Good morning, welcome to MONY Group's 2026 interim results call. We are joined today by Peter Duffy, Chief Executive Officer, Niall McBride, Chief Financial Officer, and Jennifer Cooke, Head of Investor Relations. If you would like to ask a question during today's call, please signal by pressing star one on your telephone keypad. You can also submit written questions via the webcast. I would now like to hand the call over to Peter. Please go ahead. Thank you. Good morning, everybody, thank you for making the time to join us this morning. Hopefully, you've all had the chance to watch the results video that we released at 7:00 A.M. Perhaps before we open up for questions, you'll let me just go through a quick recap. Our business only succeeds when we save customers money. In the first half, we were absolutely thrilled to have helped U.K. households save an estimated GBP 1.5 billion. It's through doing this that we delivered a strong start to the year. Record revenue was up 6% on a like-for-like basis. That was to GBP 227 million. Adjusted EBITDA was up 3% to GBP 75 million. That's our fifth consecutive interim period of growth. I think that really speaks to the strength and our breadth. We've expanded basic and adjusted EPS by 4% and 5%, respectively. We've grown our engaged member base. We've added an additional one million members to the SuperSaveClub over the last year. We now have 2.5 million members. Now went through some metrics which I think really spoke for themselves. MoneySavingExpert now has over 3.5 million app downloads and nine million people signed up for the weekly MSE newsletter. Quidco continues to offer more ways for customers to engage and save with us. I think it's also been the strategic progress as well as the strong financial delivery that's characterized the half. We are leveraging AI for growth to sharpen the customer experience, to bring new propositions to the market faster, and to run the business more efficiently. We've transformed the MoneySuperMarket app into a genuine everyday financial companion, one trusted place to compare, to switch, to save, to invest, to earn cash back, and it's going to give more reasons for households to come to us directly more often. This innovation is moving at pace. We've launched investments by MoneySuperMarket, now live. We've launched initially with a fund supermarket, enabling customers to invest from just GBP 1. We're launching our own digital broker, starting with motor insurance. That's SuperSaveClub Insurance. That's coming in the coming days, and that means customers can compare, can buy, can manage, can renew entirely within our own app. We also announced this morning that we've opened the wait list for our business banking proposition, our first dedicated SME proposition that will be launching in August. I'm sure you agree that we have been busy. For shareholders, we're planning to return over GBP 90 million this year. That's through our ongoing GBP 25 million buyback, but also our progressive dividend. Our brands, our established member base, our capacity to leverage AI, thanks to our leading platform, it's all putting us into a strong position. It's a powerful combination. We are uniquely positioned, and we're confident in the opportunities ahead, and we've got plenty more to come. With that, let's open up to Q&A. Thank you. As a reminder, to ask a question over the phone, please signal by pressing star one on your telephone keypad. You can also submit written questions via the webcast. Our first question is from Andrew Ross from Barclays. Please go ahead. Great morning, all. I've got two questions, please. First one is about double-clicking into your PPC costs and the dynamics in terms of gross margin. The online spend was up 34% year-over-year on a like-for-like basis, excluding travel and H1. Hearing you that there's been a lot of inflation through the PPC costs. When I listen to your remarks, it kind of sounds like you're starting to see now some stabilization in that market, and was hoping you can give us a bit more color in terms of what gives you confidence there. I guess particularly in the context of Google now looking to implement AI Mode into the search, why would we be so confident there isn't going to be a service squeeze on acquisition through a Google funnel? As part of that, if you can just give us some expectations about how you're thinking about your gross margin into H2 and into 2027, that would be very helpful. Then second question is on the dynamics in the energy market, and any color you can give us in terms of what you're seeing in Q3. Obviously, the price cap is high. There seems to be some pretty good savings out there on the Martin Lewis emails, but then there's also been a lot of volatility in terms of geopolitics and wholesale costs. What are the kind of big dynamics we should be thinking about when thinking about energy into H2? Thanks. Great. Thanks, Andrew. I'll do energy. Perhaps, Niall, do you want to kick off with PPC costs? I think you're right, Andrew. There's a lot of change in dynamics over the course of the last year. I think we were talking last year about uplifts Sorry, inflation and PPC costs of sort of 20%+. The reason that we were seeing that level of inflation coming through really was the introduction of AI Overviews, and you're calling out the extension of that now into AI Mode as well, into the search results. More than a year ago now, many, many results didn't have that there at all. Now, actually, Google has extended that sort of far and wide throughout its estate. What that also drove, though, was SEO was obviously pushed further down the page, and what we talked about was that we saw a lot of volatility in the SEO algorithms. You could be position one in the morning, position 10 in the afternoon, and sort of move up and down between that. We saw a different layout of PPC. PPC now looks very different to how it used to look. The classic box at the top of the page has kind of been shunted down, and many of those results now just look like any other results. That's why we saw that level of uplift coming through. A lot of that change has now happened, that sort of rollout across the estate is there. What we've seen in the first half is a moderating of that level. Obviously, we are comping it into these numbers. Part of the sort of uplift that you're seeing there is the comping from this year. If you look at H2 last year to H1 this year, we're in sort of broadly similar places, we are taking that inflation through. At the exiting the period, we're running at about 8% at the unit cost level for Google PPC. That's obviously well down on the 20%, but also I think a reflection of that settling for the moment of where Google is at. I'll hand over for energy there. Great. Thanks, Niall. Just to recap, home services was up 30% in the half. Look, our power in energy comes from three things, really. Editorial reach, our deep provider relationships, and exclusive deals. I think what we've been able to show in the first half of the year is that when we get those deals coming in, consumers are very interested in getting value on their energy. They really are up for switching, essentially. I think there have been questions in previous sessions about, is the market as competitive as it has been, say, five years ago? I think what we've slightly proven out in the last six months is that it can be. Your question, Andrew, was about the dynamics, and I think the dynamics really are about deal flow, and it's the relationship between the wholesale markets and the price cap, essentially. The price cap obviously is going up, and that sets the maximum amount that the customer will pay. We need wholesale markets to be at a point where providers are then able to offer deals that provide value for consumers. I think last Friday, wholesale markets weren't far off where they were at the start of the war. Let's not lose hope on that because it's been extremely volatile, and these things are very dynamic. I think that is the slight unknown that we face going into H2. We're very confident about our ability to reach customers. We're very confident in our relationships. We're very confident that this competitive market is attractive to consumers. It's just about the rate of deal flow. Got it. That's helpful. If I could just follow up on the first question, extend it into your thinking then on gross margins into H2, kind of how much incremental compression is baked into your guidance, that would be helpful. Thank you. I think to take a step back on gross margin, Andrew, it's always a function of mix. We've launched Which? and Blue Light Card as B2B in this half. Those will hopefully scale up as we go through the second half. We've talked about in the past, we will absolutely do those B2B deals. They add pounds, but they will affect the margin. Some of the effect that we've seen in H1 is a mix effect as well. If you think in the sort of two thirds PPC, one third business mix, that comes from a number of different areas. I think, as I say, PPC feels more stable. The business mix then will evolve as it always does. Thank you. Thank you. We'll now move to our next question from Joe Barton-Platt from UBS. Please go ahead. Excellent. Thank you very much for taking my questions. Three from me. Firstly, SuperSaveClub is obviously doing fantastically, but we've now seen the launch of the club in your largest vertical, insurance, and with a differentiated AI-powered experience. You say initially the financial contribution will be modest. Is that sort of a 2026 comment? Should we see it sort of more materially aiding the group through 2027? A little bit more color on your thoughts around that. Secondly, we've clearly seen an accelerated cadence of product innovation, I assume aided by the foundational work you've done on tech and data, but also aided by AI efficiency savings. Clearly, we'll see the benefits of that new product in due course, but do you expect to leave that cost in to keep that faster product cadence going forward? Over time, could we see some selective cost savings from here? Finally, you launched on ChatGPT earlier this year. Just if you give us a bit more color on what you're seeing around that post-launch, sort of how much traffic is coming from it, just a bit of an update there would be helpful. Thanks, guys. Thanks, Joe. Look, maybe I'll try and sort of answer your question strategically and then ask Niall to pick up the sort of numeric dimensions to it. Yeah, on SuperSaveClub, it's maybe worth just taking a step back, Joe, if you'll let me, and talk about the context of what's happened here. If you go back to what we did with MoneySuperMarket within the group, all the technology was fragmented. We brought that together so customers could essentially buy more things more easily, and that was the first base that we hit. The second base was to incentivize them to do that with the SuperSaveClub. The third base is to bring that all together in the app, the fourth base is to begin to infill some of the gaps in the proposition, savings, investments, and, in this case, brokered insurance, which will just provide a more complete offer for the customer. As a result of having what is now becoming a sort of closed user group of customers who are putting their hands up to say they're interested in saving money, we can think about how we can begin to offer our consumers more value there. We can think about how we can offer providers more value there. Brokered insurance is one way that we can begin to do that. You'll know that our model historically has worked on a CPA basis. Brokered insurance takes us into a place where we can actually begin to renew as well as just get that one-off acquisition benefit. The customer can take advantage of providers who are absolutely bidding for their business. We think it's an important strategic move. We're starting off with motor. You can probably guess where that will go to. We're going to take it step by step, which is really what the guidance is about in terms of limited 2026 impact because we have to get our pricing right. We have to roll that out. We have to understand how customers react to it. I'll get Niall to pick up on that in a little bit more. Accelerated cadence of MPD. Yeah, new product development has really kicked off, as you can see. That has, to a large degree, been facilitated by the tech and data work that we've done over the last few years. Obviously, because that is well structured, it puts us in a really strong position to take advantage of everything that is AI. I point to the fact that our CapEx of GBP 5 million is the lowest in 10 years. Equally, it's been a year of, or half of phenomenal delivery in terms of our new products there. Will that cost stay? I think that's a very low level of CapEx anyway, I'm not suggesting that that is going to wither away to nothing. If you want to have a tech stack that is healthy and is current, you're going to have to continue investing in it. Please don't hear that there's going to be some massive step up in it. It's particularly low at the moment. We will just keep that at very sensible levels. In terms of ChatGPT, you'll remember that we did two major things, I think, or we announced two major ChatGPT initiatives at the full year. The first was to put Price Optimiser onto our motor insurance journey. If you remember, that's the thing that says to customers, the answers you're giving may be driving unnecessarily high quotes. Here's an example. You're saying that you drive 8,000 miles a year, but when we check your MOT, we see that you only drive 6,000 mi a year, or you're suggesting a certain level of excess, not understanding what that is doing to the overall premium. We make suggestions to customers in that application process, and they correct their answers if that is something that makes sense to them. We've seen about 200,000 people save, on average, GBP 25. Really, really significant. At the same time, we launched initially with four products, but we've now rolled that out to all the products that are on SuperSaveClub. Travel Insurance was the last to go live just a few weeks ago now. That's for customers who want to essentially operate entirely within the ChatGPT environment. We're not releasing numbers on that, Joe, but what I can say is it's nowhere near the scale of using AI on existing journeys. We are an early adopter of ChatGPT advertising. They've got a number of companies in the U.K. who are trialing advertising with them at the moment. We're just going through that. Let's see how that begins to change things. As we sit here today, we're still getting very limited traffic coming through LLMs. Essentially, whilst you're seeing one in five searches in the U.K. coming through LLMs, really for us, we haven't got significant volume coming through yet. Niall, do you want to pick up on the- I want to, yeah. The when question that Joe's asked? I think I'm probably going to give you an unsatisfactory answer, Joe, in that I'm going to say we are doing exactly what Peter says, which is we're testing and we're learning. We've just launched it, and it's going to take us a little bit of time to kind of get it up to speed. As you can imagine, we're going to roll that out slowly and sort of ramp it up to the amount of traffic that sees it. This is for SuperSaveClub to start with. Basically, I'd say remember that comparison will still be a very big part of it. Compare is the first thing you do in broker comparison. The core product will still be there. The difference here is we're expanding it so that you can compare by manage and renew. I think we will see it scale up. The first KPI that we'll be measuring, if you like, is really around engagement. This is a connected ecosystem of things where we're bringing together for people being able to manage their everyday bills, their savings, their investments, and what we want to see is people engage in this product as well. That's what we'll be looking at, as well as just seeing getting the product right over the next number of months. A useful cover there. Thank you, guys. I guess maybe just one very brief follow-up, is it fair to sort of propose that the benefits of the SuperSaveClub are going to be seen more materially over time as you fold in your largest product area in insurance? I mean, obviously not having had insurance in there means that the benefits of the SuperSaveClub have been materially more limited. As you roll that out and you get that product market fit right, you should see the benefits of compound through to the group fairly materially. I think what I'd say to that, Joe, is to some degree, SuperSaveClub is changing. What started as an incentive program for certain products is now becoming this closed user group where we're looking at it in a way that says, how can we help customers save more across a broader range of activities? We've launched investments, an example. If you want to buy a Vanguard fund, it's cheaper to buy it with us than it is with Hargreaves. We've launched savings, for example. If you want to get a top paying instant access account, I think Revolut offer 3.5%, 3.75% at the moment, we're at 4.1%. We're on better deals. If you want to get your car, your home insurance through us, we'll guarantee that we give you the best price. The point is that this is all in one place, and we have simplified it, and we have made it really easy for customers to do it. I think the dimension that we're looking to unpack here is how do we get our existing customers to do more with us? Then, of course, there are some changes in the offer as we begin to put our own product through that, whether that's our own instant access savings account or our own motor insurance policy. The consumer has real advantage in doing that. Just one thing on the motor insurance policy. We don't charge you any extra for paying monthly. That is really rare in the market, and many customers are going to find that beneficial. 40% of customers pay for their car insurance monthly. A lot of the 60% only pay it annually because they have to pay a fee to pay it monthly. I think that's going to be a very big benefit for many people. Thanks, Peter. Thank you. We'll now move to our next question from Luke Holbrook from Morgan Stanley. Please go ahead. Thank you for taking my questions. My first is just on the motor insurance premiums, which look like they're beginning to rise if I look at external data in the past quarter for the first time in two and a half years. I'm just wondering what you're experiencing through your platform and the possibility of premiums rising and volumes improving in car insurance through the second half of this year. The second question that I have is just on the consolidation in the industry, given where the backdrop of AI valuations and where they're sitting, what's your view on the potential for more consolidation and also on M&A? It'll be useful to hear your thoughts there. Thank you. Niall, do you want to pick up motor insurance premiums and on the consolidation? Yeah. I think what we've seen in the half, obviously, is that the sort of fall in premiums has been narrowing. By definition, there will be some stuff in there that is rising. When we look at it, -9% was what we saw in terms of deflation in the second half of last year, -5%. It's definitely narrowing up. Clearly, what we saw a number of years ago, rising prices does cause people to come. I will put this in the context, though. We have a very robust switching market this year. This is a large market. We're not sat there saying that this is a sort of a very down year. No, I think that's absolutely right. We said at the full year back in February that we felt we were sailing into calmer waters and a kind of war on the other side of a world aside. That has broadly been the case, I think, in the first six months of the year. Luke, your second point was on market consolidation. Look, the balance sheet is clearly an advantage to us. I think also what you're seeing in our announcement today is a very thoughtful approach to when we would do M&A versus when we would actually build out organically. Many of the initiatives that we put in place, savings, investments, Business Banking, even the brokerage side, you could have actually got there through M&A, but we've made a decision that says with everything we've done on the platform, with everything we can do with AI, it makes more sense for us to build that out ourselves in a way that absolutely ticks the boxes that our customers require. That means the balance sheet is there when there is an opportunity which is attractive for us, which is attractive for our shareholders, which is attractive for our customers. As you would expect, we scan the market. We look at a lot, we consider a few, we do very little, which I think is just how it should be, really. Just because there hasn't been any meaningful M&A for quite a while, don't think that it is because we are not looking. We absolutely do look, we just have a very high hurdle that needs to be jumped, particularly when the share price has been at the levels that it's been over the last few months. Understood. Thank you very much. Thank you. Our next question is from Tim Ramskill from Bank of America. Please go ahead. Morning, gents. Three questions for me as well, please. Firstly, with regards to the insurance vertical, I think I'm right in saying revenue's up 4%, but you've lost nine percentage points of EBITDA margin, so cost's up 25% there. Maybe some explanation, which might link to the second question, you've already been asked about this earlier on the call. But maybe now just to clarify, you've obviously framed the inflationary dynamics around PPC now versus last year. Last year, you increased your spend online by 8% for the whole year. Then in the first half of this year, that spend is up 34%. Again, that doesn't quite chime with the observations around the inflationary dynamics. It'd just be helpful to understand what's happening behind that. Then I guess, again, Joe sort of asked you about cost-saving potential. Clearly, the improved technology stack over time has enabled you to be a lot more efficient. You can reference headcount down, et cetera. Do you see more levers to pull on the OpEx side going forward? Because I know, Peter, you referenced CapEx thoughts. Just in terms of future OpEx, where you see things trending from here, please. Well, I'll pick up the third one. Niall, do you want to pick up on one and two to start for Tim? In terms of the insurance, in terms of where we've gotten to in growth, look, we have been spending more on PPC in insurance because the market has been there to go and get. That's very much a function of where we are. Sorry, what I should say about that is, we always make those types of decisions. There is a trading decision that is going on every day about where does it make sense for us to invest to make a return that is acceptable to us. You'll have seen through the course of last year, at certain points, we decided that it wasn't acceptable to do that, at certain points we'll have trimmed spending and at other points we accelerate it. Clearly, in this half, we were happy to accelerate. We're happy with the level of return that we were making. Last year, when we were seeing those types of rises, we definitely took decisions, you'll recall whilst the car market last year was going backwards, we were growing in insurance because we leaned into other areas, such as life insurance and others where they weren't necessarily as big a PPC market. We went at those markets in different ways, including CRM. I think what you're just seeing between questions one and two, Tim, is a sort of reflection of a combination of how we generate the return that we want to generate versus where the market is at a given point in time. Should I pick up on the cost-saving point? I think the way to think about this, Tim, is AI is this is going to sound so glib, but literally changing everything. If I look at the Codex rollout in our engineering community, I look at the efficiency that we're getting, it is fundamental. If I look at our finance teams and how we're changing processes, if I look at our customer ops teams and how customers are adopting to AI-based solutions, it is literally transformational across the organization. That has, with everything else we've been doing, enabled us to realize some OpEx savings up to now. Equally, you can see us launching a series of new propositions that will all need managing. They'll all need staffing in different kind of ways. I think the efficiency on the existing business will continue to be found. How we realize that may differ. It may not always be in OpEx savings. It may well be that we put that resource then into new initiatives, which are driving revenues in different sorts of ways. The AI adoption, I think is really transformational, we are properly at the leading edge of making sure we realize that opportunity. Okay, great. Look, well done in terms of a lot of the new product launches. Sorry, just to follow up there, Niall, just coming back to your point around what you spend, and I completely get the dynamics around day-to-day decision making around which business to go after. Just to be super clear, in the half, your insurance business is up 4% from a revenue perspective. It's down 13% from a profit perspective. Does that mean if you're winning business that uses, you described acceptable levels of return, that that is acceptable to be down more than 10 percentage points in terms of profitability, and that's a new base? Because it's a massive margin delta, H1 on H1. I'm just trying to understand where we think we settle off. Again, there is always a mixing going on, Tim. Each individual decision about what is acceptable in any given time is down to that. I'm talking about, we go bidding campaign by campaign into car insurance, home insurance, travel insurance, all of that type of thing. There are a number. Last year, as I said, those markets, those big markets weren't there as strong as they were. We were very strong in some of those other segments, like life insurance, like travel, like those other areas. This year, we've gone harder in some of the more core areas. It is a mix. We will see mix effects like that. That's true every year. We mix between where we think we can get the revenue and the profit that we want to get. Just my only build to that, Tim, is insurance is a gateway product to the group of a SuperSaveClub. Once customers join the club, they hopefully go on and buy other things from us. That is the data that Niall unpacked in the presentation. We're always going to report this in a stovepiped way. There is a bigger benefit, which comes as a result of this activity. The retention dynamics change. That's totally fair. Cool. All right. Thanks, guys. Appreciate it. Thank you. Our next question is from Weng Lum Khoo from Jefferies. Please go ahead. Morning. Thanks for allowing the opportunity to ask questions. On the first one, I have two, please. The first one, SuperSaveClub Insurance. How many insurers are on that product relative to your traditional comparison product, just in case if I missed it? If it's not all providers, may I then follow up with what's the selection criteria, and do you plan to have all providers on it? The second question is on the financial companion strategy. We've seen the launch of savings investments. Could you give us an outline of what is next to come on the product roadmap? I appreciate you may not want to disclose the exact products, but perhaps what's your grander vision for this strategy would be helpful. A quick follow-up would be how much education would be required to let existing and prospective money users know that you can do much more than comparing products on MoneySuperMarket now. Thank you. Great. Thanks, Weng. On SuperSaveClub Insurance, we're starting with three brokers. We've got some more lined up. That compares with 120 providers on the main platform. This is a very different proposition in terms of brokerage versus comparison, and it will all come down to the pricing that those brokers deliver. It's of great interest because it's a closed user group. We know who those customers are. As I say, there are more providers in the wings to begin to add to that. Isn't going to get to a number like 120, but it will probably get to a bigger number than three. The financial companion strategy. Yeah, thank you for asking that because I think this is really important, and I sort of touched on some of this in my answer to Joe. We are the MoneySuperMarket here. What we want to do is to begin to fill the gaps in the proposition for the customer and find more ways for them to begin to save. Savings was a logical first step. Investments is the next step. Insurance brokerage for customers who we know and we can potentially get even better prices for is a step again, in terms of how that works. Important in terms of starting to launch own brand products. I don't think there's going to be any more big categories, but I think what you'll see is a filling out of those propositions in terms of practically how that works. The big thing that we're going to be delivering here over the next 6-12 months is the financial intelligence, the AI intelligence that sits on top and begins to prompt customers how best to manage their money, and they can solve many of those problems within this environment. The big thing is how simple it is, and it's all in one place. I think that's what we're going to keep drawing people's attention to. There is nothing quite like this in the market where you can save money on insurance, and you can put that saving essentially into a savings account, into a tracking ISA, if that's what you wanted to do. I think you're right. This is a big change for consumers. Many customers find that when they go and they use the product, and it's now clearly laid out. I hope everyone's had an opportunity to see how it works on the video this morning. Of course, it will require promotion. I'm not sure that is additional promotional spend. I think we can do that within our existing envelope. We've got some very clever people thinking about how to bring that front of mind with consumers. That's a big focus of ours on the inside wing. Thank you very much. Just to follow up on that, on the SuperSaveClub Insurance, what's the selection criteria for the three brokers and how would you move forward with that? We're not going to share that. We obviously talked to some people who were interested in doing something innovative, getting involved with us from a very early-stage concept and bringing that through, we'll expand that. It's not that people have been excluded from the process. It's more about partners who are excited in joining something at a very early stage. Thank you very much. Thank you. As a reminder to ask a question, please signal by pressing star one. Our next question is from William Larwood from Berenberg. Please go ahead. Morning. Thanks very much for the presentation. I've just got one left. It's really just in terms of the rollout of business banking and investments, just if you could give us a flavor of the pace of when you start to expect those to contribute. I know you've already touched on SuperSaveClub insurance brokerage. Just tangential to that in terms of if you could give an update on savings and how that's progressed in terms of the growth in the AUM on that. Just give us a flavor of the update. That'd be helpful. Okay. Yeah. Thanks, William. I think what we're trying to do is position these things very much as trials. If you take SuperSaveClub Insurance, if you take investments, if you take Business Banking, they're all really new. I mean, kind of days. The oldest one is days old, and others are just kind of rolling out or about to roll out. What's interesting about Business Banking is we already provide energy and insurance services into that segment. Making Tax Digital is a new requirement that the small businesses of the U.K. have to adhere to, and we think we can begin to help with this product, and we can begin to bring that package of offering together in a way that's kind of really meaningful. Investments, as I've sort of described, is a logical next step from savings. It's an infill to the MoneySuperMarket. We're starting off with a fund supermarket. You can open either an investment account or an ISA to begin with, and then we'll begin to build that out in sophistication as well. I think really savings was a toe in this water for us. It's really helped us to begin to understand how customers understand what is now available, how we communicate with them, how these things work in an integrated way. Rather like when we launched the SuperSaveClub, we're not going to be at a point for some time to say how significant or meaningful this is because we're going to have to test these waters and actually size that before we share information essentially about what that opportunity looks like. I hope what you can read from that is a real approach to innovation, a real customer centricity, a real thinking through how we can actually build our wallet share with consumers. As that understanding develops, then we'll of course, be sharing that as soon as we can. Thank you. It appears there are currently no further questions over the phone. At this stage, I'd like to hand the call back over to Jennifer for any webcast questions. Yeah, we have a question from Alastair Reid at Investec who has said, "You've touched on your established provider relationships as part of your moat. How difficult are these to implement and replicate? Do you think that providers would even want to risk integrating their systems directly with LLMs? Thanks, Alastair. I think they are difficult to replicate, I think what we deliver to providers are secure links, essentially. We capture, in the case of car insurance, a minimum of 41 different pieces of customer information, over 60 for home insurance customers. If it gets complicated, by which I mean if you have an additional driver, if you have a conviction, you go well over 100 pieces of very personal information. I think that number one is, the LLM, you wouldn't know what happens to that data if you put it into an LLM. The LLM wouldn't know what happens to that data if you put it into them. I think number two is you have to be regulated in the U.K., as we've said before, to begin to sell financial services. I'm not sure that's sort of possible at the moment in terms of how that would work from a regulatory point of view. Number three is this is a solved problem, you can see that everything we're doing in the app makes it very easy for the customer to come to us and use AI-informed services to begin to make that work. Number four, this is where the trusted brand kicks in, where customers know that we'll do everything we can to handle their data responsibly, and we'll manage things in the right way for them. I just think it's a very deep moat that we operate, and I think that's why we can then be so front-footed with AI because we're comfortable that the model is secure. Fabulous. We don't have any further questions on the webcast, I'll hand back to you, Peter. Okay. Well, look, thank you, everybody. Thank you for your time. We're going to be meeting a number of you over the coming days. Of course, we're here to catch up if you aren't on the immediate roadshow. In saying that, thank you very much for making the time for us this morning, and we look forward to speaking with you all soon. Cheers. Have a good day