With push anchored in its collaboration with Salesforce, the company is expanding into tier 2 and 3 markets and plans to double its retail footprint over the next three years
Ice cream brand and candy and cone manufacturer Dairy Day is stepping up its next phase of growth by embedding AI and data-led intelligence across its sales and distribution network, with a sharp focus on improving last-mile execution and driving higher on-ground productivity. The company is using real-time insights to enhance order accuracy, reduce stock-outs and optimise field operations in a highly seasonal, temperature-sensitive category.
Highlighting this shift, Venkateswaran Krishnamoorthy, Chief Technology Officer, Dairy Day Ice Cream, told BW Retail World that the role of technology is evolving to augment sales teams on the ground by equipping them with the right data and nudges, while continuing to keep the human element central to retailer relationships.
The push is anchored in its collaboration with Salesforce, with Dairy Day migrating its distributor management system (DMS) and sales force automation (SFA) onto a unified platform to enable end-to-end visibility, from warehouses and fleets to retail freezers. Krishnamoorthy added that company currently services over 80,000 outlets through over 500 distributors, with 100 per cent of its general trade business now tech-enabled.
“For an FMCG company, the moment of truth happens when the sales rep goes to take orders. If at that stage we can bring AI-driven efficiencies in the flow of work, at the moment of truth, when the sales rep is in the field talking to the retailer, talking to dealers, punching an order, the efficiency benefits are maximum,” highlighted Aditi Sharma, Regional Vice President and Head of Consumer and Manufacturing Industry, Salesforce India.
The company is expanding into tier 2 and 3 markets and plans to double its retail footprint over the next three years as it works towards its RRs 1,000 crore revenue ambition. As it scales further across markets, the focus is on improving demand forecasting, enabling data-backed order generation and strengthening cold chain visibility, while building a flexible system that can support evolving schemes, incentives and expansion into new regions, he added.
Enhancing Sales, Cold Chain Visibility
Dairy Day’s collaboration is anchored in consolidating fragmented systems into a single, scalable platform to drive end-to-end sales transformation. As Krishnamoorthy explained, the company moved away from “individual bespoke solutions catering to point use cases” to “bring the entire capability onto one single platform,” enabling flexibility to scale across outlets, distributors and regions
“Dairy Day has ambitions to grow and scale across multiple outlets, distributors, and regions. We needed a platform that met our current needs but also allowed us to scale and stay flexible in line with the organisation’s evolution. We want to quickly respond to what the competition is doing, introduce new schemes, new incentive structures, and so on. Our existing platforms did not allow for that,” he pointed out.
At the core of this transformation is a hybrid approach, “60 per cent ready out of the box” with the ability to “continuously customise and evolve for the remaining 40 per cent”, with Salesforce emerging as the preferred partner. The implementation began in 2024, starting with asset management, particularly tracking retail freezers across markets. Today, Dairy Day is using the platform to monitor the “end-to-end lifecycle of these freezers,” linking their performance directly with sales throughput through integrated DMS and SFA systems.
Hybrid Offline-online Model
Salesforce’s approach to solving the offline-first challenge in fragmented tier 2 and 3 markets hinges on a flexible, hybrid model rather than a one-size-fits-all system. As Aditi Sharma explained, while the platform offers “100 per cent offline capability,” most deployments today lean towards “always-on functionality with minimal lightweight offline sync,” reflecting improving connectivity on the ground.
However, she noted that real-time AI-led interventions depend on connectivity, noting that “if you need a real-time nudge… the app has to be online,” otherwise it becomes “a rule-based, post-facto recommendation.”
The solution, therefore, is use-case driven, balancing online and offline modes depending on network conditions. In areas with stable connectivity, companies prefer always-on systems, while in patchy environments, including “temporary loss of connectivity, even in metros such as basement outlets,” a hybrid model ensures continuity.
From Dairy Day’s perspective, the priority was ensuring that technology does not disrupt frontline execution. Krishnamoorthy noted that the system was built to reflect “market reality,” enabling sales reps to continue working seamlessly even in “dark spots where connectivity fails,” with data syncing back once connectivity is restored. As network infrastructure improves, the company is increasingly leaning on always-on capabilities, which he said will be “crucial for the AI journey” as it scales data-driven decision-making across markets.
Shift Towards Data-backed Ordering
Order generation in traditional trade is currently a blend of human judgement and data-led inputs, rather than being fully automated. Krishnamoorthy noted that it is “neither 100 per cent manual nor 100 per cent data-driven,” but a combination of “data and science and the art of selling.” With sales reps now equipped with real-time insights on handheld devices, conversations with retailers are becoming more informed, enabling better order quality. The larger goal, he added, is to move towards “more data-backed order generation, removing the guesswork” over time.
Sharma highlighted that the industry has “underestimated the technology savviness of the feet on the street,” with deep smartphone penetration driving acceptance of app-based systems. What is accelerating adoption further is the value delivered to sales reps through “intelligent nudges,” which directly help them improve performance and earnings, making the technology more relevant in day-to-day operations.
The role of these systems has also evolved significantly, from being tools that merely captured data for headquarters to becoming real-time decision aids in the field. Sharma pointed out that these “micro-moments are becoming very important,” as systems now actively guide sales reps during interactions. This shift is also reflecting in demand trends, with companies increasingly looking to “leap from a transactional system to an intelligent, AI-driven one,” signalling a broader transition towards data-led sales execution.
Augmenting Sales Reps
Dairy Day sees AI as an enabler rather than a replacement for frontline sales teams, with the role of the salesperson evolving towards a more data-backed yet relationship-driven function. Krishnamoorthy explained that the role remains twofold, executing in-market activities and maintaining retailer relationships, while AI helps “augment what the salesman does” by providing “the right nudges” and “the right data points” to improve efficiency.
While performance will be increasingly “maximised… using technology and AI,” he emphasised that sales reps will continue to be the “face of the company” and the “brand ambassador,” with the human element remaining central.
To ensure AI acts as a support system rather than a surveillance tool, the focus is on contextual, real-time assistance. One example is the “smart basket” approach, where the system nudges sales reps if key SKUs are missing from an order, prompting them to “review the order” and improve bill value.
Similarly, AI-driven prompts guide reps on outlet prioritisation, helping them “cover the right set of outlets” and focus on underperforming areas instead of repeatedly visiting the same stores. These interventions are designed to enhance decision-making at the point of sale rather than monitor behaviour.
This tech-led augmentation is being deployed at scale, supported by a fully digitised distribution backbone. The company currently services “80,000 plus outlets” through “500 plus distributors,” with “100 per cent of… general trade business” now running on the Salesforce-powered DMS and SFA platform. This end-to-end tech enablement is enabling Dairy Day to combine data intelligence with on-ground execution as it scales its next phase of growth.
AI ROI Shifts To Frontline Execution
AI is beginning to deliver measurable ROI in FMCG, but the biggest impact is now shifting to frontline execution rather than backend systems. As Aditi Sharma highlighted, the past four years have seen a broader transformation with companies replacing “salesforce automation systems” and “dealer management systems,” but the real gains are now emerging at the sales rep level. She pointed out that the “moment of truth happens when the sales rep goes to take orders,” and embedding AI “in the flow of work” at that stage is where “the efficiency benefits are maximum.”
While companies initially experimented with multiple AI use cases, the industry is now converging on what drives real value. Sharma noted that the consensus is clear, AI works best when integrated directly into day-to-day workflows, rather than as standalone tools. This includes improving order-taking efficiency, enhancing productivity on the field and enabling smarter decision-making in real time. At the same time, scalability across large, fragmented general trade networks remains critical, especially in markets with inconsistent connectivity.
For enterprise adoption, trust and scalability have emerged as non-negotiables. Sharma emphasised that delivering isolated use cases is relatively easy, but scaling AI “across the whole general trade business” is where platforms are truly tested. Salesforce, she said, has focused on building systems that can operate reliably even in complex environments, while also expanding into adjacent areas like “internal employee productivity” and more efficient order capture.
Addressing concerns around agentic AI and reliability, Sharma underlined that trust is built through multiple safeguards. The company ensures that “customer data remains the customer’s data only,” with intermediary layers before any interaction with LLMs to prevent exposure of sensitive information. Built on a secure platform with publicly visible uptime metrics and anchored by the Einstein Trust Layer, the approach ensures that AI outputs are grounded in enterprise data, combining both platform-level and AI-level security to enable consistent, scalable deployment.
