Hello, everyone, thanks for joining us. I'm Jon Bone, investor relations director here at Reckitt, and it's great to have so many of you online for today's event. Before we get started, I'd just like to draw your attention to the usual disclaimers around cautionary statements relating to these materials and those in the Reckitt campus, which I'd request you read in due course. These Reckitt Focus On sessions are really about giving investors a deeper look at how we're evolving the business, the capabilities we're building, and why we're confident in the model. Today is about digital science. As well as showing you more of how we're moving Reckitt forward, these events are also a great opportunity to hear from the people who are driving it. Today, we're joined by our CFO, Shannon Eisenhardt, who will set the context of the discussion and outline how digital science is acting as a key enabler for us in delivering our ambitions. Our Chief R&D Officer, Dr. Angela Naef, will explain how digital science is being integrated across our product innovation life cycle. Bastien Parizot leads our AI enterprise development at Reckitt, and will talk through how we're integrating AI solutions to drive efficiency and prioritization across our innovation function. Our Chief IT&D Officer, Nigel Richardson, will explain how we have built a scalable and advantage platform across our business. Our agenda for today is split into three parts: the why, the what, and the how. Shannon and Angela will set out the context and our fundamental strategy around instilling digital science across our innovation platform. Bastien and Angela will then take you through each stage of the project life cycle, with examples of how we're using this science across Reckitt's power brands. Nigel will discuss our foundations and journey to build our platform. Now, I hope some of you have had a chance to have a quick look at the virtual Reckitt campus. As we have done in previous Focus On events, we wanted to supplement today's presentation with some additional content. This will showcase how we're using digital science across the product life cycle for many more of our power brands. Following this main stage presentation, I would encourage you to explore the campus, where you'll discover some incremental content to show what we're doing and further bring these technologies to life. With that, let me pass to Shannon and Angela to set the context. Thanks, Jon. Welcome to everyone joining us today. Before we hear from Angela, Bastien, and Nigel, I wanna take a step back and frame how this fits into Reckitt's bigger picture. We've been very deliberate over the last couple years as we shape into a simpler and sharper Reckitt. We have a focused portfolio built around 11 power brands, all with leading positions in their categories, long-term runways for growth, and attractive earnings models. This matters. This curated portfolio of power brands with leading consumer equities allows us to successfully innovate. This means we're competing where we can win with brands that have real consumer relevance, trust, and pricing power, which is even more important for providing value to consumers in a pressured environment. What you'll hear today is how digital science acts as an accelerator to support the development and expansion of these brands with great futures and also long histories, which brings a significant benefit from years of proprietary consumer and scientific data. Digital science powers our ability to innovate faster, at greater scale, and with higher success rates. It's this enhanced innovation capability that gives us confidence in delivering our 4% to 5% growth guidance. We've been equally focused on how we invest behind that growth. We've set out a clear program supported by our four focus areas to both reduce our fixed cost base and enhance our capabilities. Digital science is an important element of this. It's not just about the benefits to our cost line. It's about strategically integrating better ways of working across the business. While we're using innovation as the most visible lens in today's presentation, what you're really seeing is a change in how decisions are made across Reckitt. Innovation is simply where this operating model shows up first and most clearly. As we introduce solutions across our functions, our simplified operating model enables faster adoption. You'll hear today how we're driving greater productivity, as well as the savings we're generating through digital science and our scale-up efforts. The results we're seeing underpin our confidence in delivering our target of driving fixed costs to below 19% of net revenue by the end of 2027. Digital science is embedded into our journey to deliver this. What's important to remember is that growth and efficiency are not trade-offs. They reinforce each other. The strategy we outlined two years ago centers on core Reckitt and driving efficiencies across our focused business to deliver long-term value creation. Our commitment to strengthening our core functions across our supply chain, technology, and innovation is not new. Nigel will talk through the journey we've been on to build a smarter and faster platform that now enables us to create and deploy bespoke solutions across all of our functions. We've talked before about significantly increasing our R&D investment around eight years ago to secure a growing pipeline across all of our power brands. Over the last few years, we've seen launches come to market as a result of this step-up in investment. Today, we feel we have the right level of investment to support our pipeline for years ahead. We're now developing greater efficiency on that investment as we optimize for our strategic priorities and utilize the digital science solutions we'll talk about today. This investment, combined with the effectiveness we're driving in freeing up our R&D teams to focus on the work that really matters, is translating into that pipeline strength, driving our power brands forward at a greater pace. The purpose of today isn't to introduce new targets or financial data, it's to give you visibility into how we operate, to build confidence in our ability to deliver against our long-term financial commitments. What you'll see today is how we're supercharging our playbook through digital science. At its simplest, this is about three things: speed, scale, and superiority. Each of these are critical in driving the growth of our power brands. Angela will talk through the outputs we're seeing today in terms of launch cadence, project size, and pipeline growth. You'll see this come to life across the full innovation life cycle as we ideate, develop, launch, and activate premium products across our chosen categories. We're gonna demonstrate how we're becoming more data-driven, more predictable, and less reliant on a few big bets. When you put that together, we're driving important structural shifts that enhance our confidence in delivering growth. We are shortening cycle times, improving launch success, and being much more disciplined in how we prioritize our portfolio. This has positive impacts across our P&L. Each of the examples we'll talk you through have delivered real benefits to our gross margin, whether that's premiumization in Vanish, the launch of completely new categories for Lysol, enhanced efficiency in manufacturing Gaviscon, or driving distribution savings with Dettol packaging. These results are already happening. These are not future capabilities. They're embedded in how we work today and delivering tangible outcomes across growth and cost. As you go through the session, I'd encourage you to look at two things. First, the individual examples, what we're doing and how it works. Second, and more importantly, the system behind it, because that system is what gives us confidence in delivering 4%-5% growth, in achieving our cost commitments, and in building capabilities that are genuinely world-class. With that, I'll hand over to Angela, Bastien, and Nigel to bring it to life. Great. Thanks so much, Shannon. Good afternoon, and thank you for spending your time with us today. First, a quick intro on myself. I have been leading R&D at Reckitt since 2020 and have had a career in R&D at several multinationals and across multiple industries for over 25 years. Over that time, I've had the opportunity to be part of and lead many new technologies. A key element across all of them is the focus on building the culture of new, innovative ways of working alongside the technical capabilities. Today, global R&D at Reckitt is powered by over 4,000 scientific and technical professionals. We're collectively driving a real cultural shift and reinventing both how we integrate and embed digital-first approaches. This change goes beyond R&D. Digital science is reshaping the way all functions involved in innovation work together at Reckitt, from consumer insight to regulatory approval to manufacturing scale-up, and this did not happen overnight. We've been on this journey since 2020. It has been about six years of building proprietary data, training brand-specific models, and genuinely changing the way thousands of people work. I'm here to tell you about why rising consumer expectations are redefining how we innovate. Our role in innovation for R&D is centered around consumer-driven, science-led opportunities. It's about good ideas and how quickly we can prove them, how confidently we can scale them, and how consistently we can repeat that success. Consumers today expect more and often all at once. They want superior performance and value for money. They want speed and credibility. They want products that feel personal. That raises the bar for innovation. That is the context in which we talk about digital science at Reckitt. What I'm going to share today is how we are fundamentally changing the way innovation happens. This change goes beyond R&D. Digital and AI solutions are shaping the way all functions work at Reckitt. Last year, Ryan talked to you about the winning playbook and the role it plays in creating and growing categories across our iconic portfolio. You'll have heard about our consumer obsession, superior innovation, executional excellence. Today, you'll see examples across our portfolio that have sustained growth, such as Lysol, Gaviscon, and Dettol. What unites them is the power of digital science and its capability that enables their strong performance from development to launch to now. Having done this for a very long time, I know that traditional R&D cycles are long, linear, and driven by physical experimentation. The pace required to meet current consumers' expectations needs new ways of working, especially at global scale. At the same time, we operate some of the most trusted brands in the world across nearly 200 markets and under complex regulatory frameworks. Our challenge is very specific. How do we innovate faster and better? That question is what led us to digital science. Digital science is all about speed, scale, and superiority. It enables us to prove value earlier, fail faster when something won't win, and scale what works with more precision. That is what underpins both our growth ambition and our confidence in sustaining it. Let me take a moment to deep dive into how we are progressing across each of these three dimensions. First, speed. By augmenting physical experiments with virtual experimentation, predictive modeling, and AI-supported decision-making, we significantly reduce the time spent on lower value iterations. Across multiple stages of innovation, we are seeing up to 70% time savings on lower added value tasks, freeing scientists to focus on what matters most, meaningful science and future innovation. Ultimately, speed enables us to be better and deliver high quality innovation earlier and more reliably. I wanted to reinforce the bold transformation we've been on. Three years ago, we were performing fewer than 100 virtual experiments a year. By that, we mean virtual experimentation through digital chemistry before testing the chemistry in the real world. Since January this year alone, we've now performed more than 10,000 virtual experiments, saving time and effort, but more importantly, expanding our design space. Now moving to scale, a key differentiator in how we make innovation happen. Today, digital science solutions are deployed across all of our power brands and in more than 20 markets. More than 2,000 colleagues across R&D and marketing have been trained on these capabilities. Where we're seeing the biggest impact is our pipeline. Scale is creating a more efficient pipeline through increasing average project size. In 2026, this has grown by around 25% year-on-year. When we talk about average project size, we mean the incremental net revenue expected from that project. It matters for growth too. Third, superiority. Digital science allows us to test more options earlier, screen out weak concepts faster, and focus investment on the ideas most likely to win. We plan for a significant portion of our growth over the next three years being driven by innovation. Our power brands pipeline, measured by net revenue from innovations, has increased over one and a half times since mid-2023. The job of our R&D function is to deliver a larger, stronger pipeline with higher quality concepts and improved confidence. To summarize, it's about speed, it's about scale, and it's about superiority. Today, you're going to hear some specific examples about how we bring this to life. To frame the impact of innovation, here I just wanted to show a handful of recent launches. Mucinex 12-Hour Cold and Flu is a great example, an innovation solving a problem no one has ever been able to do before. Our innovation on Dettol Activ Botany in China helped unlock a major consumer barrier through modernizing the most iconic format in our portfolio. There's a story of scale here too. In Q1 2026, we launched 29 new innovations. That is around a 20% uplift on the same period last year. At the heart of every one of our innovations is a cross-functional team of marketing, R&D, manufacturing, and sales making it happen. For the R&D team, building these products requires strong consumer insight translated to critical product attributes. This is where things get really exciting and what we actually do that leverages both physical and digital solutions. Now we will move to the what and shortly welcome Bastien. When we talk about digital science, it is multiple solutions and technologies. It is the systematic integration of digital, data, and AI capabilities into our R&D function from end to end. At its core, digital science combines two powerful capabilities. First, predictive science and simulation. This allows us to model the physical world from molecules to manufacturing using our proprietary data and deep scientific expertise. It replaces large parts of benchtop experimentation with virtual experimentation. Second, generative AI sharpens decisions, accelerates workflows, and removes bottlenecks that slow innovation down. Put simply, digital science allows our scientists to test more, learn faster, and decide with greater confidence before we commit significant time, money, and resources. How does this really work in the business? Everything we do is centered on the product life cycle, starting with ideate and moving clockwise to develop, to launch, to activate, and optimize. The ideate stage is the initial step, and it ties to a critical element of our playbook: consumer obsession. Here we move from insights to initial concept. At the front end, it acts as an insights generator. We use AI and predictive modeling to extract market signals, consumer language, and unmet needs, then rapidly translate those insights into testable innovation concepts. We apply concept testing and consumer liking prediction, allowing teams to prioritize the ideas most likely to succeed before physical testing even begins. The result? Faster decision-making, fewer dead ends, and stronger innovation funnels with thousands of data points narrowed down to concepts that will change the game for our consumers in the real world. At the develop stage, we build our products from an idea to something in your hands. As products move into development, our digital science capabilities accelerate and de-risk formulation. We virtually screen thousands of formulations, significantly reducing physical testing and speeding up the lab work. AI-enabled stability testing, protocol writing, and literature search shortens development cycles while increasing scientific confidence. As products scale, our focus then shifts to execution for the launch phase. We've automated forecasting, regulatory drafting, and document localization, helping teams launch faster and more consistently across markets. This is where it's so exciting that digital science moves from being an efficiency tool to a real growth enabler. Finally, activate and optimize. Here, we continuously track real-world performance. We strengthen claims over time using live data, monitor supply excellence, and include consumer feedback into the innovation engine, creating a digital active learning loop. What we've built is a connected digital science ecosystem. It reduces cost, compresses timelines, improves success rates, and scales globally, all while fueling growth. I'm incredibly proud of all the leaders and teams that have helped make this happen. I'm going to now hand you to Bastien, one of those leaders, to talk through some examples from the stage 1 of this cycle, ideate. Bastien, over to you. Thanks, Angela. For 20 years, I have been driving digital and data transformation across multiple industries. Over the past eight years across CPG and automotive, facilitating meaningful business impact through AI, machine learning, and lately generative AI. This has been the most exciting time of my career. It's a privilege to lead AI advocates since a few years, and being able to take our organization through such learning at scale. The first stage is where our consumer obsession really comes to life. As you may remember from our first Focus On event last year, Ryan shared an example of how innovation concept work that previously took weeks to month in order to generate now can be completed in hours to days. While speed is an advantage, the biggest impact coming through is the quality of the output. This is because we are able to tap into consumer data sets with tens of thousands of data points that the human brain simply isn't able to do internally. Crucially, we're connecting our own data to external unstructured data like ratings and reviews. This opens up new ways of thinking about the consumer and finding solutions for problems in their everyday life. Today, we're going to shift gears and show you how this works in the innovation development space with R&D. Our first example is Vanish. Vanish is already a category leader in stain removal, but as you'll see in the next video, leadership isn't just about performance. It's about how keeping pace with how consumers live, how they make buying decisions, and their expectations on value, sustainability, and experience. The challenge was simple to describe but complex to solve. Consumers want to save time. 33% of all wash cycles now are under 1 hour. They want to launder more sustainably, which means washing at a lower heat. Therefore, the team had to figure out how to create a powder that performs exceptionally in fast wash cycles, works at lower temperatures, and had to stay within tight costs and ingredient constraints. Let's hear from Keyla, our SVP in R&D, germ protection, and household care. Vanish already leads in stain removal. Our challenge? Go further and delight our consumers through creating a powder that excels in fast wash cycles, has outstanding performance at 40 degrees, all within cost and ingredient constraints. To achieve this, we transform how we innovate. By combining computational modeling, superior science, and real consumer insights, we create a new approach, digital active learning. We start by understanding the role of every ingredient, from enzymes to oxygen bleach. This builds a robust data foundation to power our models. Next, our proprietary algorithm help us identify the most meaningful test conditions from temperature to a stain type. This results in a smarter, faster, more predictive experimentation. We then bring the real world in. By selecting household based on actual washing behaviors, we ensure our consumer's insights reflect how people truly use our products in their homes. Consumer tell us in our research that performance alone isn't enough. The full sensory experience is equally important. That's why we then optimize the full sensory experience, such as texture, color, fragrance. These are the cues that shape perception and help drive performance. These steps form a continuous loop to design, test, model, and refine. With every cycle, we get closer and closer to the optimal formula. The outcome? The outcome is a powerful decision tool. The formula optimizer ranks the best performing options, balancing performance, cost, and ingredients, turning complexity into clear data-driven choices. This approach directly enabled the launch of Vanish Turbo. We lower the error rate when predicted consumer perception by 75%, an increased consumer overall liking from 8.4 to 9.1. This deliver very strong in-market results, including around 6% growth within an activated retailer. Our digital active learning approach travels across categories and has become the fabric of how we innovate. It doesn't just protect our leadership. It transforms R&D investment into measurable consumer-proven advantage. We've heard from Keyla. Our digital active learning model for Vanish has helped better predict consumer perception with a 75 lower error rate compared with previous approaches. Overall liking scores have increased from 8.4 to 9.1, a meaningful uplift at the top end of the scale, showing how Vanish Turbo is clearly preferred by consumers over an already well-performing benchmark. We have seen a great impact in market too, with around 6% net revenue growth with a core U.K. retailer carrying the product since activation. There's a lot more that digital science is enabling in the overall process. You can discover more examples in the Reckitt Virtual Campus room. I encourage you to have a look there later. Back to Angela to continue the product lifecycle journey. Thanks so much, Bastien. Now we're gonna talk about our next step on the product lifecycle, develop. This is how we're using digital science to help create entirely new products and categories. A recent example of that new category creation is Lysol Air Sanitizer. It has now entered its third year of sales, and we are seeing ongoing consumer uptake across major U.S. retailers, a fantastic example of a brand-new product in a brand-new category. In a moment, I'm going to hand you to Jan, our SVP for Regulatory Affairs and Safety, whose team combined regulatory rigor with digital science. Together, they partnered with experts in aerobiology and worked closely with the U.S. EPA to develop a robust testing approach that could measure how effectively Lysol Air Sanitizer reduces viruses and bacteria in the air. Let's play the video. We've talked about how we're using digital science to bring new solutions to market faster, but with the next example on Lysol Air Sanitizer, the challenge was not just creating a new product, it was to create an entirely new category. We partnered with experts in aerobiology and worked closely with the U.S. EPA to develop a robust testing approach that could measure how effectively Lysol Air Sanitizer reduced viruses and bacteria in the air. This collaboration with the EPA helped us establish a strong, credible path towards regulatory approval. The impact went beyond speed. It built trust with the external ecosystem and helped us break new ground using digital science to create an entirely new product category. This allowed us to simulate airflow, bacterial behavior, and product performance in a controlled digital environment, saving significant time and cost by avoiding upfront physical builds. This example highlights how Reckitt leverages computational design and digital science to create evidence-led innovation that drives growth and trust more efficiently and cost effectively. These models not only allowed us to launch the first generation of Lysol Air Sanitizer, but also serve as the foundation for ongoing developments of second and third generation products, delivering an even stronger pipeline. Lysol Air Sanitizer has been a really strong contributor to Lysol's overall growth in North America. In 2025, Lysol Air Sanitizer grew net revenue around 20%. It's now contributing a mid-single digit percentage of net revenue to the entire Lysol portfolio. Additionally, the premium nature of this brand-new category is delivering incremental bottom line benefits too, with a gross margin premium of about 100 basis points. Thanks to Jan and the wider team for your amazing contributions. Let's turn to the launch phase, where we scale up from lab to factory and across markets. In our industry, products or innovations don't usually fail in the lab. They fail when we try to scale them. Scaling from lab to manufacturing is a major challenge. What works at small scale isn't always predictable at large scale. In this stage, it is a time where data acquisition strategies and the opportunity that comes with it compounds. With every experiment we run, the data sets get more robust, and with more data, we can then train and strengthen the models. I'm going to turn it over to Mark, SVP R&D, Self Care and Intimate Wellness. He will share an example from Gaviscon and how we're using the fascinating capability of digital twins. Let's hear from Mark.. Moving from laboratory to industrial manufacturing is 1 of the hardest challenges we face because the science that works in a small scale experiment doesn't translate in a simple, predictable way at scale, and there are so many variables at play. The business impact of getting this right is huge, and this matters because when scale-up goes wrong or doesn't work out the way we intend, the impact is immediate and material. We see delays, rework, cost increases, and in the worst cases, our supply can become unstable. For a business, this means pressure on margins, it means missed growth opportunities, and increased operational risk. The question becomes, how do we change that? To counteract some of the challenges we face when we scale our innovations, within the Gaviscon team, we've started by reframing the problem. Instead of reacting to issues during scale-up, we've asked ourselves the question, "What if we could predict and solve them before they happen? What if we could test manufacturing at scale without physically going to full scale?" This is where simulation and modeling comes in. We build what are called digital twins, which are virtual replicas of our products, equipment, and our manufacturing processes. The models we build are calibrated using real world experimental data and continuously refined as we generate new insights. Over time, they become accurate representations of reality. This gives us a powerful capability and ensures we can simulate how we scale up before we actually do it. This means we can test different formulations, change the order of ingredients, adjust conditions and processes, and understand all of this before it affects production, and that we commit then to capital, time, and resources. In effect, we're combining virtual and real world data to stress test our decisions. The outcome is not only excellent execution, but predictable execution. Using digital twins ensures we see less delays, less rework, shorter cycle times, and a much smoother transition from small scale development into larger scale manufacturing across the product life cycle. For us, this is not an experiment or a pilot. It is a repeatable capability that we deploy at scale, ultimately, that is what this is about. It is accelerating how we innovate while protecting quality, cost, and supply, while delivering a consistent performance, stronger margins, and sustainable growth. As Mark's example shows, digital twins create real value, especially at scale-up, where failure is most costly and hardest to recover from. Across our sites, we run hundreds of these trials every year. Being able to run trials in a virtual world reduces the number of trials we have to run in the physical world. Take Gaviscon in Hull. A small batch costs around GBP 25,000 and 30 person hours. At factory scale, a large batch costs over GBP 100,000 and requires more than 300 person hours. Another example for you, which isn't on the slide, comes from Thailand. That is Durex in Bangpakong. Trials there typically cost GBP 100,000-GBP 300,000 and can take over 100 person hours. For both examples, trials can now be simulated in under two hours, saving a significant materials, energy, time, and effort. We're now able to make earlier, more confident decisions using better evidence, and that reduces risk and rework right across the life cycle. The final stage of the product life cycle is activate and optimize. Here we have two examples for you. The first example demonstrates how we use 3D modeling to make Dettol packaging more sustainable and consumer preferred, both of which are often very tough technical challenges. With digital science, we're able to optimize packaging for bottle strength, stiffness, weight, and overall performance. At the same time, we deliver on consumer appeal, including appearance and brand identity. Optimization means more than design tweaks. It involves changing geometry, introducing more sustainable materials, and ultimately using less plastic. To achieve this, we virtually test every aspect, including how the bottle is manufactured, how it's handled, and how it's transported. Let's play this next video. Introducing the new antiseptic liquid bottle from Dettol. This bottle uses PET, which has greater recyclability attributes than the previous bottle and is our first to market pack, which overcame this challenge through in-house 3D design and simulation capability. Through 3D design and simulation, switching from polypropylene plastic to PET, and by optimizing the shape, we can deliver up to 400 tons of plastic savings. This boosts pallet efficiency by up to 33%, meaning more packs per shipment, less trucks on the road, and reduced CO2. It is creating a step change in performance on shelf too, with improved purchase preference and a higher perception of quality. The simple and sleek bottle design also makes it easier to find on the shelf. This was all achieved by the use of PET material that provides a premium and eye-catching appearance, better size impression, a bottle shape that amplifies Dettol protective arch, and a bigger label. It is cheaper to produce per annum, achieved through a center parting line that provides higher mold cavitation, meaning more bottles are produced per cycle and a reduction in energy. Using simulation technology means we can predict how the bottle will perform before actually making it, which is how we managed to optimize the overall design and shape to maximize its strength and deliver key performance improvements. Through using cutting-edge simulation technology, we are not only delighting our consumers, we are building a more sustainable future. This is what superiority looks like, higher confidence in outcomes and fewer late-stage failures. It's amazing how saving 2 to 3g of plastic per bottle adds up to 400 tons when millions of bottles are sold each year. Those small reductions translate into a significant sustainability impact, and we see real efficiency gains in distribution through pallet efficiency. This example is also great in highlighting what digital science is doing for our consumers and our P&L. The newly designed bottle has driven a major uplift in consumer purchase preference, and as you heard in the video, is more cost-effective, saving around GBP 5 million per year. Angela, you know what I love about that example? That we use simulation technology to predict how the bottle will perform before actually making it, enhancing our speed and our superiority. Now, let's talk about that second example, one of our flagship GenAI solutions, WriteIt. Innovation depends on large sets of documentation such as reports, protocols, literature reviews, scientific evidence, clinical trials, and regulatory records. At Reckitt, this work is essential to ensure we deliver quality and efficacy for consumers. The problem historically is that it has consumed 30%-40% of R&D time, creating a major bottleneck between insight and market launch. WriteIt is Reckitt's GxP compliant AI platform for document drafting, literature search, and protocol generation. GxP here means compliance to good practice, critical in the highly regulated environment which we work in. Validating our system as GxP compliant ensures further robustness, repeatability, and continuous control with strong data integrity and high product quality at the center. It has been built around R&D workflows and trade on Reckitt scientific, technical, and regulatory knowledge. It helps teams grow faster, more consistently, and with full traceability. What makes WriteIt distinctive is that this is not your generic AI tool layered onto the business. It was co-developed with the respective teams in R&D and designed around process reinvention of real workflows. Much of the discussion around AI focuses on the technology and performance of large language models. Our effort has been on putting AI to work by codifying the tacit knowledge of our teams into AI applications. By tacit knowledge, we mean our team's experience-based know-how that is rarely written down. We also embed market, category, and Reckitt context into how AI is used. Applying this approach at scale and consistently across all teams is where we believe AI makes a real difference in driving our winning playbook. By embedding adoption, governance, and process redesign from day one, we built a solution that is practical, trusted, and scaled seamlessly. Let's take industrial trials. Every time we scale a product from the lab to the factory, teams must prepare, review, and approve detailed industrial trial reports across category teams, product development teams, quality, and supply. We run more than 250 of these trials each year across brands, including Lysol and Finish. With WriteIt, drafting and reviewing become significantly faster, up to 8 times faster, in fact. This means that our teams are spending far fewer hours creating documents and much more time on advancing innovation, such as the creation of tailor-made formulations. The impact is best described by the scientists using WriteIt every day. Let's hear from an R&D senior associate who is using WriteIt on the ground. I look after a team of scientists, and I find that they spend a frustrating amount of time transcribing data from one system into protocols or reports. With WriteIt, what I'm now finding is that that initial first draft stage is taking minutes rather than days. It's allowing my scientists to then spend their time building the real quality and technical information into these reports and protocols without losing any of that quality that we need for our medicinal product submissions. It's been a real game changer for us. WriteIt is already scaling across Reckitt. By February 2026, we had trained over 650 users across R&D, with further solution being deployed in the remainder of the year to extend that training to over 2,000 users, roughly half of our R&D community. This is not a pilot, it's a real capability adopted at scale. This allows us to innovate not just faster, but smarter with full control. It's one example of how we're weaving AI and GenAI into the fabric of our business. This is a great moment to now turn it over to Nigel, who will share with you how we're investing in capability and embedding this further across the enterprise. Thanks, Bastien. Good afternoon. I'm Nigel Richardson, and I lead Reckitt's IT and digital global team. I started my career as a software engineer in the automotive industry, and I've since spent more than 20 years leading IT and digital teams at two major FMCG companies before joining Reckitt. It's an incredibly exciting time to work in tech, and I'm passionate about how data and AI improves people's lives and delivers tremendous value. You've heard from Shannon, Angela, and Bastien about the impact digital science and AI are already having across Reckitt, from faster innovation to better decisions to improved operational efficiency. Now, what I want to focus on now is how we've built the capability to scale that advantage and why our early platform investments matter so much to our AI journey today. The real value of AI does not come from isolated pilots or impressive demos. It comes from having the right foundations, the data, technology, people capabilities, and the operation model that allow AI to scale reliably across the enterprise. That has been our focus for the last five to six years. We've been very deliberate in how we've approached this. Before AI can truly transform a business, you need to put the right digital platforms in place. Starting in 2020, we invested in moving to the cloud and systematically building our data foundation. Our initial focus was on enabling digital science in R&D, as you've heard from Angela, and we then expanded this capability across other functions. Because those platforms were already in place, we were well-positioned to move quickly when generative AI emerged. In 2024, we began our GenAI transformation, applying the technology to real business problems, first in marketing, then R&D, and now into supply. Alongside this, we've continued to scale and evolve our data and AI platform as the technology accelerates. Throughout this journey, our approach has been clear and consistent. Firstly, business-led decisions. Every AI initiative is tied to a clear business problem or opportunity with senior sponsorship from the outset. Secondly, disciplined investment. We apply rigorous value tracking from pilot to rollout with benefits reviewed quarterly by the executive team. Then capability building. We don't just deploy tools, we redesign workflows and upskill teams for lasting change. To date, we've trained 2,000 people across marketing and R&D, we're now extending this into supply because AI only delivers value when it's embedded in how people work every day. Infrastructure that scales, we're unlocking further value by continuously investing in our data and technology. This includes enabling agentic AI to take on more sophisticated tasks. For example, within the WriteIt solution that you just saw on the video, we orchestrate several agents, including a drafting agent, an analysis agent, and a root cause analysis agent. This combination, strong digital foundations, disciplined execution, and a constant focus on scale, is what has allowed us to accelerate our AI journey and turn capability into competitive advantage. At Reckitt, we think about AI advantage through three lenses. Knowledge. Power comes from AI trained on what we know. As Bastien explained, our solutions are grounded in tacit knowledge. That's decades of proprietary skills embedded into our solutions. That gives us uniqueness, precision, which generic solutions cannot easily replicate. Technology foundation. Trinity is Reckitt's data and AI platform. Named after the library in Dublin, it contains Reckitt's most critical knowledge and powers our scalable AI platform. Compliance is also built in by design. As Bastien mentioned, this ensures robustness, repeatability, and control so that we always meet the tough regulatory standards for pharmaceutical products. Of course, none of this could happen without our people. The most important thing is human in the lead rather than just in the loop. All our AI initiatives are human-led and AI-powered. Solutions are co-built with business and technical teams. They're grounded in workflows to enable people to get work done faster and better. Change management and adoption is at the core, with accountability to measure and capture value. Our continuous improvement approach creates a flywheel effect from fast feedback loops, so we add more knowledge, we improve solutions to become even more precise and effective over time. When these three principles come together, AI stops being a technology experiment and becomes a sustainable advantaged business capability. As a reminder, our strategy is built around three pillars to drive speed, scale, and superiority. For us, this means three things. Firstly, empower, putting AI into the hands of all our employees to make everyday work faster and smarter, making it easier to get things done. Secondly, transform, driving productivity by using AI to rethink how work gets done in key functions, redesigning workflows with GenAI and now agentic AI. Thirdly, invent, creating entirely new capabilities and opportunities enabled by data and AI, such as our digital go-to-market in China. These pillars work because they sit on a strong enterprise backbone, and it's such an exciting time to be in this job. We're reshaping our business to drive value enabled by our data foundation, our AI platform, and also our government frameworks, operating model, and upskilling our people. We've made real progress executing our strategy, but a real key point here is that we've learned a lot along the way. One important proof point of discipline is that not everything we test is scaled. We've deliberately stopped initiatives where the return wasn't good enough. For example, we stopped a media analysis solution before scaling when we found that the value wasn't sufficient, and these learnings are built into how we invest going forward. We learned that strong data foundations are a prerequisite, not an afterthought, and we learned that adoption matters as much as technology. Co-building with experts and embedding change is critical. Those early lessons now shape how we can scale with confidence. In a nutshell, AI literacy is growing across the whole business. Our people are suggesting new ideas, building improved solutions, and driving the business forward. What has been key to our success? In short, four foundations have been key to this. Firstly, data curated for reusability. Reinventing workflows with AI is only possible when there is a strong focus on data quality and reusability. Secondly, a platform built for optionality. By using open standards where possible, we continue to exploit new technologies such as agentic AI. Thirdly, responsible AI by design. A robust responsible AI framework and training are critical to mitigate AI risks. Finally, a scalable operating model. Our cross-functional teams focused on processes and business outcomes are delivering our AI solutions. This is how we at Reckitt are building a scalable AI advantage to drive growth and productivity, and they are critical enablers as we expand our scope from marketing and R&D into other functions such as procurement, supply, and finance. With that, I'll hand over to Angela to wrap things up. All right. Thanks, Nigel. Let me close by first reminding you of the speed, scale, and superiority that digital science brings us. We talked through examples and showed you evidence of how we are enhancing each of these across every stage of the product life cycle. Looking at speed, we talked about virtual experiments and how we've moved from doing just 100 of these three years ago to more than 10,000 today. This enables quicker design choices and more expansive design spaces. We talked about our WriteIt users generating reports 8x faster, freeing up scientists to do what they do best. Now, scale. I think the real insight for you here is that we are reinventing the way we do business with touchpoints across all functions. You have heard how over 650 colleagues have been trained on WriteIt and how it's transforming their working lives. We talked about how our average project size in the innovation pipeline is now over 25% larger year on year, driving real value creation. Our focus on superiority has delivered an innovation pipeline that is now at 1.5x the size it was compared to three years ago. We talked about the enhanced consumer preference for Vanish Turbo and Dettol Antiseptic Liquid, leveraging digital active learning. Overall, this progress is driven by sustained investment in the capabilities and the infrastructure that will power Reckitt's future. I have some news today too. I am so excited to share with you that in July, we will open our new global science and innovation center in Shanghai, our most advanced hub for innovation and digital science. The center will integrate cutting-edge capabilities from live streaming, consumer co-creation, and sensory testing to advanced material science. Key areas will include claims development, microbiology, predictive simulation, and scale-up capabilities, enabling faster development, stronger product claims, and more scalable innovation across our brands. Finally, I would like to bring it back to growth. Digital science is now a critical part of the toolbox in how we innovate. It has augmented how we build consumer insights into actionable concepts to how we develop and scale new products. The organizational and cultural transformation has enabled rapid deployment, and the adoption is creating new ideas every day. In every aspect of the technical work we do, the teams have been upskilled and are building the workflows and data acquisition strategies to unlock even more value. While many of today's stories focus on innovation, the real shift is broader. We are changing how Reckitt plans, prioritizes, and executes, whether we are launching something new or running what already exists. We have the data, the capability, and the world-class portfolio ensuring our journey is built for scale. We are delivering tangible results for today and building real momentum for the future. Thank you for being here with us today, and we look forward to your questions. Let me hand you back over to Jon. As Angela said, we're shortly going to take your questions. Before that, we're going to take a quick break to allow you to submit your questions through the Q&A tab on the webcast platform. For our covering analysts, if you would like to ask a question live, then please log into the Zoom link shared with you, and we'll bring you on screen when it's your turn. We'll be back in just two minutes. Great. Welcome back. Thank you all very much for joining us. We're starting to get some questions come through on the portal. I'm just going to do a first one here, and then I'll come to the Zoom call. When you're ready to ask your question on Zoom, put your hand up, and I will come to you. The first one we had come in on the portal, is asking if you can tell us about the experience of driving adoption of AI tools across your functions, and has this been met with any resistance? Bastien, maybe if you want to have a first answer on that. Sure. I think it's fair to say that everyone is being exposed to AI in their personal lives. When we started deploying solutions in marketing, working with R&D, we saw a mix of excitement and perhaps skepticism. I think what has helped us to get to what we were excited to talk about today was the fact that we co-designed the solutions with the teams, that it's solutions using AI that is embedded into their day-to-day. I think this has gone a long way for us in making sure this is part of our functions. I would say two last things, and maybe Angela, you might want to share the R&D experience, which is, one, we've mentioned this is those are solutions to have human in the lead, which is really important for us as a principle. The last thing I would say is that this is a learning curve as the technology evolves, and therefore we are trying to stay handle into what new technology and what new solutions can help us as we go forward. Maybe you might want to share more on R&D. Yeah. Thanks, Bastien. I guess from an R&D lens, you know, as I mentioned today, we started with traditional AI sort of simulation and modeling a handful of years ago. That has been, you know, a really important point in terms of some of the capabilities that we talked about today. It was about 12, 14 months or so ago that the GenAI explosion happened, right? It was sort of everywhere and, I think from my experience in R&D, I had sort of two different views. I had one view that was very much, "Let's go. It's new. Let's get on top of the latest, you know, new technology enablement." I also had another part of my organization that kind of was, "Hmm, this You know, no black boxes. We're fundamental first principles kind of scientists, let me just wait and see. It was just kind of an interesting point to be sitting there with the organization thinking, "Well, this is interesting," right? We've got sort of two different views that we need to have, how were we gonna sort of tackle it was one of just education. We, I would say as a leadership team, we decided that we needed to become students, right? We couldn't just sort of advocate and push and top-down these things. It was more, it was more important to immerse ourselves and become students in our own right? To, you know, I think as Bastien said, it's changing so fast. How could you possibly know? Once we did that, and we created sort of the ecosystem of this is a learning opportunity just like anything else, we then got into the co-development, and it became theirs. They got to put their fingerprints and their thumbprints on these solutions. Fast-forward to today, and I'm really pleased that it's now become something that we go to in terms of a toolbox of just another way to attack the problems that we have to go after. Brilliant. Okay. Looking over at Zoom, Warren, if we can bring you on first to, if you've got one or two questions. Yeah. Thank you, Jon. Can you hear me? Yeah. Okay. Thank you. Thank you for the session. It's Warren here at Barclays. The thing I was most interested in was the pipeline size. I think you said that the pipeline size is, I think, 25% bigger already in 2026, and the Power Brand pipeline I think you said was 1.5x bigger since 2023. That's quite material. Can you say maybe what that's done for innovation weights as a percentage of sales? You know, it sounds big, but what's it actually doing in terms of the sort of the revenue? How's that changing? Related to that, with the new tools that you have, how much visibility do you have on the pipeline when you look beyond 2026 when you're thinking about digital and AI? Is it allowing you to extend the duration of the pipeline beyond the kind of near term into the medium term? Thank you. That's great. I think given the innovation angle to the question, Angela, perhaps if you want to answer that one first. Thank you so much for the question. Let me start with the second 1 first with respect to tools and what we can see into the future. We actually, as part of our digital roadmap, have incorporated digital tools in terms of our overall project and portfolio management, so we've professionalized that aspect too in terms of every project, every resource is now, you know, well-articulated in terms of the digital tools that we can see the overall pipelines. We look out actually, in many cases three to five years into the future in terms of the innovation pipeline. It can depend on categories in terms of a sort of overall product life cycle, but it's actually many years into the future that we think about our technology enablement and then making sure that we have product roadmaps alongside that. Getting to the sort of valuation part of it, maybe just to say the 25% is relative to our projects, so this point around we're wanting to really ensure that we're driving good ROI-based methodology, that we're scaling the products that we actually go after so that we can go into multiple markets, multiple retailers, those kinds of things. This idea of project size growing is actually allowing us to get more from the investment that we make on the development itself. It's a critical point in terms of that overall ROI mindset. On the pipeline, it is significant in terms of the enhancements and the increase that we've been building over the last couple of years. What I can say to that is we know that innovation is a very large proportion of our ongoing annual growth, we have targets with respect to different categories and areas that then define a sufficiency approach for each of our different power brands. That's how we think about the sort of size and scale of the pipeline, as it increases into the future. Great. I think next if we go to Sarah Simon at Morgan Stanley. Sarah? Hello? Yep, go ahead, Sarah. Great. No, I just had one question, which is kind of to take the other side of the coin. Is there any downside to doing things this way? I'm just thinking obviously competitors also have AI tools. Do you find that you get a more generic nature of innovation? Just wondering if there is any downside to doing this. That's a good question. I wonder, Bastien, maybe if you sort of think about how our AI tools that we're using in the business are differentiated from what others are doing perhaps, and then we can bring it to Angela to think about the definition between our competitors. Very briefly, I think we've said that during the presentation today, which is for us what is really important is tacit knowledge and context. We have not been seeking to train our own models. We've been seeking at really understanding how our people are doing things and embedding that into AI application. The context, which is market category Reckitt, is it has taken a lot of iterations to actually make sure that our applications are delivering an output in our different categories, right? Whether it's for Durex or for Lysol or for Finish, we believe that that is what is making it unique to Reckitt. Maybe I'll let Angela and Nigel add on that. Yeah, I mean, I think the specificity of the consumer data that it's they're our consumers, right? I mean, they're, I'm sure they're consumers of other places, but they're our consumers talking to us about our products and competencies and capabilities that we think we have a right to win in. It's our products as we think about all of the sort of technical, you know, product attributes and how we think about drivers of liking, how we think about product design. They're our products. I think the third piece of it is from a brand perspective, the right to win is typically what the brand equity can carry. I think sort of putting all of that together creates its own sort of world in terms of places where we innovate from. While I agree that there are multiple, you know, tools and systems out there, I think this point around the specificity of what we're aiming to do is because it's what we're good at. It is the chosen categories, where we think we have a right to win, and then as Bastien was saying, sort of decades of consumer and product data that we can then put together to drive this multi-generational approach is the way I would think about it from a more product design perspective. Great. Why don't we go to David at Jefferies, and then after David we're gonna go to Jeremy, just to queue you up. David, you're muted our end. Hello, can you hear me now? There we go. There we go. Don't know what that was, sorry. Two from me if I can. Firstly just in terms of the bottleneck in the system all the way through to the customer, is that shifting? I mean, what I kind of envision, envisioning is that you're kind of coming through with a lot more innovation and ideas. Is the bottleneck now moved from your ideas coming through to the sales and marketing teams not being able to deal with all that in terms of the resource allocation? You just can't deal with all the good ideas that come through and how is that then being addressed? Is that the right perception? Then the second one is in terms of new technology. How do you have a or is there a process in place to keep up to date with the new systems, the new AI products, et cetera, that are coming available? How do you decide when to invest and commit and sink cost versus delaying, waiting for the next thing to come out and make sure you don't, you know, get behind or invest in the wrong thing? Thank you. Brilliant. Thanks, David. I think of the two there, maybe if we go to Angela Naef on the first one, and then Nigel probably to pick up the second one on new evolving technologies. Yeah, on the bottleneck. Yeah on the system. It so matters the timing of our launch with respect to the areas and the categories that we play on. We can't be pushing more into the market than our consumers can take and than our retailers can take and that we can support. From a launch cadence perspective, we actually, we actually stitch together the readiness of the different technology launches alongside the market readiness in terms of shelf resets, where we need to be with respect to new products and so forth. You know, China is obviously one where the speed is so fast that it's created a terrific learning loop for us where we can actually launch at a much higher cadence as part of our way to in which we can optimize. That's kind of a special way I think right now that we're able to take advantage of. It's absolutely the case that bottlenecks are changing, right? You know, I always think about you solve one rate limiting step, then all you do is you create the next rate limiting step. If you're really good, then every single time you're optimizing and defining, optimizing and defining each rate limiting step. I'm not sure we have found all of them yet, but I can say we're definitely increasing the confidence as well in terms of when we get there, the readiness will be that much greater. Nigel? Yes, and then to the second question, so I mean, technology, as you know, it's moving so quickly in this area, so one of the key design decisions we made was to build our data and AI platform as much as possible on open standards. It means that as technology does move on, you know, it's relatively easy to then, you know, move on to other parts of technology, change the models, for example, the LLMs as they develop. That's one point. We've aimed to be as open as possible to avoid vendor lock-in. That's one point. Secondly, in terms of how we think about where we develop our solutions versus looking at something that might be more generic, we have a process where we look at all of our different business capabilities and say, "Look, where are they ones where we wanna do it, you know, in a fairly standard way? We wanna be best in cost. Where are there areas where we wanna be differentiated?" Particularly it tends to be areas like we've talked about where we've got real deep, tacit knowledge, and that's where we focus on actually building bespoke solutions where we really think there's advantage for us. That's how we've been thinking about it. We just keep scanning the environment to see as the new AI develops. As I say, we've built agentic AI into it, into the most recent solutions, and we'll continue to do that as the technology advances. Great. Thank you, Nigel. If we go to Jeremy, HSBC, we'll go to Ed at Redburn after that. Jeremy, yeah. Okay, yeah. Hi there. Thanks for the session and taking the questions. Just 1 from me. You went through all the different stages of the innovation process. You gave examples of how, in each of those digital is allowing you to get better outcomes. I guess maybe similar to one of the previous questions, how do you benchmark the sort of process that you're getting relative to the relative to the competition? Because I know that all of your peers as I say have these similar sorts of tools. They'll all be getting, you know, faster ideas. They'll be able to do their digital twins as well. What is it that you can give us to say, "Well, you know, we know that we're getting a little bit more than the competition. There are certain areas where we think this is genuinely giving us an edge over the other big FMCG players that will have similar sort of capabilities"? Okay. That's a interesting question, Jeremy. I wonder, maybe Angela for some initial thoughts, and then perhaps Nigel on the, on the benchmarking side across other things if we, if you want to do that. Yeah, sure. I think the way I think about it is the benefit that we've been able to bring across the entirety of the product life cycle is sort of one level of competitive advantage, that it's not a pilot or a test or in one place we can use it. Being able to look across, I mean, we gave you two examples today, but we have many more examples across our power brands around the product life cycle. I think it's this holistic utility of digital science in all aspects of the product life cycle I think is pretty special. I feel pretty good about the fact that at each stage in the game, the KPIs that we look at are improving basis the targets and ambition that we set for ourselves. That is true. How many ideas are going through, how many launches we have, the size of the projects, the size of the pipeline, as we have talked about. As we look at it holistically, the whole thing is moving with more confidence, more credibility, so that when we get there, we know that it is going to land. To me, that is the big takeaway and big benefit that we are seeing. Just to add, more generally, certainly I spend a lot of time talking to other Chief Information Officers and Chief Digital Officers. I think one area that I think we're doing well in particular is really going deep into particular functions and processes and really driving the change in the workflows and then building AI solutions that actually fundamentally change how people work rather than just putting solutions in. Not a comment specific for R&D, it's more as we've really are deploying AI across the enterprise. I think that is an area where I think we're doing well, and we're obviously continuing to want to learn, but I think that's an area that we do have some advantage in our approach. Excellent. Can I go to Ed Lewis now at Rothschild Redburn? Yes. Thanks very much, guys. Good session. As I echo what others have said. Just a couple ones from me. I guess, you know, you're a global organization. You've clearly got a lot of presence outside sort of developed world and into emerging markets. I'd just be interested to hear sort of how sort of deployment and how sort of the benefits you're getting from what you've been discussing is, whether it's agnostic about sort of which regions of the world you're in, or whether it's particularly enhanced by other parts of the world. I mean, obviously the center in China looks interesting in terms of what that can bring. I guess a, sort of a, just thinking about the implications on the sort of the governance and regulatory side, the idea of having agentic bots sort of across the organization, the risk around that. I'm sort of unclear exactly, I'm not how that plays out, but I'd be interested to hear, 'cause, you know, it could create a, obviously a risk where if a agentic bot gets out of control or whatever it is. Just to sort of understand how you deal with that potential sort of threat. Thanks, Ed. I think probably on the first one, in terms of deployment, around our markets, globally, why don't Bastien potentially lead on that one or Angela, if you've got some comments there. Then the second one I think probably sits with Nigel in terms of governance. Sure. Happy to start. It's everywhere. I think that's the easy answer, right? The way that I run the global R&D organization is we are set up by categories, as we've talked about, I think, over the course of a couple Focus On sessions. We have a category area based approach to the organization. More central to the R&D side is our competencies and capabilities, and they show up primarily in our centers of excellence where we do the development work. As we launched WriteIt, that Bastien spoke about, we went through, you know, several different tranches of trainings, and we did it in every one of these places in terms of whether it was Thailand or Germany or the U.K. or the U.S. We actually look quite broadly that the capability is applicable everywhere. There isn't a filter that says that there's one place different than the other. We go after value, we go after business cases, and then that's what sets the priorities in terms of the categories and areas that we go after. Yeah. In terms of the governance, I would say two parts to that. One is what I mentioned about the responsible AI framework. It's something that we really put in place right at the beginning and co-developed by legal team and ourselves in IT&D. That really assesses any new idea where we're gonna use AI, and it makes sure that, you know, it is in line with all of the regulatory standards. I mean, the toughest one is typically the Artificial Intelligence Act, and that's, you know, a very good standard. That's the main one, but there are others. That ensures that we're clear on any level of risk and we've got the right controls in place for that. The second point I would say is that, you know, this point about human in the lead, you know, we are very, very conscious to ensure that with all our solutions, we actually have, you know, a person who's accountable ultimately for the quality of the work, rather than having things completely autonomous. I'd say those two things. Can I add one, just one comment? I think, yeah, I heard governance and regulatory as well. I wanted to go back to the session today where those solutions in R&D have been made GxP compliant, and this is replicating the process that we have today as we're in R&D, including connecting to data platforms that were already vetted for GxP. This is really an important for us that is again encoding to having applications that are here to help people in the functions. Fantastic. Okay. I think we've got two more to come on Zoom. I'll go to Tom at Deutsche next and then to Diana at Bloomberg. Tom, you're on. Thanks. Good afternoon, everybody. It's just relatively simple. Just to try and understand, what proportion of your R&D costs are on sort of packaging and production related versus new product lines, like I guess sanitizers being a good example? Of the R&D you deploy, what would be the split as well? Do you find that this is helping in more the packaging and production side, or is it something that is helping you identify actual new products more quickly and more efficiently? It wasn't quite clear what you said on, sorry, the cadence of deploying these. Are you saying that the cadence is going up or that you're just more confident that they will land when you do deploy them or are you actually getting both? Thank you. Thank you very much, Tom. I think again, that's probably going to be with Angela to kick off with. Again, around split of R&D costs between packaging and actually, and sort of smaller evolutions versus new category creations, a question of cadence versus confidence or a combination of the two. Yeah. Thanks. From an overall investment portfolio, we drive a very simple differential sort of portfolio investment approach across all of our categories and competencies. We definitely manage all the big sort of blockbuster things, we talk about Mucinex, and sort of small renovations that need to do to maintain the relevance of a brand. Within the full complement of R&D, we actually make prioritizations based off of, you know, return on investment, size of impact, and ultimately our ability to continue to drive market share. I won't go into, you know, sort of different budget details today, but maybe suffice it to say, what we talked about today were also examples that are in the market, and I think that's kind of an important point, right? That there's a lot more that we didn't talk about that maybe, you know, speaking a bit to more of the invention side of, you know, identifying new molecules, identifying entirely new solutions. Those we'll have to save for another day when they're, when they're a bit more ready for us to share them. Yeah, but it is going across the entirety of what we sort of consider more disruptive and transformative programs alongside more of the renovation-type programs. I think the question on deployment, on cadence of deployment, I think, relative to launches and so forth, I think what we shared today was, you know, we've had an uplift. in Q1 2026 relative to Q1 2025. It's just a data point, right? In terms of being able to ask ourselves how we're actually driving this cadence of deployment. Of course, every market, every category has a different sort of cadence of launches and so forth. When we look on it year on year, we're quite confident that we're actually driving more. I think that's sort of the proof point on the Q1 data point that we provided. I also think that this point on confidence is really important because, you know, when you're doing R&D, you don't always know all the answers. That's why we have to go through the process of R&D. Learning, pivoting, you know, relearning and making sure we are applying those learnings is just the way in which we do our business, right? I think some of these tools are just giving us that ability to do that with an additive input of utilizing these digital tools. In the end, of course, we have got scientists and technical professionals that are ultimately making the launch decisions. Great. Yeah. Thank you. Diana, if you wanna go ahead now. Hi. Thank you. I think in a way, going back to Sarah's questions in some way, in terms of the downside. Yeah, go ahead. I'm hearing some echoes. Just in terms of how Reckitt's digital-focused process or this process would assess, for instance, the risk of having falsely rejected ideas. Say a design fails the digital test, but it could potentially work in the real world, and how that balance of assessing the return with the time and cost investment would work. If I could just add a follow-up in terms of the WriteIt tool. I think you mentioned there are 650 people trained already. How, what's the proportion out of that of people that are actually using it regularly? I don't know if daily, weekly, but that could be interesting. Thank you so much. Okay. The first one there, I think that could be across any of our team here to answer. I think, Angela, if you wanna go with it, and then maybe Bastien, if you pick up on WriteIt. Yeah. Okay, this was more about the downside of maybe we might miss something or might get something wrong and maybe something that, you know, we didn't wanna have happen. The way I think about it is all of the tools that we talked about today are additive tools. They're augmenting and enabling the scientist or the product developer or the regulatory professional to have more inputs into the decisions that they make every day. There's no replacement. It's not replacing the good judgment. It's not replacing how we think about product attributes or maybe, you know, scientific principles that we need to have in terms of. a scale-up problem, for example. That's maybe the first way to think about it is it's not a replacement. We are very focused on making sure that all of our technical decisions are grounded in foundational science. We don't use these in a way that we don't understand the outcome, right? Is it, is it adding? Is it, you know, giving us new insights? Is it maybe provoking a question that would help us go down another path? Absolutely. I think that's the, that's sort of the overarching way in which they work. As you think about things like digital twins, these are all around utilizing your scientific and engineering understanding to design a digital twin that with data gets better and better over time. It's this constant loop between physical and virtual, physical and virtual, that's this sort of compounding effect of more data enhances the models and so forth. We now I think have a pretty good feel for what that needs and looks like against our different problems that we solve, because not any one answer exists for any one problem either. Maybe I can share a little bit on your question on the number of users, 650 users of WriteIt. We are going to have usage of WriteIt across half of the R&D community, which means 100% of our people can actually use that, right? The reason why I'm saying that is that WriteIt is being designed for specific purposes. I mentioned industrial trials, protocol generation. As we identify the purpose of WriteIt, we ensure that we can put that in the hands of the users who really need it. With this, it's very difficult to say the number of, I would say, active users on a given month because our people are using it when they need it. When they do, they really see the value of it. You've heard one testimonial, which is, I think, a perfect illustration of what you would hear if you were to talk of a lot more to the teams. Long story short here, it is really about deploying to everyone that has a use for it, and when they do, they use it when they need it. I think it's part of that process. Great. Thank you. We've had some more questions come through on the portal, so thank you very much for submitting those. First one I've had come in is from James Hutchinson, who says, "Great session. Thank you." Thank you all for that. "How are your physical labs evolving alongside the digital science capabilities to complement these new capabilities?" I think you might go straight to China on this one given what we talked about there. I know. We do have plans across the entirety of our lab network. Yeah, I mean, I'll share the example of China, because it's actually a place where we've built in digital science enablement from the ground up, which I think is pretty remarkable. You know, this idea around this adjacency is always how I'm thinking about it. We always have to do the physical experiment, always, right? This digital opportunity of being able to hone the thinking, hone the opportunity spaces. We actually have many of the tools that Bastien talked about, some of these simulation and modeling now live across many of our laboratory facilities. It's not just go to the lab and maybe later do a digital experiment. It's becoming the way in which we're working. Our job is to, of course, make sure that the tools and the, you know, the accessibility of all of it continues to grow with the deployments. Okay. We're just coming up on time, so I've got one more question here, and I didn't think we'd go through an event at this time of year without someone wanting to talk about the season. The question comes from Richard Lee, who asks, "Is AI and digital science used to improve forecasting? If so, what extent has it been improved? Has it been able to forecast the cold and flu season, and therefore how you expect Mucinex 12-hour to perform?" Why don't we just take the last part of that question, I think, to Mucinex, which we're obviously very excited about. Yes. Why don't we go into that? Yeah, happy to. Predicting the season is a tough job. There's no doubt about it, right? It's not because it's tough, it's because it's multifactorial, right? I think, you know, part of what we've been able to interrogate and investigate around all the different drivers of seasonality, incidence rates. I mean, there's an incredible amount of pathogen surveillance that we do, making sure that we understand, you know, communicable diseases, how they're moving across, you know, across the planet, from one hemisphere to another hemisphere. We think a lot about that. We think a lot about symptom relief and what symptoms, you know, patients and consumers are actually having, dependent, again, on what type of incidences may be happening. We do a lot of that analysis, really driving hard where we could aim in terms of where the likely issues on symptom relief should be coming. It's a tough job to do to get it all right. There are things like weather that impacts these things as well. I do think more digital tools and more simulations in that front are gonna continue to hone this and get it more precise as we go forward. I'm quite optimistic that it will continue to help us refine our own plans. Yeah, I think it's helpful, it's additive, but it's multifactorial. Another area we can continue to utilize some of these excellent solutions going forward. Okay. That is all the questions that we have come in. I'd like to thank you all for joining us. All the materials from today will be available on the Reckitt Investor Relations website, as well as our app shortly. If you do have any follow-up questions, please get in touch with Nick or me and the whole of the IR team here. A quick reminder, our next Reckitt Focus On event is on November 19th. We will be diving into our North America business, hosting from our regional HQ in New Jersey. There is sign up available for that now, we will share some further details closer to the time. We look forward to seeing many of you there, and certainly over the next few weeks at upcoming conferences. Thank you.