How AI-enabled execution became Reckitt’s ‘tenfold game changer’

A revenue growth management (RGM) plan can look perfect on paper. But even the strongest commercial strategy only creates value when it lands in store: Is the right product, at the right time, in the right quantity on the shelf when the shopper wants it?

Reckitt understood this challenge well. Through RGM, the company had already used data, AI, and modern technology to make smarter, more granular decisions on pricing, promotions, and assortment—identifying where value could be created long before products shipped or promotions went live.

But planning was only one bookend. The other was in-store sales execution.

For grocery retailers, product availability is central to consumer trust. If a shopper cannot find what they need, especially in health, hygiene, or other urgent categories in Reckitt’s portfolio, they may switch stores, brands, or digital platforms.

For Reckitt, the execution challenge was especially complex because in key markets, including the United States, store-level execution often happens through external brokers and merchandising partners.

These partners are essential to reaching thousands of stores, but they also create a management challenge: significant execution spend, limited in-store time, inconsistent visibility into what was completed, and a harder path to proving return on investment.

Reckitt saw an industry-changing opportunity: move sales execution from a downstream activity to a true engine of commercial excellence—helping field teams and brokers know which stores to prioritize, what issues to fix, and which actions would maximize sales ROI.

The company had plenty of data. In some markets, it had millions of data points from sales, inventory, store visits, displays, planograms, and retailer systems.

“Smart Execution is where data becomes action,” Reckitt’s IT&D Director Global Sales Sam Peerzada says. “It takes millions of signals and translates them into precise, store-level decisions—where to go, what to fix, and how to unlock growth. Before, we understood the landscape. Now, we know exactly where to move.”

That gap had a material impact. In the United States, where broker teams often have only 30 to 60 minutes in store, long checklists meant valuable time could be spent searching for low-priority issues instead of fixing the ones that mattered most.

The challenge was also different market by market. In Brazil, where Reckitt had a large in-store sales force, the question was how to focus hundreds of representatives on the highest-value opportunities. In the United Kingdom, the challenge expanded further: how to adapt the approach to different retailer systems, data maturity, and merchandising models.

Built with McKinsey, Smart Execution uses modern technology, including gen AI, real-time data, and a new operating rhythm to help Reckitt predict execution issues, prioritize the actions that matter most, and help field teams and broker partners solve problems faster at the store level.

For US Vice President of Retail Sales Lassamy Sesay, who joined Reckitt after years at a consumer-packaged-goods company known for strong sales execution, the capability stood out immediately. “I was blown away right away—I hadn’t seen anyone leverage technology in the way Smart Execution did,” Sesay says.

Building on the RGMx foundation of integrated data, AI, value tracking, and business-technology collaboration, Smart Execution brings those capabilities into the physical store. It ingests data from point-of-sale trends, inventory signals, promotional calendars, SKU characteristics, planograms, store-level performance, and execution feedback. AI models then detect patterns humans would struggle to see at scale, moving teams from static checklists and delayed reports to targeted recommendations: the right store, SKU, issue, and action.

A store in Massachusetts, a store in Texas, and a store in Seattle may appear unrelated, but the model may identify that they behave similarly for a specific product or execution issue. That kind of clustering would be nearly impossible through conventional analysis.

Global Channel and Sales Capability Director Tom Redfern describes the leap this way: “We’re not doing anything that hasn't been imagined or even done before. However, just one of these analyses might have taken us four days to complete, and by the time we had completed it, the answer would already be outdated and irrelevant," he says. "By comparison, Smart Execution can do these analyses for thousands of stores and items before the working day even starts.”

Critically, Reckitt and McKinsey did not approach Smart Execution as a technology project. They approached it as a business transformation—how to improve product availability, display compliance, productivity, and category growth. The modern technology and AI were built around that problem, not the other way around.

That required deep collaboration across Reckitt’s global and local business and technology, IT, market, and broker teams. Building on the trust established through RGMx, Smart Execution extended the model into a more complex environment: thousands of stores, multiple markets, different data structures, external broker partners, and frontline teams who needed to trust recommendations in the field.

The transformation was iterative by design, Peerzada says: “People went into the store, understood what was wrong, and then came back and reapplied that into technology. It was a perfect example of how AI and humans were augmenting each other.”

That “progress over perfection” mindset also shaped capability building.

McKinsey Partner Evagelos Athanassiou says the foundation for that transfer was built from day one. “We did not build Smart Execution for Reckitt; we built it with Reckitt,” he says. “From the beginning, the teams were mirrored—McKinsey and Reckitt data scientists, engineers, and business leaders working side by side—so capability transfer happened every day, not at the end.”

That side-by-side model included co-location, two-week agile sprints, daily check-ins, and structured coaching so Reckitt teams could own each step of the design and build before taking over fully.

The UK rollout, for example, was driven entirely by Reckitt technology and business teams—evidence that Smart Execution had moved from external support to internal muscle.

McKinsey Senior Partner Abdul Wahab Shaikh says scalability was also designed into the technology from the start. “The goal was never to create a one-market solution,” he says. “We designed the data and AI components to scale across markets with different levels of data accuracy, granularity, and maturity—so Reckitt could adapt the capability to new customers, retailers, and geographies.”

The results have been tangible. In the United States, display compliance improved from roughly 50–55 percent before Smart Execution, to about 70 percent or better after deployment—a “massive” change, says Redfern.

Elsewhere in the United States, identification of out-of-stock issues declined to 1.5 percent, reduced from 2.5 percent, and display compliance increased to more than 70 percent, from 50 percent across consecutive visits over two months. This created a step change in predictive accuracy. Reckitt also saw tangible commercial performance, including a sales uplift of roughly 2 percent in its largest modern-trade-led markets such as the United States, to 5–6 percent in traditional-trade-led markets such as Brazil, alongside significant improvements in visit quality, display compliance, and incremental sales.

Smart Execution also created new ways for Reckitt to engage retailers. Store managers saw fewer execution gaps, better availability, and broker partners arriving with clearer information. For retailers, Reckitt could bring insights to improve category performance, reduce lost sales, and support more reliable omnichannel fulfillment.

Redfern sees this as one of the biggest unlocks. “Smart Execution allows Reckitt to treat a large retail network as if each store were being individually studied every day,” he explains. “We can manage it to that depth—just at a scale of thousands of stores.”

For field teams and brokers, the shift was equally practical: Checklists went from 40 static tasks to 15 high-impact tasks in some stores, resulting in lower broker costs for Reckitt and less wasted effort for partners.

“It improved the quality of work for the broker,” Sesay says. “It made them more efficient and more productive in a massive way.” That matters because frontline execution is human work. The technology does not replace the broker, the merchandiser, or the store relationship, but enables them to be more effective, spending more time solving problems, selling, and building trust.

Peerzada sees that as a core principle of the transformation. “The endgame for technology is: How can technology and data make people’s lives simpler by making them smarter?” he says. “Technology has to serve people and problems, not the other way around.”

That mattered because broker and merchandising spend had often been difficult to manage with precision: Companies could see the cost, but not always the return at the store, SKU, or task level.

Smart Execution also changed how Reckitt thinks about execution investment, from a cost to manage to a growth lever to optimize—a shift that reached board-level conversations. Before Smart Execution, execution discussions often centered on cost: how much Reckitt was spending on brokers, merchandisers, and store coverage, and where efficiencies could be found. As the data matured, the conversation changed. Reckitt could now show, with store- and SKU-level evidence, where sales execution investment created growth, where incremental broker coverage paid back, and where better shelf availability could unlock category value.

“Smart Execution is a tenfold game changer’” Sesay explains. “Not because it replaces human judgment, but because it gives people the insight, focus, and confidence to do work at a scale that was simply impossible before.”