Earnings Transcript for RMV.L - Q2 Fiscal Year 2021

Peter Brooks-Johnson: Good morning, everyone. And welcome to the presentation of Rightmove's Results for the First Half of 2021. My name is Pete Brooks-Johnson, and I'm joined by Alison Dolan, our CFO. I'm going to talk through the highlights of the year, Alison is going to go through the more detailed financials. And then, I'm going to spend a little more time giving you more color on the housing market and our strategic developments. In the first half of this year, we've delivered financially and more importantly, strategically, which sets us up for the future. In 2020, COVID-19 upended the lives of everyone across the U.K. And the last time we met, there was much questioning about how long it would take Rightmove to regain its momentum, as the housing market emerged from lockdown. The Rightmove network effect has emerged stronger than ever, as home hunters continue to turn to Rightmove first and customers invest more in our digital solutions. To demonstrate how we've emerged from the shadow of 2020, I've compared our performance versus June 2019. The 2020 numbers are here too. So if you wish, you can calculate that our revenues up 58% on a year ago, but I think that's a level of flattery we don't need. Site traffic in the first six months of the year was, again, a record, with an average of 1.7 billion minutes per month being spent on the platform. The busy property market has created very different dynamics on our agency New Homes customer basis; agency customers are very busy and focused on winning the right to sell properties efficiently. Our agency business continues to grow back from a year ago, with a further 130 branches added in the first half of the year. Many of our New Homes customers are forward sold until early 2022. And, therefore, we've seen a drop in New Homes developments and developers easing back on marketing, as they have less to sell. In total, ARPA is up at £6 on June 2019 with the agency ARPA growth of £107, driven by a good mix of price and product. We've been flattened a little with some of the one-off cost savings in the first half, which Alison will detail later. As promised at the full year, we've been returning the extra cash we built up in 2020 with £128 million returned. And consistent with a long established policy, we are increasing our interim dividend in line with the EPS to £3. Beyond our KPIs, we've made much strategic progress this year. Rightmove's purpose is to make home moving easier in the U.K., will satisfy our purpose and drive growth through innovation. Our strategy has three reinforcing segments. Firstly, we are the place home hunters turn to and return to first. This, in turn, makes us the U.K.'s property platform, which delivers unparalleled ROI for our customers. And then, we are in a unique position to help both customers and home hunters by making the process more digital and also create new revenue opportunities for us. This slide gives you some examples of our strategic delivery in the first part of the year. We continue to innovate in our Hartland of property advertising and tools for property professionals. We generate return for our customers in four distinct ways. We help them save time; we have products which help them actively grow their market share. We help them market more efficiently by using the power of by far the largest audience of home hunters in the U.K., and we help them grow new revenue streams. For example, the secure video viewings platform we introduced last year has saved our agency customers nearly 6000 days in wasted time, arranging and performing viewings in the last year. We continue to innovate and deliver new efficient marketing products to our customers. I preview the advanced development this thing back in February. It launched in April, and we've been really pleased by the performance. It's delivering a lead uplift to developers around 30%. Developers also seem pleased, with it with over 20% of developments now listing using the new product. And to demonstrate the value of this innovation to us, given his trajectory, this product is likely to add £3 million revenue to New Homes next year, during a time when developers are reducing their marketing spend. As we previously talked about, we're experimenting with demystifying online conditional auctions on site, and this is to help agents grow a new revenue stream. The first provider is now live with four more in the process of integration. I'll talk more about the benefits of our unrivaled data asset and Opportunity Manager later. We believe there's much opportunity in making the home-moving process more efficient by being more digital. I'll talk more about the digital tenant journey and what we've learned for our mortgage experiment later. There's much excitement for the future. But this is all built on our strong position today. Before I give a little more color on our strategic position, I'll hand over to Alison to talk more about the numbers in detail.

Alison Dolan: Thanks, Peter. Good morning, everyone. The first half of 2021 shows a pleasing continuation of the growth seen in 2019. As Peter just explained, our revenue product and ARPA growth in the first half of this year is best illustrated by a comparison to 2019, the last full-year of no pandemic related disruption. So compared against the half year to June 2019, revenue has increased by £6 million of 4%, comprising a £4.8 million increase in agency revenues, and a £3.7 million increase in the other operating segment, what we call, our breath businesses, or partially offset by a £2.5 million reduction in New Homes revenues. The agency growth of £4.8 million was driven by strong product purchase and package upgrades on both 2019 and 2020, as agents competed to win vendor mandates in this strong market. In fact, product revenue as a percentage of overall agency revenue has increased by almost five percentage points since 2019, and is, currently, at 57% of total agency revenues. Upgrades to our premium package Optimiser 2020 contributed particularly strongly to revenue growth in the six months of 2021. New Homes revenues continue to reflect the challenges of a forward sold market, where developers are selling off plan, are struggling to build at the pace necessary to meet current demand, and, therefore, have less of a need to market new properties. In other revenues, with the exception of overseas, which continues to be impacted by travel restrictions, each of the third-party advertising, Commercial and Data Services businesses has grown relative to June 2019, and each is up by about £1 million on that date. Mortgages revenue is now £2 million. The product launched late in 2019, but is up by about a quarter of £1 million on June 2020. Year to date, ARPA at the 30th of June was up 8% on 2019. And at £1,163 is our highest ARPA to date. Within that, agency ARPA was up over 11%, an uplift of £107, driven in part, as I mentioned earlier, by product upgrades, particularly to Opti 2020. The period overall saw a big uplift in the purchase of incremental digital products, both as package upgrades and on an ala carte basis. New Homes ARPA was slightly down on 2019 by one and a quarter percent, reflecting the current market. The New Homes market tends to run counter cyclical to the agency market. This is a dynamic we have seen before and seen recovered. But we do expect the current situation to remain consistent for the remainder of 2021 and into the first half of next year. The six-month trailing ARPA has been shown here also as it shows the clear line of continual growth in ARPA since 2019, looking through all of the business disruption of 2020. The charts on this slide show the drivers of ARPA since June 2019. Of the agency 11% uplift, 45% or £49 is due to price increases over the two-year period. Included in this is £9 of ARPA from the six months to June 2021, in which we brought forward a number of price increases, which have been scheduled for later in the year. A further 36% £38 was driven by package upgrades. In this six-month period alone, over 800 customers upgraded to optimize 2020, almost 100 of which upgraded directly from the basic essential package. The average monthly revenue uplift per customer upgrading to Opti 2020 is just under £300 a month, 16% of our independent customer base were on this top package at the end of June, up from 9% in December 2020. And we have continued to upgrade additional customers to our top package throughout July. Included within this £38 is revenue from customers continuing to purchase additional products on an ala carte basis beyond their package thresholds, as they compete to win vendor mandates in a strong market like this one. Revenue from local valuation alert alone, for example, is up 30% since 2019, and feature property has seen a similar uplift. And then finally, customers choosing to take incremental products, in lieu of a price rise on their core subscription contributed a further 19% or £20. It's always a very pleasing outcome when this happens. Conversations that start as a price increase end up with a package upgrade. And we, generally, find that customers retain the additional products that they have selected in this way. We've also shown the drivers of the New Homes ARPA, which are slightly different to agency. New Homes ARPA is down £17 over the same period. What's shown here are the gross movements within ARPA, where we have seen good growth from greater product purchase, in particular, which alone has contributed £42 of ARPA since June 2019. As with agency, the proportion of new home's revenues from product has also increased since 2019 by three percentage points, in this case. Price increases have contributed £24 of ARPA the same period. However, digital marketing campaigns are where we have seen New Homes developers choose to reduce their marketing spend, which has eroded the product and the pricing growth with an 83 pound reduction to give an overall drop of £17. Digital marketing is discretionary spend above subscription listings, and that is typically where we would see the developers cutting spend in a market as strong as the current one. So although it has eroded the gains in product upgrades and pricing, those latter two provide a strong base for growth once the market returns to more normal conditions. Relative to June 2020, operating costs have increased by 5% or £1.8 million to £35.2 million, largely driven by recruitment for our product development and sales teams. As we indicated when we reported 2020's full-year results, this investment is partly in response to the recruitment pause during 2020 and partly to reflect continued investment in product and website innovation. On recruitment, we have been recruiting throughout the first half. But in the heart of lockdown, it proved more difficult to do at the scale and pace that we had been hoping to achieve. And so, the pace of acceleration in headcount costs has been lower than we thought it might be at the beginning of the year. Although, we are now running at the right pace, headcount cost in the first half has been weighted towards the final months, and, therefore, the run rate of costs in the second half will be higher. Once we've completed the recruitment catch up, one in every three of our employees will work in product development, and a further one-third in sales, which is about what one might expect from a digital business such as ours. Technology costs, which include all the costs associated with running both our sites and the business itself, so hosting licenses, traffic costs, etcetera, increased marginally up by north £0.3 million, all associated with the increase in traffic to our sites over the period. No meaningful number of employees have returned to the office yet, which has meant that staff-related cost savings have been available for longer than we expected £1.3 million in this reporting period. We are now reasonably confident that we will have everyone back in the office three days a week from September, and these costs were returned closer to normal. Overall, costs are back to their steady state of just under 25% of revenues, which is where they were in 2019 and where we anticipate we will end the current year. It is also broadly where we expect them to remain for 2022. All of this has resulted in underlying operating profit of £117 million, up 5% on 2019 and an associated operating margin of 78%. As I've mentioned already, the second half of the year, we'll see a higher run rate of costs, reduced savings as more staff returned to the office and, obviously, no repeat of the Val Mildert provision release, so the full-year margin will fall by a few percentage points. We do, however, expected to be ahead of the 71% level we discussed at last year's full-year results, given the strong revenue performance in the first half. On cash, operational cash generation remains strong at 106%, a level similar to last year. All of the cash generated in the half, along with £30 million of cash generated in 2020, was returned to shareholders with the resumption of the share buyback program and the payment of 2020's final dividend in May. We ended the half with £68 million of surplus cash, down from £97 million in December, and much closer to our expected final year balance of about £50 million. As Peter mentioned earlier, we are also announcing a 2021 interim dividend of 3P, which will be paid in October and which will amount to about £26 million. We will also be continuing the share buyback program throughout the second half. On the P&L, we've shown the impact of the various adjustments, share-based payments and the release of the Van Mildert provision, the underlying operating profit of £117 million, and the associated operating margin of 78%, which I referred earlier, exclude the share-based payments but include the provision release. The change in the balance sheet is driven almost entirely by the lower cash balance. Working capital metrics remain consistent during the period, and there has been no increase in bad debt. We continue to pay our suppliers on the same prompt terms, as previously, with average days payable for trade creditors remaining at around 20 days. Now, I mentioned that the full-year results that we would report on one element of ESG at each of our reporting periods, and today, I'll update on the S element - the social element. Rightmove sets out to make our company a great place to work, where people can bring their full selves to work, where they are helped to develop and progress and to feel supported and mentored when they face challenges. So for our management team, social is about creating the right conditions and encouraging the right behaviors to facilitate this, ensuring not just that obstacles for our people to develop are removed, but that opportunities for progression are available and that developmental training and mentoring support are accessible at the right times throughout people's careers with us. During the pandemic supporting people who have struggled with an acute need, but beyond that, ongoing measures to support the well-being of our people have always been important, as has the sense that when our people look around the company, they can see people like them and feel a sense of inclusion. There are three areas of focus for us, therefore, when it comes to employee support - well-being, training and development and gender and ethnicity. During the pandemic, we launched a wellbeing personnel and professional development program called "How We Thrive", giving all employees access to multiple health and wellbeing workshops, support networks and therapeutic coaches. And an arrangement with our insurer has made available confidential private sessions with a mental health coach, which have been accessed by over 20% of our staff in the past six months. Our 16 mental health first aiders, triage mental health issues that are brought to them using non-judgmental conversations either to resolve an issue or to guide those involved towards external support. In appointing them, we have also sought to remove any perceived stigma and provide an inclusive environment, where people can openly seek support. One of the biggest jumps people make in their careers is the move from team member to manager. As Rightmove, this can often mean a move to managing people who were previously peers, which can be quite daunting if done with no training. So management training and ongoing development is a key area of focus for us. Managers are trained across a six-month period on how to provide and receive ongoing feedback and how to have open conversations about performance, pay, development needs and behaviors. Over $5,000 of this training were delivered in 2020 alone. Getting the gender and ethnicity balances right has always been important at Rightmove. But this year, we have actively delayed a number of senior hires, until we could find females who were right for the role, and we will do more of that as we seek to maintain equal female representation to close our gender pay gap. The mean of which is currently at 24% and the median at 33%. It is a gap that we are determined to close. In the meantime, we are proud of our gender balance Board with three non-exec directors from ethnic minority backgrounds, exceeding the Parker reviews recommendations. And then lastly, supporting the communities in which we operate, namely Milton Keynes, London and Newcastle and supporting causes, which are important to us, such as promoting the study of STEM subjects, particularly for girls, has been part of our agenda for the last few years. Of note, this year, we've signed a partnership with generating genius, whereby our employees provide a mentoring network to the participants of the black women into tech program. We awarded a university scholarship for someone to study a STEM subject to degree level. And we donated computers to needy schools in Milton Keynes during the pandemic. There's a lot more to cover, but I hope this has given you a flavor of the social issues that are important to us as a company and the sort of culture we try to foster as Rightmove to ensure that our people are proud of the company they work for. Part of the remuneration for us as executive directors requires that over 90% of our people think that Rightmove is a great place to work. And I'm happy to report that 93% of them thought so in December 2020. We also, of course, want our culture to be a reason for new people to come to work with us. So that's it for me. I will now hand back to Peter to take you through the strategic update.

Peter Brooks-Johnson: Thank you, Alison. I'll start of by briefly covering what happened in the housing market in the first half of the year. The frenetic activity towards the end of last year and early this has now started to manifest in transactions be completed, as you can see in the top chart. In fact, according to HMRC, there were more transactions completed in the first half of 2021, than in any six-month period since before 2006. For many of our agency customers, the high number of transactions at a slightly higher commission rate has enabled them to rebuild their balance sheet after the turbulence of 2020. The big question many have been asking is what will happen now the stamp duty holiday starting to taper in England. As a reminder, in England at the end of June, we saw the stamp duty-free threshold for all transactions being cut from 500,000 pounds to 250,000 pounds. In September, they says relief will be withdrawn. And once again only first time buyers will be stamp duty-free up to 300,000 pounds. It's impossible to determine how much of the market has been driven by the stamp duty holiday. But our proprietary data shown on the bottom left gives cause to cautious optimism posts the withdrawal of the holiday. This graph shows the level of demand, the number of sales being agreed and the number of new listings all indexed versus 2019. The gray line is a forward-looking measure of demand. This is the number of people actively looking to buy in the market. And looking at our most up to date data despite the end of the holiday in July demand is still 30% higher than the same time in 2019. And this demand is feeding through into sales agreed. The orange line is the number of sales we believe are being agreed. And at the moment, that's typically four to six months ahead of the property completion, putting deals agreed now beyond the end of the September deadline. Again, the latest data in July shows that sales agreed are exactly the same as 2019. To give more comfort, we've only seen a slightly high number of transactions fall through this year, which given the significant increase in deals agreed is very positive. In fact, as July has progressed, the fall through rate has returned to the same level as 2019. This is consistent with surveys we've run, which suggests that only around 4% of buyers would run it - would pull out of a transaction if they fail to meet the stamp duty deadline. Most people are moving to find a home which meets their needs. They don't see it as an asset. I spoke back in February about the lack of new stock coming to market. You can see from the tail line this trend has improved a little since then, with the relaxing of lockdown and possibly the return to school. But we're still seeing new listing levels around 10%. The supply and demand imbalance is impacting the market in a few different ways. Prices are being driven up, the average asking price is nearly 7% higher than the start of the year. And the property market is becoming more efficient. Typically only around a half of properties listed will sell. This has risen to nearly 70% recently, and that supporting increased rate of sales being agreed. New agent formation is slower than one might expect at this point in the cycle is, of course, a new agent has to win all the new stock to begin trading. New Homes developers are also experiencing record demand with 20% more development selling out in the first half of this year, compared to 2019. Taking together as you might imagine, it would be advantageous for Rightmove if the end of the stamp duty holiday did call the markets a little bit. So now, I'll talk more about Rightmove. I'll look at each of the following in turn. I'll lead with home hunter, our packaging product sales and the number of customers. I'll then wrap up by looking at some of the innovations we've delivered. So, starting with our lead with home hunters; time spent on Rightmove was up nearly 60% last year. And if we needed a demonstration of the instinctive place that home hunters turned to, we recorded a new busiest day ever on the third of March, the day of the budget, with over 70 million minutes spent on the platform in a single day. You can see from Comscore chart that our market share of time has nudged up despite quite a lot of marketing activity from our competitors. I think it's worth just alighting on the Comscore chart for a moment. Firstly, you'll notice the Comscore has yet to begin tracking booming. We also use similar Web tracking was, perhaps, not as comprehensive. It has the benefit of being more real-time. Similarly, shows booming share of time to be less than 1%. Secondly, you may remember the Comscore methodology change back in February 2019. Comscore have alerted us to another upcoming change, which I expect will reverse some of that step up. So for those of you who follow Comscore, don't be alarmed if you see that happen in the next six months. But all of this adds up to one way we deliver value to our customers - quality leads, which were up over 50% year on year. In order to win sellers, agents are aiming to differentiate themselves from their competition. The strong sales market has focused agents on winning the right to sell a home more effectively and more efficiently. I'm particularly pleased with the sales rate of Optimiser 2020. The graph on the bottom right compares the package numbers of Optimiser 2020 with its popular predecessor Optimiser 2015. Since launch, it has consistently outperformed the previous package, with COVID only slowing it down for a few months. The right hand end of the graph graphically demonstrates the value our customers see in Optimiser 2020 to help them succeed with record numbers of customers upgrading during the first half of this year. And that sales momentum has continued in July. The pie-chart shows which package those customers were upgrading from. And as ever, the majority are upgrading from Optimiser 2015. But nearly a third of upgrading from our middle enhanced package, having seen the benefits of our products and wanting to make the most of their marketing budget. For now, our focus remains on the Optimiser 2020. But enhanced is still playing a very important role in our package leather. And that's one of the design features of our package structure. Customers can upgrade through it as they feel more confident. And as they upgrade, they get even more value. And as Alison discussed, those upgrades can either come as a result of a pricing conversation, or is the result of a customer request. I'm content with how we've structured our pricing conversations this year. And given their success, we've put forward some of the conversations we were intending to have later in the year. So far, we've spoken to around twice as many customers as we had intended to, and we've added some more activity in the second hall to get a head start on the 2020 plan. So moving on to customer numbers demonstrating the underlying value of our proposition to agents in both slower and stronger markets; we've had the lowest number of first half levers since 2014 which has driven our retention rates to the highest it's ever been. I think it's also testament to support we gave customers last year in the depths of lockdown. However, as I mentioned earlier, new agent formation has been slower than might have been expected, as the lack of this stop coming to market is making it very difficult for new agents to start up. In addition, the higher sales efficiency means there are fewer instructions which end up moving agents, meaning there are less second bites of the cherry for agents who miss out on an instruction first time. I think there are a number of existing agents waiting for slightly more stock to come to market before committing to a new branch too. The second half, I expect branch numbers to continue a gradual rise, unless we see a sustained growth in new listing numbers. In New Homes, as you will have heard to say many times, developers are already forward sold well into next year. It's simply a story of reduced supply and increased demand having to work through the system. The number of New Homes listed was 15% down in the first half, compared to the same period in 2019, as the difficulty constructing against the backdrop of social distancing, reduced output. And on the flip side, we've seen over 50% more demand for New Homes. In terms of number of New Homes developer companies advertising with us, that hasn't changed, so I fully expect the number of developments will rise when we move out of this particular supply and demand imbalance. Looking forward for the next six months, I think we'll continue to see a drop in development numbers for most of the second half, as the new home construction start numbers are still down over 20% in the first quarter compared to 2019. And as I showed earlier, we don't see any significant drop in demand. So how are we innovating to help agents and developers succeed in this market? I thought I'd show you one of the products, which our customers are using to win. Our unrivaled knowledge of home hunters allows us to help agents uncover the potential sellers within the pool of potential buyers they already know. And using our data, we can predict when the seller is ready to consider instructing an agent. Rightmove is in a unique position, as two-thirds of the people sending leads have yet to select an agent to sell their home. Opportunity Manager is exclusively available to Optimiser 2020 customers. It uses a real time algorithm to help the agents prioritize their work with a simple color code. In stock constrained markets like this, it's even more important for agents to be able to talk to potential sellers before their competition. Opportunity Manager correctly identified around a third of the properties which came to market in the first six months of the year. Typically, given the agency use the product a month head start over their competition, of course, we can only highlight the opportunity; the agent still has to win the instruction. But to give a sense of scale, that's a commission pool of over £0.5 billion. Looking a little further out, we're going to make home moving easy by helping home hunters be more transaction ready and by being digital. As I've talked about in previous presentations, we've been focused on improving the process of renting the property for the last few years. We believe because of our unique position in the market, with more people searching for rental properties than anywhere else, we're the only company to be able to truly change the rental process for the better from end to end. We've just delivered the first element of the second phase of the plan. We've enabled one click reference ordering from Rightmove Plus. When a tenant has requested an appointment via Rightmove, an agent can request a reference from us via Rightmove Plus, and the property and tenant details are pre-completed in the reference order. The agent can also see the status of all their current references from Rightmove Plus, with the obvious efficiency benefits. The streamlined flow from lead to keys is what we'll continue to focus on in the coming months. There's more integration planned in the second half of the year to enable the tenant and appointment details to feed into the agent CRM systems. And we're also working on the contract part of the process and expect to launch that early next year. The digital insurance flow for tenants who pass referencing has just been launched. You may remember, we've been relying on the telephone up until now to make sure we could learn the basics as fast as possible. We've got a lot to learn around the digital conversion for tenants insurance as a key part of the services helping tenants understand that they're not covered by their landlords insurance. This need for confidence has been backed up by notable number of tenants going through the digital flow and then getting in contact with us directly to purchase. And all of this activity has been supported by rebranding Van Mildert to right move. We think our trusted brand will bring more confidence to tenants in what is clearly a difficult process. Referencing is key to the digital rental journey. It's both the moment when a tenant finally gets a reasonable level of certainty that they will be able to move in. It's also the doorway for a number of new opportunities for us. We know the timing is crucially important when selling ancillary products like contents insurance, broadband and energy. And referencing is the best indicator of a tenant when a tenant wants to buy these products. Not only does referencing generate revenue in its own right, this timing indicator may well give us an advantage when marketing solutions to tenants and landlords. Today for every reference we complete, we generate just over $4 of revenue from tenants and landlords, taking into account the current tax rate. We're early in the journey of scaling this revenue opportunity. So, I wouldn't want you to get carried away just yet. This is the first time we've created a revenue stream from consumers. I see this as becoming meaningful perhaps in a couple of years' time. And we're working on three levers to grow this revenue. Firstly, increasing the number of references we complete. We only process around 7% of references in the market. Referencing is currently a competitive industry with little differentiation. We believe that by continuing our integration and simplification, we will be able to create a best-in-class reference, which is simple for tenant agent, which balances speed and accuracy. In the last year, we've increased our referencing speed by 20% - is pleasing to see that our tax rate for insurance is climbing, and that's indication of our strategy of working hard on offering the right product in the right way. But we can see that we've got more work to do, particularly with our new digital flow. And, finally, we can offer more services to tenants. We're still in the early stages of learning with T.V. and broadband as you can see from the chart. And beyond that, I'm sure you can all think of things that we might be able to successfully offer to tenants. And in doing this, we're not only creating new revenue stream for ourselves, we must keep making her moving easier for tenants and much more efficient for our agents. We're also working on making home moving easier for our buyers. Earlier, I said I'd share some of the learnings for our mortgage experiments. And what we found won't surprise anyone who's been searching for a property and who needs a mortgage. Not only can the process be daunting, hunters tell us it significantly adds the stress of finding your next home. The fundamental reason is that as someone looking for a home, it isn't until very late in the process, you become certain that you'll actually get a mortgage, and, therefore, that you'll actually be able to buy the home you're looking at. If one looks at the current flow shown on the top, you can see that most home movers only really starts to be confident that they can get a mortgage that they point or the point they want to make an offer on the home they've chosen. That's the fourth step along. That's pretty inefficient, also can be pretty upsetting. There's many homeowners will pretty much set their hearts on the property by that stake - was as much activity in making mortgage application more digital, we're in a unique position to bring the mortgage earlier in the home-buying flow. Looking at the flow we're building on the bottom, not only do we believe we can increase the homeowners confidence earlier in the process, probably before the viewing by offering easy, personalized lender-backed decisions in principle, we also think we can remove some wasteful steps too. So far, our work gives us confidence in our approach. By working on our tools, content and placement, we've doubled the number of leads we sent to our lender partner in the last year. And those leads are of better quality with three times as many receiving a lender-issued decision in principle as a year ago. But, again, we'd much to do much to learn in this space. We will continue to focus on the process up to decision in principle, with more innovation in bringing together mortgage and property search tools and joining up helpful content. By doing this, we will increase the number of leads and the conversion rate to decision in principle, which in itself, will lead to more, which is being written. We've got much to do and much ambition. But let's bring it back to 2021 and wrap up with the outlook. The network effects of the ARPA business are stronger than ever with record traffic and leads. At the moment, the property market is busier than it's been for a long time. What's the level of impact to the end of the stamp duty how holiday will have been hard to predict? The indicators we have given cause for optimism the demand we see in the market is founded on a desire to move, rather than just the holiday. Agents continue to recognize the value in our unrivaled audience and products with a strong ARPA growth being driven by a combination of pricing changes and agents upgrading package to grow their business. Looking forward, we see that momentum continuing in the second half, and we are planning bring forwards in 2022 pricing changes into this year. This won't impact our revenue this year but will mean we hit the ground running in '22. Mindful of the lack of stock, I think it's prudent to assume we'll see a small growth in agency branch numbers over the rest of the year. And equally, I think the very good conditions will continue for New Homes developers. So for the rest of the year, I'd expect the number of developments on site to fall as they set out. And as you've seen, there's no slowdown in our ambition to make home moving easier, or indeed a long-term outlook.

Peter Brooks-Johnson: Good morning, everyone. I'm joined by Alison Dolan, our CFO; and Miles Shipside, our Housing Market Expert. Hopefully, you've had chance to see the presentation. I thought maybe to start us off by taking a couple of minutes to give you a quick summary of all those words and pictures. I guess one can't start without saying that COVID has, obviously, upended the lives of everybody. And it's had tragic impact on many people. And this time a year ago, we were in the middle of the unknown. And I know one of the key questions, a number of you have asked me over the subsequent months is, "How well right moves emerge from the lock downs?" So, rather than compare against June 2020, which I think might be a little flattering to us, I'll focus my comparisons against June 2019 to give you a better sense. And from a Rightmove's perspective, I think we've emerged stronger and as, hopefully, as some form of normality returns, you can see that the network effects at the heart of our business are stronger than ever, with record traffic and leads. Traffic over the last six months was over 60% higher than in 2019, which is partly influenced by the very strong heart - housing market. But if you need a demonstration of the instinctive place that home hunters turn to first, we recorded our new busiest day ever on the third of March. That was the day of the budget, with over 70 million minutes spent on the platform in a single day. And according to Comscore, our market share of time has nudged a little higher. Turning to the housing market; I'm pleased to say that the leading indicators that we show you every time had been proven right with HMRC recording more transactions in the last six months than a period before 2006. And there's been a lot of discussion about the impact of the taper of stamp duty holiday in the second half of this year. We're now over the first step of the taper at the end of June. And our indicators do give cause for - cautious optimism. Demand is still higher this July than it was in July 2019. And sales agreed are running at a rate similar to about to those two years ago. So there's no sign of a cliff edge at the moment. And actually, there's still a lack of new stock coming to market. It's a bit better than it was back in February when we spoke, but it's around 10% down on the usual run rate. And that's having a number of effects. Firstly, most, obviously, house prices are being driven up. Average asking price is now 7% higher than it was at the start of the year. Probably surprisingly, to many the markets becoming more efficient, typically, sort of long-term run rate. On average, only half of property's listed at sell. At the moment, that's risen to nearly 70%. And it's that increase, which is supporting the increased rate of sales being agreed. Couple of dollar impacts, new agent formation is, perhaps, slightly slower than one would expect at this point in the cycle. There's - of course, a new agent has to win all their stock before they can begin trading. And New Homes developers are also experiencing record demand, with 20% more developments selling out in the first half of this year compared to 2019. And you can see all of that rolling into our membership numbers. We've had record-high retention rates in the first half of the year. So despite the slower agent formation, we've still grown agency branches by 130 since December 20. And whilst we have the same number of developer customers, the strong New Homes market, alight to a slightly slower build rate means that development listing numbers are down 211. And looking forward, I think we should expect more of the same in second half, slightly up maybe broadly flat branch numbers and a slight fall in development numbers while that supply and demand imbalance continues. I think it's difficult to predict the impact of the final end of the stamp duty holiday at the end of September. And actually I think from my point of view, it would be good for the market and probably good for Rightmove if we saw a slight cooling. Moving on to ARPA, I think we've seen strong trends in both the agency and New Homes ARPA. I think they're easier to see an agency. Agency offers up £107 compared to June 2019; that's 11% up. And it's been driven by a combination of product adoption and price rises. We've had over 800 branches upgrade to our top package Optimiser 2020 in the first six months of the year. And for context, we upgraded just over 1,000 in the whole of 2020. And like-for-like, Optimiser 2020 has been selling faster than its predecessor Optimiser 2015. And we still got lots left to go at for Optimiser 2020. We've got around 1,700 branches on Optimiser 2015. And we've seen around 40% of our upgrades actually to Optimiser 2020 coming from our essential and enhanced customers. Before I summarize, I think we were rightly cautious at the start of the year. But as ever, we've responded to the environment. So we've doubled the number of customers we're going to talk to this year, with the majority of those conversations now complete. And looking at New Homes ARPA but the headline level is down £17 since 2019. But I think the really the really strong market is hiding the underlying picture. ARPA from our subscription listings and subscription products to New Homes developers is up £96, and that's been offset by a reduction in the one-off digital marketing, which developers often use to sell the last few plots on development. And at the moment with such a strong market, they simply don't have them. And in terms of innovation, we've delivered the three new products we talked about back in February. Online conditional auctions, a solution for the built to renters category went live a few days ago, and the new advanced listings for New Homes developments, which is selling well, with over 20% of new developments now buying that listing. And we've made goods - good progress on our strategic integrated innovations in the first part of the year. In the digital rental flow, Van Mildert has now been revamped - rebranded Rightmove to leverage the trust consumers nations have in our brand. And the team have just delivered one quick tenant referencing audient - ordering for agents in Rightmove Plus, which, obviously, will save them time and rekeying errors. We've also made references 20% quicker on average over the last year with no less of quality. So that's important for agents. We've also launched our new digital tenant insurance flow. We just launched that, and we've got much to learn in the flow. But it's certainly an exciting start. And we continue to make progress in our mortgage experiments. Over the course of the last year, we've doubled the number of leads we send to our lender partner. And we've increased the quality of those leads with three times as many now receiving a lender-backed decision in principle. We've got more innovation planned over the course of the second half too, but I guess I can talk about that when we meet, again, in February. So looking ahead to the second half; I think we'll see more of the same old membership numbers and further strong ARPA performance. Those of you who followed us for a while know that usually in the second half, ARPA increases around £10, £11. I think this year we'll do a little better than that. But I think actually looking at your numbers; most of you have worked that out. Where one of the joys of the subscription-based businesses is you can get a head start on subsequent periods, so we're also making sure we hit the ground running if in '22, by pulling forward some of - some of the pricing activity that we would traditionally have done in early '22. And, finally, we're continuing our progressive dividend policy by increasing the interim dividends to 3P. So that's it for me. Now over to questions, perhaps, Nadi [ph], just remind people how to ask a question.

Operator: Thank you. [Operator Instructions] The first question comes from line of Joe Barnet-Lamb from Credit Suisse. Please ask your question.

Joe Barnet-Lamb: Jerry?

Peter Brooks-Johnson: Hi.

Alison Dolan: Hi, Joe.

Joe Barnet-Lamb: Joe, by the way. So that's fine. That's no problem, whatsoever. Three questions.

Peter Brooks-Johnson: Have I stopped calling you, Jerry?

Joe Barnet-Lamb: That's fine. I'm down. I quite like Jerry. Anyway, three questions for me, if I may. So Alison, you gave some great color around the drivers of ARPA in your presentation. In new home, it was discussed about they put 83 pound drop from digital marketing, which is, typically, on top of subscription. What percentage drop in digital marketing was that, i.e., how much digital marketing revenue remained in new home in 1H '21? And could we see sort of a greater drop off from that perspective going through the year? Secondly, on Opportunity Manager, and thanks for the color on how the product works and the value it drives? I don't see you explaining much about monetization, except to say that it's only available to optimize the 2020 customers. Can you talk a little bit around the incremental charge associated with that, and how material and opportunity it could be? And then thirdly, you mentioned bring forward a number of the price increases, which was scheduled for later in the year. From your comments just a second ago, Peter, I'm assuming that, therefore, isn't actually implementing price rises sooner. It's just having the conversation to lay the groundwork for them sooner? Is that a right way to sort of think about it, you're not actually doing the price rise sooner, you're having the discussion to alert them of the future price rise earlier? Is that a correct way to interpret that? Thank you.

Peter Brooks-Johnson: Thanks, Joe. So we maybe we work backwards up those questions. So the first one, the price rise, we're actually going to implement them sooner. So it's not just conversations conversation and action but you won't - it'll it given the timing in the year, it won't impact revenue and after really in '21. It will be a '22 impact, hopefully, that's clear. But we will we will be doing, we bring forward some of those conversations and action. Opportunity Manager; Opportunity Manager is part of the Optimiser 2020 package. So it's one of the things that you get as part of the package is not an extra - there's no extra monetization. But it's, certainly, one of the reasons we see customers upgrade to an Optimiser 2020; too many O's. Optimiser 2020; it - it's a package, in many ways, like our others. The two - the two features are access to Opportunity Manager, as part of the package, but also exclusive access to our Sold By Me product, which is an additional charge. So it's those two things, which are driving the upgrades to Optimiser 2020. Does that make sense?

Joe Barnet-Lamb: Yes, absolutely. That makes perfect sense.

Peter Brooks-Johnson: Right, and then, perhaps, over to Alison to answer the -

Alison Dolan: Hi, Joe.

Peter Brooks-Johnson: Just the marketing question.

Alison Dolan: Yes, yes. Hi Joe, morning. So digital marketing would normally be about £4 million. And this drop represents about £1.5 million of that. It can be quite seasonal. We'd normally see a drop into the November-December time anyway. But just to put it in context, that - that's the sort of quantum that we're talking about.

Joe Barnet-Lamb: That's ideal. Thank you very much, team.

Alison Dolan: No problem.

Operator: Thank you. The next question comes from line of William Packer from Exane. Please ask a question.

William Packer: Hi, Pete and Alison. Thanks for taking my questions for me briefly. Firstly, the tone around costs on the webcast is quite cautious with phase - like phasing definitely highlighted. But I think the substance of the commentary is actually quite positive. And I wanted to check my understanding is correct. So you said OpEx would be 25% of sales for '20 - FY21 and FY22, which compares to consensus expectations of 27% for FY21 and '22, is that a fair summary? Secondly, you mentioned agent formation is a bit slower than you would have expected in this market context; do you think this is structural or cyclical? I.e. is it cyclical because of the lack of stock or is it the dynamics around online estate agency mean that you have a more consolidated market? And then, finally, monetizing vendor leads has been a critical growth driver of your business in recent years. Scout24 have made quite an interesting acquisition in the space in Mocap24 [ph], where they directly share in the agent commission. Are there similar businesses in the U.K. or is this the kind of models you might experiment within due course? Or is the U.K. market structure less integral [ph]? Thanks.

Alison Dolan: Thanks, Will. On the cost side, yes, absolutely, 25% of revenues is what we're expecting for the full year, which will be a little bit more than double what you've seen in the first half, for a couple of reasons. We will still be recruiting in the second half just recruitment, given the work issues in lockdown has been a little slower than we expected. We have also had the vendor provision relief, which is £2.4 million, which will be a reduction in costs in the first half that won't repeat in the second half. And then, finally, savings from people being at home have just gone on for longer than we thought that they would, and feel not all of that will be available in the second half as well. So think of full-year half as slightly more than H1. But, I guess, very much within the guidance that we've given you towards the low end of that guidance, in fact, and I think we've been quite specific on the percentage of revenue guidance.

Peter Brooks-Johnson: Okay, talk about the other ones. So as your information, cyclical or structural, I think it's cyclical, will - it's very hard to win stock. There's a couple of - couple of elements. New sort to market is, is obviously tight. And that means there's an existing agent, you sort of get your first go. So as a new agent to win a - to win a portfolio to sell, if you like, is really tough, so people are staying out. And actually existing agents aren't talking. Some of them are sort of poised to open new branches, but they want to see a bit more stock return to the market. And there's, perhaps, a slightly secondary effect. So - because now 70% of properties are selling, the markets much more efficient; there's many fewer properties that are moving from agent to agent. The agents call those secondary instructions. And that sort of second bite of the cherry is going away because things are selling so quickly. So, I think, that's what's going on definitely cyclical. Structurally; no one's asked me the question yet, but I thought I might as well give you the answer because someone will. The hybrid agents, market share the sort of whole segment market share pretty steady at around 8% of listings. So we haven't really seen any change there. I can talk about that more - if someone wants to talk about that. Vendor leads; yes, really interesting. Some of the stuff that Scouts are doing. There isn't really an equivalent to - in Macau, in the U.K. I mean some very, very small things, but nothing at that sort of scale. I think, partly, it's due to the structure of the agents in Germany versus the agents in U.K. The sort of sharing in commission, most big agents in the U.K. bit - and that's, typically, my understanding is that agents in Germany are a lot smaller. So there's - this - there is - one they have big networks, the individual realtor, if you like more like U.S. model is much smaller. So they're much more willing to share on commission than our agency businesses. I think sharing commission's an interesting topics that are very small, and certainly one of the things that sort of on my radar as maybe a way of helping people start up at some point in the future. Does that - does that answer it, Will?

William Packer: Yes, yes. Thanks for the color. That's helpful. Appreciate it.

Operator: Thank you. [Operator Instructions] The next question comes from the line of Andrew Ross from Barclays. Please ask your question.

Andrew Ross: Great, thank you, and morning, everyone. I've got three as well. First one, I've just pin you down on a bit more detail on the offer guidance for this year. You said it would - you'd expect it, clearly, to grow in the second half versus the first half more than normal, but I think the full-year results you talked about growth of 8% against the December 2020 base or around 1190. So is that still sensible for this year? That's the first question. And the second one is to come back on - before for the price increase for '22. Can you give us any kind of steer as to how many agents have been pulled forward? Or, I guess, a different way of asking that question is normally you think about also growing, let's say £80 year on year on absolute terms, how much better than that might you do for 2022, as you see here today? And then, the third question is on Van Mildert. Can you just remind us why you haven't made the earnouts? So kind of what the targets were or why was there any color on that would be helpful. Thank you.

Peter Brooks-Johnson: Thanks, Andrew. Yes, I think we saw of those. So up guidance, yes, I think - I think what we said back in Fed was expect slightly softer than 2019 in on the - on that December off a number. Yes, I think we'll do a bit better than that. So 7% to 8% probably would be - we'd get to sort of numbers. So I don't - I don't shy away from what you've said. In terms of portfolio price rise, it's about double the number that we were expecting, in total. It'll be about double the number this year that we were expecting. What does that mean, for our next year? That Plus, the product stuff we're doing, I think, we'll do better than the - than the sort of £80 to £90 next year, in absolute terms. And then, Van Mildert, the earnout. So Van Mildert, the earnout based effectively. It was a sales number over an average of two years. Probably really important to point out that most of the earn out went to the former owners, not the management team. So I'm really sort of - I'm really okay that the management team are still super motivated, because this doesn't really affect them at all. And, obviously, through the pandemic, last year, the sales rate was the referencing rate was significantly hit. The housing market, obviously, including rentals was shot for nearly three months, it then took a little while to get back going after that with all the obvious concerns about health and well-being. So that's really what's led to the Van Mildert earn-out. And as I say, I feel really comfortable that the management team are protected and motivated. So that's what's most important to me. Does that - does that answer your questions, Andrew?

Andrew Ross: Yes, that's great. And maybe just to follow up quickly on my second one there. You said you're increasing about double the number of agents you normally would. What kind of typical amount just to transfer the number [ph]?

Peter Brooks-Johnson: And so, it was - it was - it's double. It's - I'm not going to give you the number; but it's double the number we set off at this year. So it's - I think, probably at the start of the year, we were still pretty uncertain about what the market was going to hold. So we were a bit cautious. So we - we're getting closer to our usual run rate for a year, actually.

Andrew Ross: Okay, cool. Thank you.

Operator: Thank you. The next question comes from line of Miriam Adisa from Morgan Stanley. Please ask the question.

Miriam Adisa: Great. Morning, everyone. Two for me. Firstly, just on the additional opportunities that you mentioned that you can get from the referencing product that you spoke about an additional £4 of profit per product? Could you just give us a sense of where that may be able to get to over time? I know you mentioned we shouldn't get carried away, but it would be good to sort of see how you think about that evolving over the next couple of years. And then also, you did mention that the market at the moment is quite competitive. So how do you think you can create that differentiation? And how do you plan to increase the tax rate? And then secondly, just on the competitive environment. I think you said you saw boom and have sort of 1% share of time spent? Have they done anything that surprised you? Or - and what are you seeing in the broader competitive landscape? Thank you.

Peter Brooks-Johnson: Good morning, Miriam. So £4, where could it go? Certainly, north of £4. As I say, please don't get excited about it yet. We - we're still very much in the foothills of that - of that journey. I think we we're probably less than - less than half of what it could be, maybe a bit more than that, actually, in terms of - and that will be a combination of things. That will be about - increasing the attach rate. As I say, we've currently been doing most of that insurance work via telesales. And that's deliberate because it allows you to earn - to learn very quickly, in terms of what's important consumers because we've never sold anything to consumers before. So for us, this is a real journey into the unknown, makes it super exciting for me. So the launch of our new digital flow, which only launched a couple of weeks ago, that's one of the key ways that we will increase the attach rate. And that will be a process of learning. As I say, we've got to learn. We've got a really good team, who have experiencing this, but we've got to learn how it works in this context. So that's probably how we will - we will move the attach rate up. And then things like T.V. and broadband, we've only just started talking about T.V. and broadband. So there's lots - there's lots to learn there. One of the critical things that we - and one of the reasons why we use the telephone, first is what we've discovered is timing is critical, which is one of the exciting things about doing the reference because you know the moving date, and I think maybe some of you have heard me say this before. What we discovered is that you have to talk about broadband much earlier in comparison to the move date than you talk about insurance in terms of how a tenant mind works, which makes sense to me, go where we'll be if we didn't have broadband at the moment, whereas insurance feels like a bit of a sort of admin task. So, one of the learnings is actually what's the perfect timing for the conversation. And when I mentioned differentiation, that's really - that's less about those products is more about referencing itself; so actually winning the reference business from [indiscernible] agents. And the way - the way we're doing that - and it's the path we've started out, and you can see, I suppose it started all the way back with our first version of the passport - is integrating that referencing flow starting at much earlier and integrating it into the tenants property search. So that's sort of how that will flow out. And I think we've, clearly, got a big advantage over every other referencing company because we got, by far, the biggest number of tenants sending leads and, by far, the largest number of rental properties in the U.K. So you'll see us integrate more and more and more, hence the reason one click ordering - reference ordering in Rightmove Plus, so that's sort of how we get - how we're thinking about that. In terms of competition, I think booming has - I probably said this before, yes, I think I totally respect the Bruce brothers marketing ability. So they've come out strongly with some really good marketing messages. I think one of the lessons for us, which the team has taken on board is, I suppose the language I would use is sort of merchandising, how do you describe your features in an exciting way? And I think they've done a great job on some of that stuff. So I think there's some lessons for us to learn there. And I think - I think there's sort of some detailed things in there but, perhaps, not surprises, but some great - some really great learnings for us. As ever, competitive environment, the real challenge for any competitor is getting consumers to go to your website. And you can see from our numbers, yes, as I said, it's Rightmove they turn to first that's the result of 21 years now, believe it or not, 21 years of marketing, spend brand building, delivering on our promise to consumers to show them properties, and to be 100% available and fast and all those really boring stuff, things that sort of interest me, but then interest many other people, but - so it's all of that comes together. So that's the real challenge. Can you get a consumers turn up your website because of course we're free to consumers; and I think that's - that will continue to be the challenge for competition. Is that --does that answer it, Miriam?

Miriam Adisa: Yes, it does [ph]. Thank you.

Peter Brooks-Johnson: You're welcome.

Operator: Thank you. [Operator Instructions] The next question comes from line of Adam Berlin from UBS. Please ask your question.

Adam Berlin: Yes, hi, everyone. Three questions for me. First one is a numbers question. Can you break down the £1,103 offer to December '20 in the New Homes and agency offer just to see how the - each of those have progressed since December? Second question is, can you comment on the new strike kind of model, which is allowing consumers, essentially, to list their properties on Rightmove for free? Is that a path towards a kind of more digital model for selling houses online that doesn't require kind of a normal agency? How are normal agents responding to that? Is that having the impact on the market? Yes, a second question. And then, the third question is, how do you think the market changes from its current dynamic when you've got such high demand and not enough new listings and the number of listings keeps falling, what's going to - what do you think could change that dynamics that we get back to a more normalized number of listings in the market? And how long do you think that might take? I know the crystal balling but just interested in your opinions on that question.

Alison Dolan: Thanks, Adam.

Peter Brooks-Johnson: Thanks, Adam.

Alison Dolan: Yes, I'll take the output for inflation [ph]. So Adam, I would discourage you from breaking out the £1,103 in that way because the £1,103 is a one-month ARPA number. And so, it didn't really make them to compare it to the six-month trailing ARPA just because there's too much in the mix. And there's also a unique dynamic to the 2020 number, which was the extent of ala carte product purchase because those numbers were still discounters, as well as the unwind of the discount, which was going on over October, November, December of 2020. So, I think the more clarifying comparison actually is if what has happened between June '19, December '19, and June '21. And so, maybe just to give you a little more color on that. So the agency ARPA increased by £107 between June '19 and June '21, £12 of that £107 came between June '19 and December '19. And then the remaining 95, as the market recovered from December '19 to June '21. And I think really 2020 was so disrupted that trying to compare any particular monthly specific points to the six-month trailing '21 number doesn't make sense and actually won't give you a proper picture of what has happened in the insurance.

Peter Brooks-Johnson: Okay. I'll - why don't I answer the - your question about Strike and then maybe you can bring miles into to uncover his crystal ball about the marketplace. So, in terms of - in terms of Strike, I think the way to think about Strike is Strike's an estate agent. They have - they're very similar in set up to the other hybrid agents, so Purplebricks [ph]; they're exploring a different charting model. So whilst it's free to list with Strike, you do have to commit to using their own services. And I think this is - this is a story that we're seeing a lot of and probably predicted for a while. We - the hybrid agents are all experimenting with different charging models. It's not a - it's still quite nascent. So you probably have seen Purplebricks now have a - have also have an option, where - whilst you committed upfront, they also have a refund option if you don't sell, so this - they're experimenting. And actually what we're seeing interestingly is traditional, if I can call them that traditional estate agents, experimenting the other way. So talking some agents, there are some now who have three packages. They have the sort of well known, no-sale no-fee model, they have sort of small upfront fee, followed by a smaller on-sale fee model or they have a wholly upfront fee. So the - there's sort of a - there's a whole bunch of experimentation going on. But I think the fundamental story is Striker hybrid agent, just trying to compete with the other hybrid agents. I think they will have some success. It is fascinating, as I - as I said at the top that the hybrid agent as a - as a total marketplace is about - still about 8% as it has been for quite a long time now. So I think they - they're winning within that context rather than versus hybrid agents. Maybe I'll hand over to Miles to give you his prognostications on the marketplace.

Miles Shipside: What changes the dynamics on listing numbers? So predictions are always difficult, but, particularly, over the last 18 months have proved difficult with the many commentators predicting doom and lower house sales and low-house prices rather than what we've seen. So looking forward, patiently, demand is exceeding supply. So we need a change rather than that to actually increase listing numbers and increase the supply. So usual dynamics is still happening, new households are being formed, first-time buyers buying, partnerships split up. You've got the 5% deposit coming in on first-time buyer mortgages, so increases demand as well. And obviously, the ongoing effect of people looking to have new housing needs and the house provided for things from the lockdown still on ongoing. Supply, more new build is, obviously, one route forward. But as Peter said, a lot of that's forward sold. So we need more sellers coming to market, who aren't buying to increase supply. But I actually think these most stock levels kind of continue because we're seeing a buoyant market. What's happening though? As Peter said, more efficient market properties selling quicker. So agents have gotten used to operating on lower stock numbers and, essentially, operating on a - on a faster, faster turnover. So plays into Rightmove's hands because obviously if they want new listings, advertising online movement, using Rightmove's products, very important for attracting new listing so a bit of a woefully answer but, I think, more of the same for a period to come. Obviously, a slump in the market means that you get more properties and sold and that increase the stock numbers but we can't see the signs of that at the moment.

Peter Brooks-Johnson: Thanks, Miles. Anything else, Adam?

Adam Berlin: Just for Alison. She said, of the £107 of agency ARPA since June 2019, £12 came between June '19 and December '20. Is that - is that correct?

Alison Dolan: No, no. June '19 and December '19.

Adam Berlin: All right.

Alison Dolan: And then, the move from December '19 to June '21 with £95.

Adam Berlin: Okay.

Alison Dolan: Thanks, Adam.

Operator: [Operator Instructions] The next question comes from line of William Packer from Exane. Please ask your question.

William Packer: Hi, Peter. Just a quick clarification. It should, hopefully, be quick. Just understand your precise wording around ARPU - ARPA for FY21 and FY22. I think you said there should be better than 7% to 8% growth versus FY19. FY21. And I think you said it'd be better than £80 to £90 of absolute growth for FY22; is that correct?

Peter Brooks-Johnson: 50%. Thank you for the clarification. So, what I think I said was better than 7% for '21.

William Packer: Okay.

Peter Brooks-Johnson: A little bit better so…

William Packer: And as verse - and as versus FY19.

Peter Brooks-Johnson: And as so is '19. Sorry, no, versus the - versus the number that we gave you at the start of the year.

Alison Dolan: The £1,103.

Peter Brooks-Johnson: The £1,103.

William Packer: The growth - it's versus - the 7% growth rate is versus FY19?

Peter Brooks-Johnson: No, it's versus the £1,103.

William Packer: Okay, thanks.

Peter Brooks-Johnson: And then in '22, yes, better than £80 to £90.

William Packer: Thanks for color.

Peter Brooks-Johnson: Thanks for checking.

Operator: Thank you. [Operator Instructions]

Peter Brooks-Johnson: I think - I think we're done. And I think everyone's asked all the questions.

Operator: There are no further questions. I would like to hand conference over to Mr. Peter Brooks-Johnson for closing remarks.

Peter Brooks-Johnson: Thank you, everyone. Thank you for coming along. And as ever, if you've got follow-up questions or detailed questions, don't hesitate to get in touch with Alison or myself; we'd love to talk to you. Have a lovely day.