S4 Capital plc (SFOR) Earnings Call Transcript & Summary

Welcome to Bonhill, our new headquarters here in London for MediaMonks. We've got a quorum, and I'm delighted to declare the meeting open. I'm joined here in person by -- on my right, Mary Basterfield, our CFO. On her right, Colin Day, who is Chair of the Audit and Risk Committee. Sue Prevezer, on my left, the incoming Chair of the Nomination and Remuneration Committee. And on her left, Miles Young, Non-Executive Director responsible for Culture. Paul Roy, who is our outgoing Chair of the Nomination and Remuneration Committee; Rupert Faure Walker, who's our Senior Independent Director; Scott Spirit, our Chief Growth Officer, who is in Singapore; Wes ter Haar, our Co-CEO of Content, will join us virtually. And both Scott and Wes will be making a presentation virtually on AI after the statement and a few comments from Mary. As you are aware, this is a hybrid meeting of physical and virtual meeting, which enables all our shareowners to participate either in person or virtually. And I'm pleased to welcome those of you joining us in person today, and I continue to value the participation by all of those of you who are joining us remotely. Before providing a brief update on the business and on current developments, we would like to take this opportunity to explain how the formal business of the meeting will proceed and how you will be able to vote on the resolutions proposed in the notice of the Annual General Meeting. The notice of the Annual General Meeting was sent to all shareowners on the 9th of May 2024, and at the same time was made available on our website. I would, therefore, ask your permission to take the notice of the meeting as read. On that basis, we shall be taking a poll on each resolution. And I now propose formally that each of the resolutions as set out in the notice of the AGM is put to the vote of the meeting. Resolutions 1 to 15 are proposed as ordinary resolutions and require a simple majority of the votes to be cast in favor to be passed. Resolutions 16 to 20 are proposed as special resolutions, which require at least 75% of the votes to be cast in favor to be passed. Following the Board's decision to streamline its size and its composition and to be more in line with U.K. listed company practice, you will have seen our stock exchange announcement released just now that Naoko Okumoto has decided not to offer herself for reelection and will be stepping down from the Board with effect from the conclusion of the meeting. Accordingly, Resolution #8 is withdrawn. In addition, we have previously announced on March 27 and April 29 that Scott Spirit, Christopher Martin, Wes ter Haar, Victor Knaap and Paul Roy are also stepping down from the Board. I would, therefore, like to thank all of our retiring directors, both the nonexecutives and the executives for their commitment, contribution and indeed, all their hard work for your company during their tenure on the Board. Turning now to voting. All resolutions at our shareowner meetings are decided by poll. And our share registrar is present as the poll scrutineer to count the votes at the end of this meeting. The directors are unanimously in favor of each resolution and recommend that you vote in favor. On a poll, each member present in person, by corporate representative or by proxy is entitled to 1 vote for every share held by them or the shareowner whom they represent whatever the case may be. a person entitled to more than 1 vote need not use all his votes or cast all votes he uses in the same way. All shareowners, proxies and corporate representatives here today should have been provided with a poll card on arrival. If you don't have either of these, please raise your hand now and a steward will assist you. Please fill in the full name and address of the shareowner in block capitals. If you are a third-party proxy or a corporate representative, please write your full name in the space provided. In addition to that of the personal company you've been appointed to represent. If there are 2 more persons present representing a joint holding, the person whose name appears first on the register of members should complete and sign the poll card. Their votes will be accepted to the exclusion of the other joint holders. Please indicate your vote for each resolution by putting across in either the for or against or vote withheld box. Please note that the vote withheld is not a vote in law which means that the vote will not be counted in the calculation of votes for or against the resolution. Please ensure you sign your poll card and hand it to a steward as you leave the room. For those shareholders joining us virtually via the online platform, the list of resolutions should have appeared by now in your screen automatically although depending on your device, you may need to select the voting icon from within the navigation bar to see the resolutions. You should also see for or against and vote withheld voting options. To register your votes, please select one of these options. If you change your mind, you just need to select another option. You can change your mind as many times as you wish up until the close of the poll. There is no submit button. When you click on your preferred voting option, the icon will change color to indicate that your vote has been submitted. I will close the voting once we have concluded the Q&A session that we're going to have. And I will give you a clear prompt later in the meeting to warn of the close in voting. As a reminder, Resolution 8 has been withdrawn and is not being proposed. So the poll is now open. We will also take the opportunity for general business questions and discussion from the floor and online. Online voters can start submitting questions now using the online meeting platform, which will address -- we will address later in the proceedings. If you wish to ask a question, please click on the message icon located at the bottom of the screen if viewing from a phone or top of your screen if viewing from your laptop. Type your message into the Ask a Question box and click the arrow button to the right-hand side of the message box. So I'd now like -- having gone through the procedures, I'd like now to bring shareowners up-to-date on current developments. 2023, which was our fifth full year, was a difficult year with slower market growth and continuing macroeconomic uncertainty. The first half saw a mixed performance with less momentum and an anticipated second half uplift, which did not materialize amidst continuing client caution together with economic and geopolitical challenges. Overall, we saw clients very much focused on the short term particularly in relation to larger transformation projects, which resulted in longer sales cycles along with lower regional and local opportunities. Our stated 'whopper' strategy of building broad-scaled relationships with leading enterprise clients continues to drive our business, and we ended the year with 10 against our target of 20 such relationships. We remain focused on a disciplined approach to costs and the number of months and operational cash generation. Trading in the first 4 months of the year continues to reflect the impact of volatile global macroeconomic conditions, general client caution, continuing amongst technology clients and a reduction in activity with some of our larger technology services clients, although there has been some improvement in the profitability of the content practice during this period. We continue to develop our larger scale relationships with leading enterprise clients and are increasing our focus on margin improvement through greater efficiency, greater utilization, billability and pricing. We maintain our targets for the full year. And as in prior years, financial performance will be significantly second half weighted, reflecting both our normal seasonality and an expected improvement in market conditions. We want to take this opportunity to remind everyone of our company's definitive and differentiated strategy which continues to be based on 4 core principles: Firstly, digital only; secondly, data-driven; thirdly, faster, better, cheaper and more adding more for AI; and then finally, a unitary structure. We say our company as it is both your company as our shareowners and our company as your management team. We are still tightly aligned with you with around 40% of our share capital connected to our Monks and/or our directors. We are purely digital because that is where the growth is, even more so in a post-COVID-19, 24/7 always on digital world, fueled by AI. Our data-driven business model derives insights from first-party data and platform signals, which fuel the creation, the production and the distribution of digital advertising and marketing content through our data&digital media planning and buying, our programmatic and our performance executions. We continue to expand our capabilities in technology services to fully provide digital marketing transformation services for our clients. The transformational shift to artificial intelligence, or AI, is incorporated into our strap line of faster, better, cheaper and more to reflect our enhanced ability to integrate AI across our uniquely unitary operations. We remain confident in our strategy, in our business model and in our talent. These together with scaled client relationships, position us well for growth in the longer term with an emphasis on deploying free cash flow as and when appropriate, to improve shareowner returns, particularly now that all significant merger payments have been made. New business activity continues at healthy levels, particularly with the current focus on hyper-personalization or personalization at scale, which has been accelerated by AI. Marketeers are starting to split their activities into two areas or levels. Upper funnel, strategic, big idea creative which is in turn, then activated by a lower funnel creative content platform or factory. This involves significant organizational change and is being embraced by clients that are both successful, and those facing disruptive change, all being accelerated by the deprecation of third-party cookies, which now has been postponed a little bit into next year. New business wins in the first 4 months include Burger King, Panasonic, FanDuel, AliExpress, Decathlon, Santander and ICBC. In addition, the company continues to capitalize on its strong AI positioning, winning multiple exploratory excitements as clients experiment and explore applications and develop use cases. These are currently focused on visualization and copywriting, on hyper-personalization at scale on media planning and buying, on general client/agency efficiency and democratization knowledge across the firm. Since our last AGM in June of 2023, our company has reduced the number of months in the company to around 7,600 compared to around 8,600 at this time last year, a reduction of approximately 1,000, reflecting the ongoing progress made in aligning our cost base with the demand we are seeing from our clients. We will maintain a disciplined approach to managing our cost base with an increasing focus on driving efficiency across the company as well as utilization, billability, and pricing. For 2024, at a practice level, we continue to expect content to show a profitability improvement, reflecting the benefit of cost reductions made in '23 and in 2024. Data&digital Media will show a similar top line and bottom line performance to the prior year with some modest margin improvement, while the outlook for Technology Services remains challenging, and the performance will be lower following a significant reduction in activity with some key clients. But the company as a whole, given the current outlook for technology services and wider market uncertainty, we continue to target like-for-like net revenue -- net growth to be down on the previous year -- sorry, not target, like-for-like net revenue to be down on the prior year with a broadly similar overall level of operational EBITDA as 2023. And this is a result of cost improvements made last year. The comparatives with 2023 will continue to be relatively harder in the first half and will ease in the second half. We continue to expect the year to be heavily second half weighted with improving end markets and our normal seasonality. Our net debt is expected to fall in the second half of 2024 reflecting positive free cash flow and is significantly lower combination payments. Our targeted range for the year-end, that's a net debt remains GBP 150 million to GBP 190 million. We continue to aim for financial leverage of around 1.5x operational EBITDA over the medium term and noting that our key covenant is 4.5x. Over the medium to longer term, we continue to expect our growth to outperform our markets and operational EBITDA margins to return to historic levels of around 20%. Our talented people have responded positively to the challenging trading conditions and our drive, in turn, for efficiency. We have continued to make progress in three areas of our ESG strategy : in People Fulfillment; in Our Responsibility to the World and in our One Brand. We are grateful for the ongoing support of our shareowners and remain confident in our strategy, in our business model and in our talent, which all together with scaled client relationships position us very well for growth in the longer term. As we announced on the 27th of March 2024 and to continue aligning our governance to a more traditional Board composition, Chris Martin, Victor Knaap, Wes ter Haar and Scott Spirit, will all be retiring from the Board at the conclusion of today's AGM, with Wes ter Haar becoming a Board observer. They will retain their current roles within the company, including any involvement with the Executive Committee. In addition, as announced on the 29th of April 2024, Paul Roy will also step down from the Board at the conclusion of this AGM as a part of reducing his commitments to commercial and corporate interests. Finally, in the consideration of the Board structure as well as our time commitments, Naoko Okumoto will not seek reelection from shareowners at today's AGM and will step down from the Board at the conclusion of the AGM. I want to take this opportunity again to take -- to publicly thank all of our retiring directors, both executive and nonexecutive for their contribution, their commitment and their hard work to S4Capital during their tenure on the Board. Now before we go into Q&A, our CFO, Mary Basterfield; and Jean-Benoit Berty, who has been recently appointed Chief Operating Officer. Mary will give a presentation on our trading update and Jean-Benoit will be available for questions when we get into the Q&A session, after which, Scott Spirit will give you an update on the market and on our clients. He will then be joined by Wes ter Haar to talk to you about AI. So with all that, at this stage, I'll hand it over to Mary for our trading update.

Thank you, Martin. Good afternoon, and thank you for joining us today. I'm going to take you through our financial performance for 2023 and recent trading update for the first quarter. I'll start with the financial highlights for 2023. After 4 years of strong growth, 2023 was a difficult year. Net revenue was GBP 873 million, down 4.5% on a like-for-like basis. We delivered growth in the first half, but the third and fourth quarters were more challenging. Profitability came under pressure due to lower revenues, but operational EBITDA was in line with revised targets at GBP 94 million. We made significant cost reductions to deliver an operational EBITDA margin of 10.7%. EBITDA improved as these cost reductions took effect, and we delivered GBP 57 million in the second half compared to GBP 37 million in the first. Adjusted operating profit was GBP 82 million, and adjusted earnings per share were 5.7p. We finished the year with net debt of GBP 181 million, reflecting expected combination payments made in the year. This was at the bottom of the targeted range due to tight cost control and slightly lower-than-expected combination payments. Leverage was 1.9x. Moving on to 2024. Our first quarter performance was in line with our expectations. Net revenue of GBP 186 million decreased 15% reported or 12% like-for-like. Operational EBITDA for the quarter was also in line with our expectations reflecting both lower activity levels and the benefit of cost reductions made in 2023. Clients remain cautious in the near term with ongoing market uncertainty but we are managing through this, and we maintained our full year targets. At a practice level, we continue to expect content to deliver an improvement in EBITDA and margin driven by cost reductions made last year. And data&digital media to perform in line with 2023 on both top and bottom line, with some margin improvements. Technology Services has a more challenging outlook, and we expect both lower revenue and EBITDA following a reduction in activity with some key clients. We continue to maintain a disciplined approach to cost management, including headcount and discretionary costs as well as a focus on utilization, billing and pricing. Given the outlook for Technology Services, we continue to target a decrease in like-for-like group net revenue versus 2023, with a broadly similar level of operational EBITDA. The comparatives with 2023 will be tougher in the first half and ease in the second. We also expect 2024 to be heavily weighted to the second half, given our natural seasonality and expected improvement in end markets. Net debt at the end of March was GBP 206 million or 2.2x operational EBITDA. This is after GBP 10 million of contingent consideration payments in the first quarter. We maintain our full year net debt guidance of GBP 150 million to GBP 190 million. Thank you very much. With that, I will hand over to Scott.

Thank you very much, Mary. Good morning, everybody. So from a market perspective, if you look at our addressable markets, it's actually a contrast a picture -- the digital media market is very healthy with a strong end to 2023 and a good start so far to 2024. Growth is expected to moderate over the course of '24, but the major platforms will still grow by double digits. Obviously, that contrast with our own performance so far but we are still seeing caution from our clients, particularly technology clients. If you look at Nektar's Q1 results, for example, their revenues were up 27%, but their sales and marketing expenses were down 16%. For technology services or digital transformation, this is a more difficult market, which slowed in 2023 and is flat in 2024. We actually had a very strong 2023. So those negative market trends have caught up with us in 2024. Most of our client categories were stable in quarter 1. Technology is still by far the main category, representing 43% of our revenue. We saw a decline in financial services category, and that was largely driven by the decline in spend with one of our clients in the tech services practice. We saw growth from our existing clients in auto and fashion, which saw us gained traction in those 2 categories. Whilst the majority of our top 10 and top 50 clients did, in fact, show growth, we saw declines overall in the average size of our clients, driven largely by continued softness in technology clients and a specific large financial services clients in tech services, which has reduced spend. Given our overall revenue decline in Q1, our top 10 clients declined slightly less, our top 20 in line and our top 50 slightly more. When we look at the distribution of our clients via spend, you'll see we have 7 clients spending above GBP 5 million versus GBP 8 million last year, with Mondelez being the one which dropped down below that threshold. On a positive note, over the course of 2023, we saw a significant decline in the number of clients in the 0.1 million to 1 million category, which tends to be project-based clients, local client relationships and new business but this has stabilized in Q1. And we do see a healthy pipeline of pictures and wins, including Burger King, Panasonic, FanDuel, AliExpress, Decathlon, Santander and ICBC in Q1, which should support our H2 weighted outlook. We're also excited to welcome Justin Billingsley as our new Chief Growth Officer, who will be focused on driving new business. One of the key areas of opportunity for us is artificial intelligence, which remains the main topic of conversation with clients and partners. And for this, I'm going to hand over to Wes, who will give us the latest update. Over to you, Wes.

Thanks, Scott, and hey, everyone. Yes, has dominated the conversation of last 12 to 16 months, it's quite interesting to look at the bull versus bear case. On the bull side, it will be AI story. We're seeing people reference it as potentially the most important technology ever created, and there's quite a lot of messaging towards this being an industrial revolution. From a bearish case, the total revenue generated through AI is only slightly in what supplies the business has generated out last year. If you look at it through our space, if we go to the next slide, it gets really practical. We sell about 75% of our business into the CMO organization, and the CMOs are under pressure to deliver productivity gains, especially in the current macro. So the expected productivity gain sits at around 15% to 20%. It's opening up quite a lot of practical opportunities for us because if we go to the next slide, I think we've gone -- we've done a good job of positioning ourselves in this space. You can go back to early last year where we were fast and first commit the change. We've been repeatedly fast in our go-to-market, both in some of the first really visible work and also some of the biggest changes from a workflow perspective. That has made us Adweek's first-ever AI Agency of the Year, which is -- and I'll show some work later on quite important because it does show that this is also a great tool for creativity and customer experience. This isn't just about productivity guidance. But we've also managed on the other end of the spectrum to position as a strong consultative partner. So according to the Business Intelligence Group, we were awarded for AI excellence, both in technology innovation and strategic planning. As you can imagine, that's quite key. This if you look at the space through the lens of our clients, you can broadly split it into the back end and the front end. And if we go to the next slide, we'll talk a bit about what we mean with the back end and why our consultative practice and our technology expertise there is so important. So for our clients, typically, their ability to promote their product, their services looks something like this. It's a lot of steps, a lot of silos. It starts with a moment of insights or a piece of customer data or market data, and that gets translated through strategy and creative to content at scale to activation and media, to the markets. For many clients, this is a process that has become quite complex. We have clients where this process from start to finish at the global level can take up to 18 months. And to be honest, if we go to the next slide, for most clients, it actually looks something like this. Lots of brands, lots of regions, lots of channels and realistically, if you look at the current landscape, we sort of hit what I would call the limits of solving the complexity of global marketing purely through minimal labor, which is opportunity. And if we go to the next slide, I think the technology assets currently in market, but also clearly, the foremost of the -- it's that we can start rethinking what marketing actually looks like. Still using the same building books, we still go from inside the strategy to creative to quantitate scale, adaptation creation, transforation -- to deliver into performance, but there has to be ways to do that where we're speeding it up. We're helping our clients go to market more quickly. We're helping them save some spends. And mostly, we're just taking out some of the complexity that, in many cases, has made it really difficult for global marketers to respond in a timely manner. To do that in a way that combined, if we go to the next slide, both our talents and our knowledge and expertise in technology, we've launched amongst low. We did this at CES at the beginning of the year have had really good feedback. And this is really the orchestration of talent and technology to help solve the complexities of the global marketing org -- we're doing this in very close relationships with our technology partners like NVIDIA, Google, Meta, Adobe, we'll have an announcement on that in a few days in time. And it's a positioning, I think, where we as a company are quite strong and quite uniquely positioned from a marketing perspective that knowledge of technology and that the partnership with those companies, I think, is something that sets us apart from most of our competitors. If we go to the next slide, -- if you look at the current landscape, there really are 2 goals and questions that the global marketing work is being asked to answer, one is they need to deliver growth and they need to deliver possible growth, which puts a lot of additional pressure on measurement and making sure you're making data-driven decisions. We believe growth is available by speeding up 4 of these blocks through technology enablement. Insights and strategy, this is really about taking back control clients owning their data set, the ability to get the insights more quickly, but also for a much deeper sets of data being able to automate delivery, that's quite an important part of our go-to-market. If you look at our media team, it's very technology native and then continuously drive additional performance because we're collecting more data. We're generating more signal. And because of that, we're helping our clients get more value from their media buying. And then if we go to the next slide, our clients aren't just being asked to live in profitable growth. They are often also being asked to save some of their spend. We heard so Martin referenced these content factories realistically, consolidation allows for innovation. So we're helping our clients consolidate the four cones in supply chain all the way from creative to adaptation. And because of that, it's possible to generate savings. Partly that's in the simplification by taking out many agencies and going down to a single partner, and part of that is because of technology, which is a mix of automation in AI. If we go to the next slide, I'll hit these relatively quickly, but this is more of a general note that in each of these steps, we have packaged both our talent base and technology in ways that really help our clients deliver both on the growth and savings components. Insights really interesting. We're doing this for a variety of clients now, where we're capturing much deeper sets of data and using that. So we will culture our insights more quickly, allows our clients to be better in how they're going to market and what they're marketing. If we go to the next slide, we have a tool that we brought to market specifically for strategy, for persona. This is where we're combining large third-party data sets bring that into our clients' organization, combine it with first-party data, and it allows us to build digital twins of our clients customer bases, the ability to talk to those digital wins in real time. We have statistical proof that this replaces quality and quantitive testing. Again, this is about speed, granularity of information and this is the ability to move in a more agile fashion processes that used to take 3 months, especially testing can take a long time, is compressed down to date, which is really quite powerful. We go to the next slide, we have 2 slides that really talk about speed and scale of content creation. We have own [ TrackingFox ] and digital twins very closely aligned to NVIDIA and Omniverse. And then if we go to the next slide, this is really about adaptation. What's interesting is that we're getting closer to the original performance of visual advertising because a lot of that promise is built on the amount of content you can create and in this case, generate. You need more content to be more personal, more targeted, more contextual. Historically, the question has been, is the juice worth the squeeze? Our ability to now speed that up is actually quite impactful because it allows our clients drive more value from the media buy, which brings us to the next slide. As a technological people in the room will know, when we started S4, we brought in MediaMonks and MightyHive. MightHive is a very strong digitally native technology-first media and data company. We're doing a lot of media automation for our clients because if you're generating 10 assets instead of 1 or even 1,000 instead of 10, it's really important that the traditional ways of media buying and planning and activation go through that same type of technology transformation. And I think that is, in general, something that will have a lot of impact on our industry in the next few years. And then the last part is about performance. The interesting note here is we are running some tests with our media and technology partners, Google and Meta, which will be showcasing [indiscernible] which show meaningful upswings in performance from existing media buys through the use of our technology. So that is the back end. Main takeaway, we're helping our clients solve the speed, scale and spend complexities of the global marketing work. That's a combination of our talent, which is digital native and best-in-class in this space with technology where we're very closely aligned with the major players in our industry. If we got to the next slide, the other part of the AI story is the frontend. This is the customer experience. There is [indiscernible] view of what will happen. And this help in view is the future of customer service is 2 AIs endlessly e-mailing each other back and forth alternately demanding and refusing. Refunds, I think there is a point of view that's a little more you talking. And if we go to the next slide, and there's a little irony to it, which is that AI actually allows us to make the customer experience fundamentally more human. Our ability to understand the intent of some of these behavior within an ecosystem, our opportunity to reply to that intent almost in real time, be highly responsive, do so with high levels of fidelity, more personal and sequential, I think it's really powerful. We'll see a lot of meaningful and I think in some cases, unimaginable change in customer experience. And we've been on the front end of some of the standout examples of that. If we go to the next slide, launched the $1 million whooper contest for Burger King about 3 to 4 months ago. As far as I can tell, biggest creative generative AI comparing that's been launched so far and for a standout brand. What we did here really speaks to how AI will help brands scale their interactions with the customer, but also create personalization at scale in real time. So in this case, we have a brand model, which means we have a model that's trained specifically on the brand. That model allows us to generate perfect looking whoopers. It also is trained on the voice of the brands, that's both tone of voice and the actual voice. There's official voice that's used in all commercials. And it allowed us to create a campaign where people were able to generate the whopper of their choice with complete freedom and then submit that for the competition, created a lot more flexibility. Normally, these [indiscernible] are very restricted. In this case, it was completely open interesting takeaway. All of the moderation in this work was also done by AIs. A great example where AI allowed us to do something at scale that normally would not have been possible purely through traditional ways of working. Really successful piece of work. It's interesting to see the stickiness of this type of customer experience. And a lot of that's driven by the personalization component. If we go to the next slide, another example of our team being involved in some really standout work. This is for Dove. Dove has been running one of the most famous campaigns in advertising for 2 decades now called Real Beauty. To celebrate that milestone, it was important to have a point of view on generative AI. Our team was the top leader in that space. And I think that some of the industry's best work to really understand the complexity of both the [ bias ] and the machines, but also our ability of humans to talk to machines in different ways and actually get from the machines, the diversity that Dove has always been focused on in their advertising. So a beautiful piece of work, this campaign, and this was a key part of that has just been announced as a short list at [indiscernible] Titanium, which is a category for truly breakthrough advertising. So we're excited to see what this does in [indiscernible] about 1.5 weeks, but also the rest of the year. One more note,if we go to the next slide. Burger King, the iconic advertising and marketing brands where we've been able to do some really groundbreaking work. I want to be really clear that a lot of the exciting work that happens in this space is because of data. And I think our data practice and how native it's been technology from day 1 is really key there. [indiscernible] is a great example of how data through the lens of AI technology opens up completely different ways of working. This is for Starbucks, and it's a great example of how AI actually allows us to represent the customer in our decision-making. So we run the mobile app for Starbucks in quite a few markets. As you can imagine, it's one of the most used mobile apps on the planet, which also means you get a lot of customer feedback, especially in the app stores across different countries and regions. What we did in this case was feed all of that content into an AI customer analysis process and use that analysis to impact our product road map. The goal being how do we get from our current star rating to a plus 1. And instead of doing that purely through manual labor, which would be a piece of research. And then based on a percentage of data make, I guess, a very well position guess of course, through expertise, but still a guess based on a smaller percentage of data. We were able to do that on a massive data set. And it had a meaningful impact on the star rating and the customer experience in a really short time frame. This is, I think, key to a lot of what we're doing internally. This isn't only a transformation model for our clients. We're also making sure we're empowering our own teams and our own workflows with this type of additional value add because it just makes us better at our job. If we go to the next slide, we'll end it on what happens next. Again, we've, I think, been quite good at providing updates every few months or so. We'll do a big update in time. If we go to the next slide, we have a variety of updates and launches planned with our technology partners. We're hosting our Monks.Cafe together with NVIDIA. It will be, I think, a hotspot for a really interesting combination of C-level folks, both on the marketing and from the technology parts of our clients' organizations. We'll talk in depth about AI and how we go from pilots to scale and to profits. I think the biggest conversation in our industry is about scale now when it comes to AI, and we'll be announcing specific use cases with Google, with Salesforce, with Meta, and we have an exciting announcement with Adobe as well. So we'll be continuing, hopefully, to position as a leader in this space, and we're excited to talk about that next time we meet. And with that, I will hand back to the room in London.

Thank you, Wes. Thank you, Scott. Thanks, Mary. Now the AGM is an opportunity for Sharon is to express their views and to ask questions to the Board and your Board committed to an open dialogue with our shareowners, and we're pleased to have this opportunity during the meeting to engage with you directly. We ask, however, that your questions relate solely to the items of the business before the meeting. If you do have a question, a microphone is available for your use if needed. And it would be helpful if you could state your name and indicate if you are a shareowner or if you are representing a shareowner. And if so, if you could indicate the name of that shareowner before you ask that your question. [Operator Instructions] Now before I take the question from the floor or online, we would like to respond to three questions that we have been sent in by TEA. That's the engagement appeal. And their questions are on youth financial literacy and Scott Spirit is going to -- as our most youthful person, is going to respond to it. Scott, can you respond to those questions?

Yes, sure. I think there are four actually. Thanks, Martin, and I appreciate the youthfulness comments. So I think the first question was around how informed younger generations, especially Gen Z, are about personal finance. I think despite your compliment, I'm actually technically a boomer. So I think I would be expected to be dismissive of Gen Z and their knowledge of personal finance. But I think the reality is actually when I looked into this a bit today, they are far more curious and educated than you'd think. Responded survey from NASDAQ saying in the U.S., 73% of Gen Z owned stocks and 47% of them own cryptocurrency, which is obviously very popular given their understanding of technology. They're heavy users of technology and especially social media. They follow financial influences, which are very important to them. And I found a report from the U.K.'s Royal Mint that said almost 1/4 of Gen Z follow financial influencers on social media. Although that's also an area of caution given that this does open them up to potential scams or get rich quick schemes. So I think actually, the answer is that Gen Z are quite well educated and curious about personal finance. The second question was, given our difficult economic circumstances, what impact do you think, if any, could individual investors deliver to a company's overall performance? I'm not sure I fully understood that, but I think I would encourage investors whether they're individual, retail or institutional investors to engage with companies that they've invested in, it doesn't need to be in the form of activism. But certainly here at S4, we spend a large amount, particularly I spend a large amount, of time communicating with our investors, both large and small, and have received valuable insight, advice and client leads actually from investors, so I would encourage investors to communicate with the companies. The third question was, in our opinion, what matters most to the younger generation of individual investors? And I think that's very clear actually. I think despite the fact that in the past couple of years, ESG seems to have been given slightly less priority by institutional investors, that's certainly not the case with Gen Z. And I found the Morgan Stanley report that said 90% of Gen Zers believe that companies have a responsibility to address environmental and social issues, so that's a very key thing for younger individual investors. And then the final question was around what more, if anything, can companies like us do to educate, empower or engage individuals on basic personal finance and investing? Whilst I'm not sure it's our role to do this on our behalf, it is certainly something we can and do, do on the behalf of our clients, particularly those in the financial service sector. So we work for a variety of clients in that sector. And a lot of what we do, do is digital marketing, particularly social marketing, to educate younger generations and inform them of kind of products and services that our clients offer. So that's those 4 questions, hopefully answered.

Thank you, Scott. And I think that there is a representative of TEA here. Okay, does that answer your question? Anything you want to comment on?

No, I think TEA as an organization is very much trying to encourage and promote engagement between retail investors and corporates, so thank you. Yes, TEA as a normalization has been set up. It's an altruistic organization, but nonetheless commercial, but it's trying to promote greater engagement between retail investors and corporates. I'm an IR professional myself. And I think too many IR are focused very much on the institutional investors because they're the ones who are easier to reach and easier to get onboard with your corporate narrative. And we think there's a lot more that can be done to reach out and engage with retail investors. So it's an agnostic platform, a two-sided platform, designed to promote those altruistic aims, so thank you very much for the answer to those questions.

Good. Okay. Anybody else? Yes, I think the one at the back there first.

[ Roger Mayhew ], and I hold my shares through Hargreaves Lansdown. I've got two questions. One is just a very practical one about your website. I don't find your website easy to navigate compared with other investor relation websites. Things like the financial calendar, the way it's laid out, it isn't easy. And I think that it needs to be improved, especially for a company like this.

On that, I'll just do it one by one. If you've got any specifics, maybe have a word with Caroline, and we can get it maybe -- and we spend a lot of time on this, actually. And it's disappointing to hear you say that because the time and effort has been spent on it has been quite significant. So if we can go through some of the specifics, that would be helpful. Thank you.

The second thing is, obviously, the multiple on the share at the moment is somewhat disappointing. I think there is huge opportunity in AI from what I've heard in your presentations. But I don't know what scale is required. Obviously, I think S4 has some very valuable assets. Otherwise, people wouldn't have come along last autumn to try and buy them cheaply. And I think it's quite right that you didn't sell them to others at a very discounted price. I wouldn't have wanted fronted that, and I don't think that would have been a good idea. But the question is for the medium term, given the scale of opportunity and the level of investment, which is needed, do you think it will be beneficial for all shareholders potentially, and I'm not saying immediately or even in the next year, to merge with a larger company where, on decent terms first, we get shares in a larger organization. I mean I look at companies like Trade Desk in the States where, over the last year, the share performance been very good compared with ours. So I'm just trying to understand what -- and our risk in this business because there are benefits of AI. But AI, it could be, is not so good for some areas of business because people will think they don't need agencies. I don't think that is the case given the complication and the level of expertise. So I'm just trying to see whether the Board will look at possibly merging the company into a large organization.

Well, I think just to comment on that. I mean the Board will look at anything and everything that's credible from our point of view and not anybody else's. On the question of scale, it depends on what you're talking about. I mean the Trade Desk that you mentioned is a very different business. That's a platform business. We're not a platform business. We're not a tech business in the sense of taking, I'd put it, bets or making investments in technology. We're users of technology. And on AI, I mean we have to make investments in streamlining. I mean as you heard in the presentation, we're making investments in speeding up the workflow. I mean essentially, Monks.Flow is about automating processes, which are hitherto be much more complicated. And those can be in the areas that we indicated in the statement and you heard in the presentation. So to act to grow and develop in that area, you don't need the huge investments that have to be made by the hardware companies, the picks and the shovels, if you like, like an NVIDIA or a Microsoft or an Adobe, an Oracle or a Salesforce or any of the 6 platforms. So it's not a huge investment. If you look at what we have in terms of AI, it's in its infancy. NVIDIA, who is the biggest hardware manufacturer, I guess, in some sense, in that area, has built capacity out. And the key issue is to establish the use cases. And you saw in our 3 use cases or 3 examples of use cases in the presentation with Burger King and Starbucks, et cetera, and there are others. But in order to capitalize on that, what we need is for clients to embrace those use cases at scale. And to date, what we've seen is, I would say, audits, workshops, your starters for, let's put it crudely, $0.5 million or $1 million or $1.5 million or $2 million. The big digital transformation projects that come out of that will take more time. Now having said that, as we speak, there are companies, particularly in the U.S., where it's the biggest scale market in the world. People are starting to look at this area. And it's that sort of second of the 5 areas that we mentioned in the statement, the personalization, where there is definitely a strong searchlight or microscope being trained on this area. So with the deprecation of third-party cookies, with personalization at scale driven by AI, in particular, you're looking at what I refer to as Netflix on steroids. You are seeing that. Now the question is, in order to do that, companies have to go through major organizational change. Some of them are very successful. So people take the attitude of if it ain't broke, don't fix it. Some of those people are challenged. They might be being disintermediated and they're disrupted in their own industry and they will probably risk more and turn everything upside down in order to do it. So it will take some time. I don't think it will take a long time. But almost as we sit here or as we speak, people are examining these opportunities and they are significant and at scale. But to do that, it doesn't demand huge investments from us and we can do it within the P&L we have. If you said to me one of the areas where we need more capability, I think we need more capability probably on the tech services side, which is the smallest practice we have. And in Asia Pacific, those markets were not at the scale that we would like. So ideally, if you could find a company which was a tech services business in Asia Pacific, you would fulfill those two criteria. If there was anybody else out there that we felt would be good to merge with, as you put it, we would do that. I mean there's no barrier to doing what we think strategically would make sense. And we've laid out yet again our strategy and our thinking on our positioning, digital-only, data-driven, faster, better, cheaper and more, and unitary, so we've laid them out. But does that answer your question?

Who would you regard as out there who are not your smallest, your greatest, competition in the areas that you work?

We always describe it. Scott, do you want to describe our competition? I mean we always referred to 3 levels of competition. Do you want to fill that out?

Yes. Sure. So I mean there's certainly the holding companies, which have a very large market share in the marketing and advertising side of the business. So that's the Omnicoms, Publicis, WPPs of the world. Then you have the consulting companies. So that's particularly Accenture, but some of the other larger consulting companies as well. And then I think you have some of the more direct competitors that really do or try and do what we do that specialize particularly in digital marketing or digital transformation. So that might be a company like Brandtech Group or DEPT on the digital marketing side or Globant or Thoughtworks on the digital transformation side. So it's pretty broad. But I would say from a scale perspective, you have very large companies. I think we're probably, in our industry, considered the midsized companies. There's also a lot of very small companies as well, and many of which do very well. So I don't think scale is the key factor in success or failure.

I mean in talking to clients, if you take those -- for example, we did a big event with Google here in London earlier this week and actually with Google in Ireland last week at the Dublin Tech Summit. And if you talk to clients there, I wouldn't say the larger the holding companies have any particular advantage. They're seen, if I can say so, as being a little bit cumbersome. I mean we know that NVIDIA, for example, finds it easier to navigate a company of 10,000 to 12,000 people rather than 100,000 people, which is probably naturally the case anyway. So I would say that the holding companies necessarily are, in fact, talking to clients one last week, I don't think, which uses the holding companies and ourselves. They think agility is absolutely key, and flexibility. So I don't think that the big holding companies have a natural advantage. The tech companies, the sort of companies like JB, like Capgemini, EY, Globant Accenture, Thoughtworks, EPAM, and Perficient, which has just gone private, they're probably a little bit more competitive but very much in the tech services, CIO, CTO slot because we deal with sales and marketing. They deal with IT. They're trying to get into sales and marketing, we're trying to get into IT. Those 3 functions are unifying. I would say within the holding companies, there may be particular digital agencies that are smaller, which probably, by definition, a little bit more agile, maybe they. I think probably the specialist, and Scott mentioned Brandtech and DEPT in particular, we probably come across them more. But in talking to Google this week and last week, I would say, specialization actually, which we've positioned ourselves as, as AI specialists or first movers, is probably the most important positioning. So I'm sure there are people out there that would make sense to work together with, but there's nothing obvious that comes out from that analysis, I think that you say immediately that's something you want to do. Some of the private equity houses have good entrants. But they've seen pressure as well, and they don't mark their investments to market. And so they have their investments in their private portfolios at historical multiples which, when you look at the market, is not reflected in the market anymore. The other little subset to that is they're raising money against unmarked-down portfolios as well, which I think raises some issues. Anyway, anything else you want to raise on that? No. Okay, I think we have a gentleman in front of you.

[ Michael Claricky ], I'm a private investor. I have two questions really. First one is on your balance sheet. I note that your net debt is at the higher end of your expectations? And if the tech sector doesn't recover as quickly as you hope, do you have any plans for diversification?

Diversification of?

Of area, unless you've got a very strong coverage in America. But your Asia Pac seems quite weak. Although I do note you said you've won business with Ali Express. That's the one question. And the second part has to do with your large language model.

Let's deal with one at a time. So Mary, I answered that a little bit on the APAC on a previous question. We do think that should be stronger. So that's an area where I think we would agree with you, okay, on that. Do you want to talk about -- if you can [ add some more ].

Complete with the drilling, yes. So on the net debt, as I said earlier, we've maintained our full year guidance of $150 million to $190 million. What we saw at the end of Q1 was our expected natural seasonality. I think it's important to remember, we have now completed materially all of the contingent consideration payments that are due. So we do not have those payments going forward. And therefore, as we go into '25 and '26, we will see the debt profile change significantly.

Does that answer the first question?

It does. The second part was about your AI and your large language models in terms of the [ plan you're approaching ] for your customers. Can you elaborate your ownership on the IP on the large language models?

Yes. That's not an issue for us. That is an issue for media owners and for clients. And it's an issue that is very much high up on their agenda. You've seen with the media houses, negotiations take place in relation to IP. And I think most people think that, that is an economic discussion. In other words, there is a price, which we've seen it before, for example, in Australia between Meta and Murdoch on news, paying for news. So that's an economic thing. But I mean, you're right in the sense that some clients have been hesitant to use LLMs. We don't make LLMs. We use them. I've been hesitant to use them because of the IP infringement, copyright risk. And in negotiating contracts around AI with our clients, procurement departments do raise that as an issue, and it is still an issue. I wouldn't say it's hindering. I think I'm right in saying one of the platforms actually did, I think it was OpenAI, offer at one stage to take the IP risk themselves. In other words, they took the risk off. It is an issue, but I think it's an economic issue that will be resolved in time. And we don't make LLMs or create them, we use them. Okay?

But in your use, do you own the actual structure of it?

No.

Okay.

I mean in creating the sort of workflow tools that you -- Wes, do you want to add any more from a Monks.Flow point of view?

Yes. You're hitting the main ones that we built on top of the foundational models that are created by big tech companies. The big tech companies have indemnified the use of their models broadly. So yes, we're utilizing the technology. We're tuning the models for brand-specific goals, but we're not building models ourselves, at least not the foundational ones.

Okay. Right, I think there was one over and then we'll go over there.

[ Nick Steiner ], a private shareholder. Quite a lot of the questions I had have been answered. But again, on your sort of strategy worldwide, and again, looking particularly at sort of Asia, the question of language is Mandarin, the Indian languages and all these sorts of things, are they a distraction or something that you can actually work through?

Language differences are gone. I mean with AI and with the [ Monks ] technology, gone, finished. I mean there's not a barrier.

So how does it work for a Mandarin speaker who doesn't speak English?

In terms of what?

Well, in terms of sort of the work we do. I'm just trying to sort of understand it.

Obviously, we have Mandarin speakers. We have English speakers. We have people who can't speak English and people who can't speak Mandarin. But in terms of the technology and translation, I mean that whole barrier has gone.

All right. Okay. That's good. The second question, I was looking at Page 19, Not guilty: exposing bias, and particularly using AI to pick up -- I'm not quite sure how -- generated scripts on abuse at home and so on and so forth. And I know this wouldn't be core, but does that mean that AI could actually assist with online abuse and that type of thing and obviously, companies you work with have to deal with this? Is it something that this company could do?

Is it something that we can do?

Yes.

Wes, do you want to talk about online bias? I mean we are worried about bias and we have an ethical policy. Do you want to explain it?

Yes. I think that's the standard watchout with the models that are trained on massive amounts of data, but within that data is, I would say, humanity's bias made real, to a large extent. So that is important, why it's so important to have people in the loop. The rules and regulations that are rolling out mostly focus on the bias part, making sure that the models aren't left to turn on their own and, because of that, make bias decisions. There's some really interesting components of that in financial marketing, for instance. We're not allowed to have, in the U.S., the models make advertising targeting decisions because you might have a model be biased against certain demographics. So it's definitely an important space. I think we were very early to market with our ethical guidelines. We have a really strong internal legal team that also actually helps our clients in this space as well. So I think we're well set up to navigate the complexities of the landscape. But yes, the way we use the models that there is low risk of that impacting our output in meaningful ways.

Yes. Just on language, it just reminded me, I think it was at the Google I/O conference. There was a demonstration of a woman speaking to a man. The woman was speaking in English, and he answered in French, I think it was, in German. So I mean, these things are, I think, languages in terms -- I mean there were devices ironically before AI. There was a Chinese application that you have. You just stick an audio plug in your ear and you hear instant translation. So I think language as a barrier, both between people and in terms of communication, has been reduced drastically. Anything else?

No. Thank you.

Thank you. Over there.

[ Scott Wallace ], a long-time shareholder, now something called a corporate representative, unfortunately. Just first question, there's two questions, first question is about -- lovely to see you, Martin, by the way, and long may you reign, but it's about succession planning. I mean is there several young Martin Sorrell types to take over from you? Or is there a Sorrell from Brazil in incubation?

You haven't seen any young Martin Sorrells today?

Well, they may not have incubated yet.

Do you want to answer that, Paul, on the line? It'd be invidious for me to answer it. Paul, do you want to respond to that, if you're there?

Yes. Thank you. I mean I think, at the same time, we'll talk about the combined role that Martin has with Chairman and Chief Executive. I mean we feel that we continually review the overall composition of the Board. We consistently update our Board's skills matrix and our skills and knowledge, independence and diversity. We do, and have done over a period of time, have very continuous discussions about succession. We have no immediate plans or thought of any formal succession plan as far as Martin's position is concerned. I think if you're clearly listening to him responding to our questions today, it's very obvious that not only has he been integral to the creation of S4. And the committee and the wider Board considers, obviously, that he continues to be very much so and that it's in shareholders' interest for him to lead the company. And at this point in time and at this point in the development of the company and the changing environment that we've been discussing here today, that it is in the interest of share owners for him to continue to lead the company in a combined role.

To your answer, I mean if you look at the annual report and you look at the ExCo, you look at the practice leaders, you look at the geographic leaders, there is a significant and talented management team there.

Thank you. The second question is, could you give us a little bit more of your Philip K. Dick vision of where we'll be in 10 years' time with this generation of AI and the connected world? I mean I find it a little bit frightening that as Genghis Khan and Napoleon, Hitler or Stalin took over this whole system, like what happened to the Canadian truckers when their bank accounts turned off. Could you give us some view of that?

We aren't aware of the Canadian truckers. I never heard that story before. What happened with the Canadian truckers?

I think the union for some reason, the government, they stopped them drawing money out of their accounts.

I'm not sure what quite relevance that has to us.

Yes, but I'm just talking, in terms of a connected world, what you see in 10 years' time how we'll be living.

Well, I think it's quite difficult to figure that out. I mean it's difficult enough to the end of this year, let alone 10 years' time. Yes, if I can be philosophical about it, I think the world is very different to what it was a few years ago or for the last 30 or 40 years. I think it is more fractured and fragmented. And I think you have to take a much more -- I mean the answer to your question, I think, is you have to take a much more selective geographical approach. One of the questions we got asked is why we got so much in North and South America. And I think the answer is because that's one of the best places to be. The other best place is to be, in my view, for what it's worth, is Middle East and Asia, with a qualifier on China because of Taiwan. If you're big in China, do you really want to be bigger? We're not big enough, so we want to be bigger. But there are other places in the world where in Asia, you should be. India is the obvious one. I mean the election result perhaps makes that a little bit more tenuous. But still I would bet on Modi and bet on India, certainly, as an alternative or a complement to a big position in China, and then as Vietnam, Indonesia, Singapore, Malaysia, Philippines and Thailand, et cetera. It's a new Asia rather than old Asia like Japan, Australia and New Zealand maybe. So I think you have to pick your growth spots. I didn't mention Europe, and I didn't mention it deliberately, because I think Europe has got some big issues, not Canadian trucker issues, but difficult issues. So that would be one thing. The other thing would be AI does give us the opportunity to run businesses, governance, institutions, whatever, more efficiently. So I think there are two things that really top of mind with clients as a result of that fragmentation and a world which is going to grow slower, have stickier inflation, have higher interest rates than we've been used to, I think that will continue. The first is pick your markets. So I do see that North and South American block being really important and the other parts of the world that I mentioned. Africa is probably too volatile for my taste than other people's taste. That's one thing. And the second thing is AI, as Wes has laid out the sort of things that we see in our own industry. And our own industry is actually, along with the health care industry, being most affected by AI so far. There are very few other industries that have been as dramatically affected.

I mean we've been in a sort of cyber war for some years now, probably over 20 years, and we only hear what's happening to us rather than what's happening to our opponents. But that's quite a worrying thought of disrupting your technology and also tracking; another thing, turning off hospitals.

We can go back to Wes and ask him what he thinks about it. My own view is that the big are going to get bigger. NVIDIA, you saw again, there was another lurch upwards today, and it's trending towards being the world's most valuable company in a very short period of time. You are going to see those shifts taking place. So the 6 platforms, the 3 Western and 3 Eastern, plus NVIDIA, plus Microsoft, plus Apple, plus Adobe, Oracle and Salesforce, I think are going to be the powers in the sort of AI space. The regulators will not be able to keep up with them unless the regulators go the extreme route of actually treating them like AT&T was treated or like Standard Oil, by actually breaking them up. Unless they do that, I think what will happen is these companies will continue to expand because the resources, to go back to the previous question, are so needed and so high that those companies -- Musk must will be another player in that, along with Tesla and Space X. So I think the big will get bigger. The issues you raised about cyber and bad actors, nuclear brought bad actors, okay? You see that with Oppenheimer. You will have bad actors in this space. Self-regulation will be really important. So those big companies I mentioned, I think, are going to have to self-regulate because I don't think the regulators can keep up with them. And you do see signs of that happening. So with power comes responsibility, and they're going to do that. So that will be -- Wes, do you want to add any more about how you see the future from our industry point of view?

Well, I think we're heading into the next version of the industry. And that happens now and again. I think this is probably at least similar in scale to the impact of the Internet on our industry, but it's going to be in a massively compressed time frame. So if you look at the impact of the Internet on marketing and advertising, it really took a good decade plus for that to roll out at scale. And of course, now it is pretty much the defining part of our industry. This will have at least similar impact, but the speed of change will be compressed. And I would say it's at most a 5-year process where Internet impact at scale was close to 10 to 50. It will move a little slower because enterprises are about risk mitigation, so the big global enterprise will move a little less quick. So it's probably in 2- to 5-year window that we'll see massive amounts of change in our industry.

I think once clients start to move, we hinted at this before, I think there's a lot of people thinking about things, once one client in a vertical starts to make a major change, FOMO, fear of missing out, will move at light speed in this case. I mean I can't remember -- the personalization scale model, I can't remember a model as potent as that for many years, if any. This is really huge stuff. But to do it, you have to turn your organization upside down. And if you're doing okay, if it ain't broke, don't fix it. If you're disrupted, maybe there's a little bit more impetus. But once people move, others will follow. The other thing is the suppliers like NVIDIA are concerned about use cases. I mean the task that we've been given by the suppliers, if you like, for 2024 is to develop the use cases. Once you get 1 or 2 in a vertical moving, it's like -- what did Boris say when he resigned? The herd has moved. The herd will move.

Is Mistral a challenger?

Well, Mistral is now linked with Microsoft and OpenAI. It's part of the pivot that Microsoft has made. It's part of that wave. But it's interesting, Mistral, OpenAI, G42, which was the Middle Eastern company that Microsoft bought, I think, in its entirety, if I remember rightly, are all examples of people who didn't have the wherewithal to develop the LLMs. And through their lot in -- and Microsoft have done a brilliant job of pivoting without acquisition, in theory. I mean they've invested in the companies with a minority stake, and then they put money in. And the money, in a way, round trips back to Microsoft because OpenAI and Mistral are going to use their technology. Anybody else? Yes, the one behind here.

Thank you, Chairman, the man behind. I'm a great admirer of entrepreneurs like you, Sir Martin, having built a company from scratch. I'm also a great admirer of Elon Musk, who's done pretty well the same thing. And incidentally, I have to vote at their AGM, vote my shares, to decide whether to pay him $50 billion this year. But that's not my question. What I cannot find in your annual report, which has a lot of corporate information, is a clear list of your clients. I can't pin them down. You've talked about the [ sub ] people and the coffee people, but I'm not clear who they are. And finally, Sir, can you pick up these shares from crawling along the ground?

The answer to the last one is we'll do everything to do that. Scott, can you run through our client list?

Yes. I mean part of that is deliberate because some of our clients prefer to be confidential. But we do work for most of the world's large tech companies. So pretty much any technology company you've heard of will be a client of ours, particularly, as I think you're all aware, that Google, or Alphabet, I guess, the broader Google company, is our largest client. When it comes to financial services, we have strong clients there. We work for Visa, JPMorgan and others. When it comes to consumer goods, we work for Procter & Gamble, we work for Unilever. Wes has showed a case study there from Unilever. We work for Starbucks as well. We work for Mondelez still. When it comes to auto, you're aware of our large relationship with BMW in Europe, but we work for various other auto companies around the world. It's BMW Mini actually. We work for some of the EV manufacturers in China. So that's a selection of our clients. But the reality is, apart from when we show cases like Wes did, I think increasingly, our clients prefer the work to be confidential as they see it as a sort of competitive information.

Being a bit more -- I mean Scott has mentioned Google or Alphabet, I guess there's one big tech plant, which we're NDA-ed on, which we can't mention, but it's one of the biggest. You might guess who it is. Meta, we work for. He's mentioned BMW and Mondelez. We work for a big cosmetics FMCG, which is NDA-ed. But if you look carefully at our website, I think you'll figure out who it is. Another big financial services company on the tech services side, which again is NDA-ed, but you can work out who that is, too. HP, we also work for, that's not NDA-ed. And then you go to Walmart, T-Mobile, Amazon, Disney and PayPal. Those are our top 13. They're about 55% of the business -- 50%, call it, of the business. And I think Scott mentioned before, tech is about 44% of our business. So we're heavily tech focused and, in our view, should continue to be so despite the fact that the tech companies have been perhaps, going back to your last comment, which is one of the reasons why we've had the compression last year, the tech clients have been reducing their marketing spend at the time when the ad revenues from the platforms that they own and develop have been actually increasing. So we don't hide the clients. I mean they are out there. You do need a microscope to find them out. Anybody else?

Sir, there's a question we've received online from Sheryl Cuisia, who's also a representative of The Engagement Appeal. She asks, what is the Remuneration Committee Chairman's view of the average U.S. CEO pay being $16.7 million versus the U.K.'s average of GBP 3.8 million?

Paul, do you want to comment on that?

Certainly. I think it illustrates that there's a fair proportion of the U.K. PLC being under remunerated. It's quite interesting that we have made, and I'm sure a lot of other corporates have, a fair number of representations to various bodies with regard to sort of best practice on U.K. remuneration policies, including proxy agencies and so forth. And one of the key points that I think a lot of respondents from the U.K. have made is that there needs to be more of a leveling up, particularly with regard to the competition for talent and particularly with regard to the fact that, if you take S4 as an example, we are competing for talent globally and in places like technology as well as marketing where the demand is very great, and it is extremely highly remunerated particularly so in the U.S. and less so here and indeed in the rest of Europe. Does that answer the question?

It's triggered another question from the gentleman here.

I think it depends a bit on the industry. But the only retort to that would be this, why is it, if it's such a competitive market, that so many more CEOs of U.K. companies aren't, let's say, getting GBP 3 million. Why don't we see so much more news of them being bid by American companies being paid $20 million? I mean it's a reasonable question.

I mean to be fair, there are one or two companies that have switched to America and where compensation is [ sorted ] as a result. I think there are 3 or 4 of those switched their listing to America. And in fact, there was an article last week about one of them where, as a result of that, the compensation have changed quite significantly. I mean it's an interesting question. I think also, in the context of NVIDIA now, is worth more than the whole of the FTSE 100 -- actually, the whole of the U.K. stock exchange, I think it is, not just the FTSE 100. So okay. All right. That's it. Okay. I just have to remind you all to vote. So for those of you voting online, I would like to remind you to cast your vote in accordance with the instructions given at the start of the Annual General Meeting. If you have not already done so, the meeting will be concluding shortly. For those of you voting in person, our registrar should now have collected your poll cards. Have we done all that? Anybody, poll card are there, please, thank you. You're raising your hand. Thank you. Thank you, [ Lorna ]. [Voting]

Okay. I now propose the resolutions as set out in the Notice of the Meeting with the exception, of course, of Resolution 8, which has been withdrawn, as I explained earlier. These are shown now on the screen together with the proxy votes received prior to the meeting. Do you see that?

Do we have a bigger screen?

Yes, you can come forward and look at them here at the screen, if you like. All these votes will be posted on the stock exchange shortly. When, [ Tim ]?

I think later this afternoon once we collected then the votes that they cast in the room. It will be released this afternoon.

Okay. All right. Ladies and gentlemen, the poll is now closed. And the provisional results are that all resolutions are carried. The results will be available on our website, as I just mentioned, and announced to the stock exchange, as also mentioned, in due course. So that concludes the meeting. Thank you for attending today's meeting, and we wish you all a very safe journey home. Thank you very much. Thank you.