on a Hop basis versus your traditional grocery business.And then just a couple of quick ones. Can you just remind us of the grocery as a percentage of total GTV in Q3 and perhaps where you are in terms of sort of the number of grocery sites in the U.K. now as well?

[William Shu, Founder, CEO & Director]

Thanks for the questions. Let me take the first one. Yes. The Hop model is different, right? We're operating in a site. Morrisons is functioning as a wholesaler, although I would say our relationship is much tighter than just a pure wholesale relationship in general. We work with hundreds of their sites, and we have a very close relationship. I think your question really is around pricing and gross margin and ultimately what level sort of makes sense for this business line to become profitable. And I think the short answer is there's a lot of assumptions that go into that, right? There's a lot of assumptions on capacity. There's a lot of assumptions on shrink. There's a lot of assumption on labor efficiency in the warehouse. And there's a lot more software that we want to build as well, although I have to say the team has done an outstanding job building what I think is really world-class integration with Morrisons supply chain already. So it's too early to answer that question is the short of it. And so we'll be testing various mechanisms to see where we think the gross margin levels are. But of course, we're learning as we go on, and it is a different business model with more fixed cost than our traditional business model. And so we're keenly aware of that. But we are very, very optimistic around the consumer trends we've seen so far.Adam, I think there's a second grocery question around penetration.

[Adam Miller, CFO & Director]

Yes, happy to take that. So we did talk about at the half year that our on-demand grocery business represented 7% of total GTV for the group at the end of H1. I think as we've talked about this quarter, we have continued to expand that proposition, both the existing on-demand grocery proposition and the launch of Hop, which Will just talked about. So at the end of H1, we said we had more than 9,000 grocery sites live. We've now said that's more than 10,000 at the end of Q3, and we are very excited about the continued expansion of this business line.

[Operator]

This concludes our question-and-answer session. I would like to turn over to Will Shu for any closing remarks. Please go ahead.

[William Shu, Founder, CEO & Director]

Thank you, operator. Thank you all for the questions. If we did not have time to take your questions, David Hancock and the IR team can follow up later. And I just want to end by thanking the people we work with every day. So our restaurant our grocery partners, we talked about some of the challenges that we're all facing; our riders; our consumers; and then the team at Deliveroo, who again have executed so well this quarter; and then to all of you for your interest. So thank you. Have a great day. Bye-bye.