Rentokil Initial plc (RTO) Q3 2024 Earnings Report, Transcript and Summary

Good morning, ladies and gentlemen. Welcome to our conference call for Rentokil Initial's Trading Update for the 3 months to the 30th of September. This call is scheduled for 45 minutes. And now let me hand it over to CEO, Andy Ransom. Please go ahead, Andy.

Thank you, and good morning, everyone. I'm here with Stuart, and in a few minutes, we'll be pleased to take any questions. But first, let me just say a few words covering the third quarter and importantly, the actions that we've been taking following the September trading update. In the third quarter, the group delivered total revenue growth of 3.6%, of which organic revenue growth was 2.6%. There was good momentum sustained in the group's international regions, which in aggregate, delivered organic growth of 4.4% in the third quarter with Europe delivering organic growth of 4.7% and the Asia and MENA region delivering organic growth of 6.5%. Year-to-date, our international businesses outside of North America, have delivered a combined organic growth rate of 5%. Turning now to North America, where in the third quarter, overall, we delivered organic revenue growth of 1.4%, with North America Pest Control also up by 1.4%. Clearly, we were very disappointed to announce in September that the business was underperforming our expectations. And since then, we've put in place a focused action plan on a number of fronts. Firstly, in our operations, we've taken decisive action to mitigate cost overruns. Given our elevated workforce costs this year, we've reduced our sales and service headcount in addition to the normal ongoing off-season headcount reductions. Over the last few weeks, we've removed around 250 roles from service sales and back office functions with an annualized cost saving of around $22 million. We've been tightly managing overtime and labor as we've entered the off season. And as we said in September, material and consumable costs in the North American business have been higher than expected, partly due to inflation. There was also an impact from a new ordering process for Terminix branches and a weaker termite season that resulted in elevated inventory. To help mitigate some of these effects, we've now implemented strict ordering controls at branch manager and regional director levels. Some additional costs due to inflation is expected to persist. It's around $7 million on an annualized basis with around $10 million of other material and consumable costs expected to unwind during the fourth quarter and next year. Secondly, we've progressed our initiatives to increase organic growth through our RIGHT WAY 2 growth plan. Whilst we've seen a recent positive improvement in digital inbound lead flow, this is coming from our paid search activities supporting the Terminix brand in particular. In the fourth quarter, we're looking to deliver improved leads for a number of our other master brands, whilst also increasing lead generation from our organic search initiatives. In the fourth quarter, we'll also be piloting the opening of at least 10 new satellite branches. These are smaller branch offices from which colleagues can cater for the needs of our customers and in particular, potential new customers in key target cities. We'll also be assessing the value of a physical presence to the visibility and digital presence of our brands and services. In sales, we're introducing a back-to-basics approach with increased focus and accountability on executing the selling basics such as improving the speed from lead to inspection and to proposal and increasing the average number of sales proposals per sales rep per day. At the half year, we highlighted the need to focus on customer retention initiatives and this program continues to build up as we aim to drive improvements to all phases of the customer experience. We're adding 3 senior leaders in this space in addition to the 40 team members that we've added to the dedicated customer sales team. Customer retention for the overall North American business improved slightly in the quarter to 79.9%. We're also pleased to see a continued improvement in our North American colleague retention, which moved up from 77.8% to 78.5% with good improvements in both service and sales roles. Other actions have included the appointment of a new Chief Marketing Officer and Chief Operating Officer in North America, and the appointment of an interim CFO for North America, whilst the search is underway for the permanent replacement. You may also have noted that we are recruiting for at least one new Board member with specific experience in U.S. network-based services and/or B2C marketing. Thirdly, on the integration front. The program proceeded very much the plan in the third quarter with systems and data integration in another 28 branches with combined revenues of $136 million, taking the total number of branches where we've integrated the systems to 36 with revenues of around $172 million. Importantly, at these locations, there was minimal disruption to operations with customer retention stable and colleague retention, very strong. We're now in a very busy and important fourth quarter for the integration with systems migration continuing in a further 23 branches with a total revenue of around $130 million. Importantly, we're also on track to undertake the first rerouting of our technicians in the integrated branches as well as the piloting of our new service and sales pay plans across 8 branches, affecting over 250 technicians and about 40 sales colleagues. Over the past year, our U.S. business has experienced significant change activities as we've implemented our integration and RIGHT WAY 2 growth strategies. Whilst we remain confident that these strategies will lead to a stronger and faster growing organization, during the first quarter of next year, we'll be reviewing our optimum branch network footprint, which will be informed by the early results of the new satellite branches that will be opened during the fourth quarter. We'll also review the effectiveness of the new technician and sales pay plans being piloted later this year. The review will result in 2025 synergies being pushed out by approximately 2 to 3 months, and we'll update the market on this review at the full year results in March. So we're taking action on many fronts to address costs, to focus on growth, to drive the integration forward and to add talent. There's no change to our 2024 guidance, and we remain convinced about the long-term attractive growth potential of the North American pest control market, which continues to grow at around 5% per annum and in the significant opportunities that we have to capture this growth. Our international businesses, which account for around 40% of group revenue and profit continued to perform well. With that, I'm going to hand back to the operator, and Stuart and I will be pleased to take any questions. Thank you.

[Operator Instructions] Your first question comes from the line of Simona Sarli from Bank of America. Your next question comes from the line of Suhasini Varanasi from Goldman Sachs.

Just a couple for me, please. So it's interesting to see the launch of the satellite branches in Q4. If these work and revenues do improve, can you maybe share some color on how many more you can launch in [ 2024 ], and maybe 100 more, a couple of hundreds more and the costs associated with that. The second one is on the appointment of the Chief Marketing Officer, Chief Operating Officer in North America, can you maybe share some details on who they are, their background, please?

Thanks, Suhasini. Look, on the satellite branches, and so back up one. To drive organic search volumes from the Internet, you need three things in place, all right? You need high-quality content on your websites of which we have lots, hundreds and hundreds of web pages, and we're working hard to refresh and update those. Google has made another change to its algorithm using generative search so it's important that the content of our web pages aligns with Google's most recent algorithm. The second is the importance of 5-star Google reviews. And we've had a massive improvement in 5-star Google reviews across our master brands in the last few months. The third factor in local search and organic search is the location of your facilities. And so what we're doing here is reflecting on the fact that we have got a lot of facilities across North America, but many of those facilities are not in the prime real estate. They're not in the biggest urban areas, the metro areas, and some of them are a little bit further away from dense population. So what we're piloting here is initially in the fourth quarter, we said at least 10, and certainly will be at least 10. They're in prime real estate metro areas. And we'll be putting in real facilities, real offices, with real people answering phones, moving some staff from our other branches. The technicians will still live locally as they do servicing customers locally. And we're hoping that we will see a positive response to that to drive an improvement in local organic search. So the reason we're piloting it, if we were so certain so saying that this would work 100% in all of the cities, then I'll be able to give you a proper answer to how many and over what period. But we'll be piloting over the next few weeks and months, and we're hoping to see some really good positive results. In terms of how many, that's a complex one, because it really does depend on our existing branch network, many of which -- of our branches are perfectly located, some could be optimized. So that's part of the reason we're calling the review in the first quarter. Give us time to see how do these pilots work, are they effective? And if they are, how many more do we think we should put into the network. They're -- I can't give you a dollar number. So we're saying, they're relatively inexpensive. So we're not talking about the opening of massive real estate footprints with huge facilities and massive car parks. We're talking relatively small properties. So we'll be able to give you a much better answer to those questions after the review and at the time of the prelims. On the CMO and the Chief Operating Officer. And Chief Operating Officer is our most experienced senior multisite operator coming from outside of the industry over a year ago. He's been with us, as I said, over a year. A very effective operator. He's been running the region that has been going through the integration. He's been running that extremely well. And we've got him in charge of the overall COO role and delighted with the start that he's made there. Chief Marketing Officer, actually comes from a mixed B2B and B2C background, having worked for quite a number of years in HD Supply and with Brad, the CEO when he was over there for quite a number of years. So very, very experienced, very effective and has made a really fast start. So delighted with her.

We'll reopen the line for Simona Sarli from Bank of America.

So I have a couple of them. First of all, you have been mentioning that there is some cost overrun, but you're managing to contain some of that. What is the total amount of this cost overrun and how much you will be recovering overall next year? Also, of the incremental net synergies that were initially expected for 2025. Can you remind us what is the amount? And also how conservative is your expectation that this will be pushed out only by 2 to 3 months? And lastly, if you can kindly provide an update on PestPac. So in July, you mentioned that 40% of North America colleagues were already using PestPac. How much progress did you make since then? And how much of this 40% is actually Terminix versus Rentokil colleagues?

Yes. So thanks, Simona, apologies. You asked a number of questions there about costs, what does that mean for 2025 synergies. So I'll try and sort of respond in the round, I think. So first thing, we haven't given guidance on 2025. We don't -- we never have at this time of the year given the North America performance we've seen, it certainly wouldn't be prudent to start now. But there are a number of moving pieces, and I'll try to give some color on that. Firstly, of course, we've got the organic growth trajectory, the impact of the integration, both positive and negative. So once we're through the other side, we'd expect growth to improve, but as we've said many times, as we're going through an integration, we get a short-term disruption. So key concern, of course, is the lower route density we're experiencing due to our volume decline and its impact on gross margins and through the integration where customer churn can be slightly elevated. Clearly, that's a concern, too. So we've got a few moving parts there. You referenced synergies. Look, so what did we say previously? We've said $65 million we gave that guidance at the prelims. But then at the interims, we said that of the additional $25 million marketing investment, marketing and customer care investment. In the second half, $10 million of that would flow through into 2025. So you've got a net there of $55 million is where we stand. I'm not really in a position right now to say quite what that looks like in 2025 because, of course, the point of the 2- to 3-month review is to determine what that future trajectory might look like. But there's the sort of the polls around which you can [ have that ] but clearly, there's an impact as we will be delaying. And most of those 2025 synergies are operational synergies so Q4 was always going to be higher than Q1. So it's a Q4 that you lose rather than a relatively low modest Q1. And then, yes, the review of the cost base, we've said we're finding about $22 million of cost reductions through FTE, the 250 heads that we've already done and about $10 million, we think, over Q4 and into 2025 should flow back. So of that GBP 50 million overrun, we've identified about $33 million that will come back the other way. But there's other items moving around as well. I've already referenced the density point from organic growth. There's other moving pieces. We've got small acquisitions, relative growth rate of the product business, which obviously impacts margin and rebuild of bonus provisions and those sorts of things as well. So quite a lot of moving pieces there. Not really ready to give guidance on 2025, but hopefully, there's enough information there that you've got some points around which you can think.

Simona, you said how conservative is the 2- to 3-month period, don't really know. I mean, I don't see -- I don't think it's either aggressive, nor conservative, I suppose, that's our estimate. We'll know a fair bit on the pay plans, I think, during the fourth quarter. We get some incident reaction. As we launch those pay plans, we'll get -- we'll do some feedback questionnaires with the colleagues do they like it, do they not like it? Are we hitting the spot with that? I think in terms of the satellite branches, so we just talked about that, that will take a little bit longer to see how effective that is and whether they're working to plan. So we're probably going to need that January, February. And then we're updating the market numbers at the first week of March, I think the [ pro 11 ], so we're certainly going to need that period in the first couple of months of next year to see what are we learning from that. . I think it makes absolute -- I think of all the change we've put through to the business in the last year or 2, I think it makes absolute sense to take that moment of reflection and review and then start and push on as we go into second quarter. So absolutely no reason to believe it will be longer than the 2 to 3 months that we're calling. But equally, I don't think we're calling it 2 to 3 months and actually think it's 1 month. I think we're going to need all of that time to review those changes that we're making. Your detailed question on the percentage of people on PestPac is not in my head at the moment. That is a detailed question. We, I think I've said before, every night, as I go to bed, what a sad person I am, I get an e-mail from the teams who are integrating the businesses at about 11:00 at night, I get the daily feedback on the integrations. And I have to say, it's really encouraging. They're going really well. So the conversions from mission to PestPac, they've had their issues. They've had their glitches and their gremlins, but they're pretty small and they've been fixed pretty quickly in the main. So delighted to see those revenues coming on. We'll have over whatever it is, $250 million, $300 million of the revenue by the end of the year, that will have been fully converted. And that's a very good reference point. And as we do put more mission, which is the Terminix system to PestPac conversions, they should get easier and easier and easier and easier because the gremlins and the nits and the nats that have come out of the system will all have been resolved. So I don't have the percentage answer for you. But as I mentioned in my remarks, is it's really good to see very high levels of technician retention, branch manager retention, sales colleague retention, very high levels. So that's a good sign. That's a positive. And customer retention is also stable. We're not seeing any significant impact on customer retention as we've converted those. So a really positive set of migrations and integrations we've seen so far.

And...

Go on if you have a follow-on.

So I was wondering also like on the engineering customer retention that you have indicated decreasing its holding. If you can split it a little bit between Rentokil and Terminix. Are you seeing a similar improvement between the two?

No, I'm not that split. And just to be clear, that figure is a slight improvement. We're talking about 0.1%, I think, up from the previous quarter. That's not good enough, Simona. I'm not -- we're not putting the champagne on ice on customer retention at all. That's why we've invested the 40 extra people in heads. That's why we've got 3 senior people across this. We are looking at this with a laser focus on an end-to-end in the customer experience because we've got opportunities to improve that right across the piece. I don't have the split between Rentokil and Terminix in my head. But as a general rule, as I've shared before, the customer retention with residential is always the weakest. It doesn't matter what country, it doesn't matter what brand, what business, it's always the weakest. So that's the area that we have to focus on most acutely. Residential customers who took out the service when they had a problem, who signed up to a contract when they had a problem, who no longer have a problem and therefore, in their minds decide that they no longer need the service. That's our biggest challenge. So trying to convince people that the value of the service when they're no longer seeing the ants in the kitchen or the mosquitoes in the yard that caused them to book us in the first place. But I don't have the split. And again, we'll update more with the prelims.

And your next question comes from the line of Annelies Vermeulen from Morgan Stanley.

I have three please. So just firstly, on the synergy review in the new year. Could you comment on the timing? I appreciate there's a lot going on in Q4 with the branch integrations and pay plans, et cetera. But do you think the review can be truly representative, given you're basing it on sort of that quarter in the low season? And equally, how straightforward would it be to speed up if you make the integration again into Q2 once the review is done? Or do you think it's more likely that you wait until you get past the peak season later in the year to ramp it up again? Then secondly, on the satellite branches, it sounds like this is part of how you're thinking about lead generation. At the last update, I think you mentioned that the group was a little over reliant on paid search and cost of leads. Is this a way of addressing that? And again, with the review, do you think that, that is enough time to see whether those are successful? I don't know how quickly you'd expect to see an uplift from opening those satellites? And then just lastly, very briefly, to the extent that you can comment on it, have you seen much impact to parts of the business from the elevated hurricane activity in the U.S., particularly through early October. Any comment you can give there?

Thanks, Annelies. I think great question. Great two questions. The first couple there related to say how much you're going to learn. Of course, if I knew that, I wouldn't need to do the pilot size but I think we'll know enough to see whether it's working. Is it part of your question, is the satellite branch part of the overall organic search strategy? Yes, it's part of it. Unashamedly, yes, it's part of it. And being local, as I said the answer to my first question, you got to get three things right. The web content, the 5-star review and the location of your facilities. I'm hopeful that we will see enough evidence that, that is an important part or it isn't. And we'll -- all I can do is tell you, we'll update you when we know whether we'll have the definitive answer, and back to the very first question, therefore, how many satellites might you need? I don't know. We're going to know the definitive answer, but I'm hopeful that we will know enough to give you an educated answer to the question as opposed to a guessed answer to the question when we next speak. In terms of the -- are we slowing down and then can we speed back up? Just to be clear, we're not stopping. We're not suspending. There will be a wall of activity, an immense wall of activity that goes on through the first quarter. And I know I've never really done a good enough job of explaining what this integration is all about. But there is a massive amount of preconditioning of data cleansing of training of all manner of things that have to go on to get a branch ready for integration. So we're not going to stop -- we're not going to idle the engine and stop that. We're just not going to push forward into the full-blown integration until we've tested out these things. The satellite piece could have a bearing on the optimum branch footprint. So if it's going to have a bearing on the optimum branch footprint. We better know that before we fully deploy the integration as is currently planned. Does that mean we go to a different region first? Does it mean we go to a different approach to smaller branches and larger branches? So that's quite an important question, but it's not as if we're going to idle the engine down to 0 and then have to ramp it up again. Yes, you're right to point out the high season. We are mindful of the high season. So whatever the phasing of the next part of deployment, we'll take that into account. One of the things we've been careful to avoid is doing a lot of integration in termite territories when the termite season is there. So you should assume, as we recommence integration fully in the beginning of the second quarter, end of the first quarter, we'll be avoiding doing that in big termite territories. But overall, and that's the reason I've stressed how well the systems integrations have been going. We believe we can integrate systems through the high season as we've been doing now for some time. So will we learn enough? Not sure, but I'm hopeful we'll learn a lot. Will it really -- will we have to slow down and speed up? I don't think we're going to lose momentum in that sense. So I don't think that's a major concern. On the impacts of the hurricanes, we've had two, as you know, in tropical storms and hurricanes absolutely terrifying events for our colleagues and our customers. Most importantly to me, at least. We've had no injuries to our colleagues. That's the first thing I always ask about with the teams so everyone was safe. Everyone is fine. We've had 6 colleagues that have either lost their homes or had serious damage to their homes. So we're working with those. In terms of the impacts, well, the one that hit in the third quarter is effectively in the numbers we've presented today. Didn't have a major impact on revenue. Shut down a bunch of branches, quite a lot of branches for 3 or 4 days, but most of that work was caught up and caught up over weekends as well. In the hurricane we just experienced, Milton, similar story. It's come early in the quarter, and we have had quite a lot of branches shut down for up to a week in the Panhandle area. But that horrific though it was, Milton, turned out not to be as disastrous as many people had first feared. So the ability for the business to recover pretty quickly has been proven again. So some impact, not dramatic. And on the offset, it's fair to say, we do have our vector control business. Vector control business basically in this environment puts planes into the sky a few days after hurricanes because hurricanes leave a lot of standing water on the ground in hot humid climates, which means the breeding of mosquitoes becomes a massive problem for the municipalities, particularly in areas like Florida. So we do have -- if we want to talk about it just in financial terms, we do have a bit of a financial offset in that we have vector business that is able to be paid to deal with the potential for a serious mosquito vector-borne crisis for the people after hurricane. So I wouldn't expect, unless we get significant more hurricane activity in the remainder of the quarter, I wouldn't expect the hurricane to have a -- hurricanes to have a material impact on our numbers.

The next question is from the line of Sylvia Barker, JPMorgan.

Two questions, please. Firstly, on growth. So can you just confirm Pest Services was about 1.4%, which is very similar to the previous quarter. And possibly a little bit slower if we take into account the comps from last year. But how do you -- what are you monitoring now in terms of kind of data for Q4, and what you're seeing on the distribution side and on the services side. Secondly, just around cost -- so we've talked about the synergies. I guess the review show whether you actually reduced the scope of the integration to any degree if you think that you needed more of these branches for the organic search, et cetera. But on the cost side, so where are you now on kind of the marketing -- digital marketing spend, could that -- would that need to go up again? Or do you think that you're there? And then any additional costs around hiring your four traditional heads senior leaders? And also, do you need to maybe pay up more in incentives to any of the colleagues? Is there any -- should we think as we think about kind of what's the level of earnings for next year, even if you're not guiding on that yet, kind of what are the moving parts to consider around costs? And then finally, on the refinancing, Stuart, would you be able to just update us on kind of where that is? What the additional costs from that might be into next year?

Thanks, Sylvia. Let me have a go. I'll dive into the middle of it and come back to you, Stuart. On the cost piece, we're not saying anything different on the digital marketing budget at this point in time to that what we've said before. So those with long memories, we originally put an additional $25 million into 2024. We then subsequently increased that to $50 million, of which $40 million would be spent in 2024, and $10 million going into the run rate of 2025. So we haven't changed that. And your question is, might we, could we? Yes, we could, I suppose, but we haven't. And we're putting together budgets for next year even as we speak. But at the moment, we're not calling any material change to that number. Let me remind us what the component parts of that $50 million included. It included a big investment into the brands, top of funnel marketing, brand marketing for the first time for a couple of years, if not more, in the Terminix brand that will continue. That was effective, raising top-of-mind awareness for the Terminix brand so that continues. The additional heads is included. So any additional cost of the sales team is included overall in the investment. So that's not -- and you don't need to add an additional 40 heads. That hasn't changed from anything we've said previously. In terms of the three additional heads, there might be some additional cost there around the margin. I talked about three senior heads that wasn't originally in the plan. But so there's a little incremental investment there. Sorry, you did ask a question about incentives, which I've lost the plot on, sorry. So I'm going to have to skip that one. Stuart, do you want to cover -- and you can come back and ask me what that incentive point was, Sylvia apologies.

It was are you having to pay more to your new people in essence, do you have to incentivize people to a greater degree?

No, we haven't made any changes to the incentive plans at all through 2024. I have, of course, talked about the planned piloted changes to the pay plans more generally for technicians and for sales, which we'll be piloting in the fourth quarter, which I've talked at many, many times before, which in the main will result in lower paid colleagues getting higher pay and some people at the very high end on technicians getting a lower pay over time. But other than that, there's been no changes specifically this year. And at the moment, other than the planned pipeline changes I've talked about, I've got nothing to add to that. Why don't you Stuart take refinancing, and I'll come back and address Sylvia's question on growth.

Yes, sure. So we've got a EUR 400 million bond that matures on the 22nd of November. I think the swap rate is about 3.2%. We are retiring that out of cash, so we're not refinancing this year. We've got plenty of cash sitting on the balance sheet, so we're making ourselves slightly more efficient. So we'll probably be in the market Q2 of next year, I would imagine, because we've got the $700 million term loan that needs refinancing next year. And obviously, rates will be what they are, it will be in Q2 next year, but those are -- no news this year, apart from retiring that bond.

Yes. And not so much additional color on the growth, 1.4% for Pest Control, 1.4% for North America in total. What are we monitoring on a daily basis? Well, what are we monitoring? On a daily basis, we are monitoring lead flow. So we're monitoring inbound lead flow from paid search. We're monitoring inbound lead flow into the websites from organic search. We're monitoring technician leads from our field force on a daily basis. We are monitoring the sales close rate, that's the rate at which sales people sell those leads. We're monitoring the average dollar value of those leads as they come in. We don't have daily visibility yet of customer retention. We have daily visibility of customer cancels, which is a different thing because when a customer cancels, you still have the opportunity to save it, hence, the investment in the sales team. We should have daily visibility of customer retention, I'm hoping between now and the end of the year. So I think it's one of the bigger changes heritage Rentokil colleagues such as myself, whereas historically, we typically didn't really need to look at data much more than once a month and data was highly predictable on the commercial side of the business. On the residential side in the termite, it's much less predictable, as I've said many times. So we've moved the organization to that daily tracking, daily visibility. I'd like to get off daily, if I'm honest, because it's a bit we're not retail. We're not a supermarket. I'd like to get it to be more of a weekly cadence and the predictability of the numbers has got better, and that's the plan over the coming weeks and months. There's not much of a story on distribution. Distribution's been quite lumpy as it tends to be in that business. Business is performing reasonably well, but it is a seasonal business and then we have -- we're coming to a season now where the manufacturers of chemical products for pest control and lawn and ornamental will start doing what they call the early order program. So we'll see how that goes. We have it every year, but basically, the suppliers offer discounts to customers to buy that product this year so they can get it in this year's numbers. So we'll see what the early order program looks like this year, but no significant story on distribution.

Your next question is from the line of Rory McKenzie from UBS.

It's Rory here with two questions on behalf of Nicole. Firstly, now that you're about to pilot the new pay plans in the 8 branches, can you give more detail on what those pay structures look like and how they compare to the legacy plans? And then secondly, following up on Annelies' question about what you learned from the Q1 review. I wanted to ask about the branch network. I think you'll be up to 59 branches migrated by the end of Q4, and then you have the 10 new satellite branches. Can you just talk about how that's spread geographically? Is that all focused in specific states or regions? Just trying to think about if in the Q1 review, you'll be able to assess a kind of fully transitioned model in any local areas or whether it's more about the national portfolio of branches overall?

Yes. Thanks, Rory. Yes, very good question. I'll deal with the second one first. So in terms of the 59, 60, whatever it is, branches by the end of the year, those are all in one region. They are all geographically in one region. And the pay plans that we will be piloting will be piloted in that region. The 8 branches is a subset of the 59 branches. So yes, increasingly, we'll be able to answer the question that you just posed, which is okay, now you've done the systems, now you've done the pay plan, now you've done branding. What are you -- what is the -- what does the post world look like in that region? I'm not sure we'll be able to answer that question fully at the time of the prelims. But yes, so we're not gone scattergun with these branches. They're all in the same geographical region, and that's where we'll be piloting the pay plans. As to the satellite, no, they're not in the same region. They are in a selection of prime metro cities around the United States, which we've selected, handpicked, if you like, deliberately because we're testing something different there. We're testing to see if you have a physical location closer to dense and affluent potential consumers, does that drive up a better local presence than having branches elsewhere. So that's a different theory we're testing there, Rory, that I'm sure I haven't got the list in front of me, wouldn't tell you if I did. But I'm sure there'll be at least one, if not two, in the region that we've integrated the systems. On the pay structure, I think it's fairer to tell the colleagues the details of the pay structure before we tell the analysts to be fair, and we haven't told the colleagues yet. But broadly speaking, as I've so shared before, if you look at the service colleagues. You've got a bell curve of distribution. And on the left-hand side of that bell curve, you've got a reasonable chunk, the 20%-ish, haven't got that chart in front of me, of colleagues that whose pay needs to improve, whose pay needs to increase. I'm not going to use the phrase underpaid, but there's certainly our belief that they need to earn a higher basic pay. And we believe that, that will significantly improve the retention of more junior colleagues that have only joined us in the last 6, 12, 18 months. So that is part of the investment we've always talked about. At the right-hand side of the bell curve, there is a smaller cohort population who are paid very well. And they understand they're paid very well. And that's an accident of history. It's an accident of the territory that they operate in. We've started to talk to those colleagues about the implications of change. We will not implement those changes overnight. We will do it on a phased basis. We'll do it on a gradual basis. But there will be some colleagues at the far right of that bell curve who will be disappointed because they won't be able to earn quite as much as they used to be. On sales, a little bit different. And so just to address a question that wasn't asked, does that mean that your very best technicians are going to get a pay cut? No, that's not what I'm saying at all. Does that mean that our most highly paid technicians may see some reductions, that's what I'm saying. And just because you're the highest paid in technician land, doesn't necessarily mean you are the best. It may mean you've got a very productive route. On sales, a little bit different. On sales, if you're the highest earning sales colleague, that probably means you're one of the best sales colleagues because you're selling more than anyone else or you're selling at the high end of the cohort. And so the sales pay pan has been designed to continue to incentivize our highest best-performing salespeople to continue to be our highest best performing salespeople. But in the main, Rory, the main change that we're doing is to give more fixed compensation and less variable, but still having a very healthy level of variable pay because it's the variable pay, which incentivizes the discretionary effort of salespeople. So I know I've sort of half answered your question, but the detail needs to be given to the colleagues before I can share it publicly.

Thanks for answering the unasked third question.

I think we've got time for one more question, please.

And your final question comes from the line of James Rose from Barclays.

Just one for me. Coming back on to customer retention, which is creeping upwards, but there's still a way to go. Could you run us through what's been done so far to improve the service levels and processes across Terminix? And the second part of that is, do you think it's not necessarily possible to address service levels processes until we get through the branch integrations? Do we think about it being a fairly long time scale to walk up that customer retention rate as you go through the integrations?

Yes. Thanks, James. Very good question as a small, I must say. The honest answer is, you said one question so I could do an [indiscernible] on this one and I'll try not to. Look, customer retention, you've got to track the customer life cycle journey from the very first point in time when a customer touches our website or phones a call center all the way through to the very end of that life cycle when the customer says, I'd like to terminate. And through that journey, there are multiple, multiple, multiple touch points. Every time they call us and want to change their appointment, every time they call us and want to query a bill, every time they want to book an additional service, every time we want to tell them, we can't turn up this afternoon because something's happened. So in terms of process, we are looking at that end-to-end process because there are so many opportunities, which is why I put 3 people across this. Why three people? Because there's big chunks of this process. I mean it is the end-to-end customer experience. So there is a ton of things that we can do to impact positively the customer experience. And over time, we believe they will have a positive impact on customer retention. Some of that will relate to systems. So some of that has to be post integration and some of it relates to process. And again, some of that process has changed post integration, but a bunch of it's changed pre-integration. So I think it's a great question to finish on because for me, it's arguably the single biggest most important thing we've got to get right, hence, the investments that we're seeing and hence, the focus we're putting on it. If you look at the far end of the book end, what happens when you get to a customer wanting to terminate, that's a change we've made. So we had 80 people in the sales team. Over the balance of the third quarter, we've added 40 people. So we've now got 120 people in the sales team. And we would expect, over time, to see an improved rate at which customers who were terminating, actually decide to stay with us. So there's one absolute tangible thing that we've made -- changed. Our customer service protocols for our technicians is another area that we've changed. That is a process change. That is something that we're doing pre-integration and post integration. And that is more about getting a single playbook across Terminix and Rentokil colleagues, Western colleagues, our LED colleagues, a single playbook approach to what are our customer protocols, what do we do when we get to a customer premise, how should we deal with customers. So that is a bit more of a slow burn because we've got to get kind of 20-odd thousand people across the group, all thinking in the same way and acting in the same way. So I'm conscious I haven't really answered your question in detail. There are a bunch of changes we've made, bunch of investments that are going in. We are working with experts in the field. We're not just doing this on our own. So well, what save offers should we offer to a customer when they want to leave, what's working, what doesn't work? We've invested in a system, a customer service, CX system, working pretty well. We think we can get more out of that system. So honestly, it is a -- I think this one is a multiyear journey. It will get a bit disrupted by the integration, which is why I focus so heavily every night before I go to bed and see, well, how is customer retention doing and the branches that we've migrated. I don't think we'll be able to do the complete migration and have 0 impact on customer retention. I will say at the moment, it looks pretty good, but we still got the rerouting, et cetera, to go. So incredibly important area, keep asking the question because I certainly do. Lots of things we've already changed, but I think we're really just beginning this journey. I'll just remind us for what it's worth, the rest of the group operates at about 85%, 86% customer retention. And here we are in North America across the estate operating at 80%. I'm not sure we can get to 86% anytime soon in America, but somewhere between 80% and 86% over the next 2 or 3 years, we certainly should be targeting. Rentokil North America used to be about 82.5%, I think it was at the time we did the acquisition, something like that. So the opportunity is material and it's incredibly important that we seize it, that's why we're putting the investments in, and that's why we're changing the processes. Thanks, James. Thanks, operator. Thank you very much, everyone. Appreciate the questions. We get to do it all again for our American colleagues. Everyone's welcome, I guess, at 1:00, I think, our time. And so now we'll leave it there. I look forward to updating you again with the prelims. Thank you very much.