Key takeaways from a recent event of Industry leaders organized by FieldAssist
The changing face of FMCG
FMCG as we know is going through a flux of change. Things are moving a little too fast for the Fast-Moving Consumer Goods industry, faster than many can manage.
India is a peculiar market to operate in at the moment. First, almost every single metric is moving ahead at a rapid pace. India recently became the most populous country on Earth, leaving China behind. India’s e-commerce and q-commerce markets, though not currently as large as India’s population would indicate, are exploding.
Gone are the days when brands dictated consumer choices. Today, it’s increasingly the opposite: consumers dictate brands’ choices.The power dynamic today has shifted from where it was even 10 years ago. Consumers have much more power in their hands, and brands need to adapt or risk becoming irrelevant. This has made it imperative for CPG firms to understand the brand down at a very deep level, right down to their micro habits.
To discuss these challenges and their solutions, FieldAssist recently had the honor of hosting the titans of industry in a series of two panel discussions in Mumbai. Industry thought leaders shared key insights on the past, present, and future of the CPG sector.
AI: The new world, or just the new buzzword?
AI today is a major topic of discussion in almost every conference – in every industry. But in recent years, especially for CPG, it has gone beyond being just a buzzword, and has started achieving positive outcomes. Panelists shared their experiences and success stories of deploying AI and ML, the need for the right data, and connecting it back to the right models and the front-line staff.
The industry has been observing AI for almost a decade, but it has taken time to take off. One of the main reasons for this was the extensive time and effort required to train it, Sachin Patil, Infrastructure Lead at Microsoft Azure. This is where LLMs have been immensely helpful, argued. They have enabled the creation of a built-in learning model that you can adapt to your specific project needs – at scale. Sachin gave two interesting examples of two successful LLM implementation:
- When McDonalds went to China, it deployed AI and LLM for achieving operational excellence in language contextualization and translation, as well as overall customer experience and inventory management. This was a case where we saw early success in
- deploying LLM.
Back home in India, ITC, MARS implemented LLM in one of their apps called “Kisan Mitra”, where a farmer on the ground can call up using his own mother tongue, and get responses in the same language about ITC’s products. This app supports 21 Indic languages.
Suchindra Khaidem, Director IT South Asia, Kellanova (formerly Kellogg Company), emphasized the importance of building trust in an AI system, which is subject to how clean your data is.
Data is (still) the new data
The general consensus was that the industry today has no shortage of data. The challenge is to extract predictions from the data using ML technology: placing the right filters and getting the right insights from the data, analyzing past numbers, and picking up trends for the future. As Sachin put it, “Do you want to master the data, or do you want the data to master you?”
Vinayak Vetekar, Senior Vice President, Flair Writing Industries, argued that shortage of data is not the problem. Gathering and accumulating data was a challenge 20 years ago. Today, the industry generates multitudes of data daily. And it also possesses the right tools, such as Power BI, to scrutinize and analyze that data.
Brands should also be clear on what they are mining for. How well firms are able to tailor technology to their specific requirements will determine how well and how fast they are able to drive growth.
Internal data can also be highly useful in predicting future trends. Eureka Forbes India for example, under its Head of IT Kunal Handa, puts a lot of effort on persona-based decision making. As an example of the sort of deterministic variables the firm deals with, consider that in the Indian urban landscape, even Tier I cities such as Delhi have a wide variation on water quality. This in turn has a deep impact on everything from sales planning to execution.
Digitization in sales planning and analysis
For increasing market penetration, CPG firms are looking for increased digitization due to the high level precision required. However, customization is also important, instead of a one-size-fits-all approach.
A key question today is what level of digitization should be adopted for specific local conditions and maturity levels. The answer depends on where the brand stands – and where the market stands in relation to the brand. In the industry today, we are seeing both smaller digital tweaks as well as large-scale digital transformations in CPG.
Rishabh Rastogi, Lead Center of Excellence – FMCG, Miebach, talked about how CPG firms are incorporating a demand-sensing angle to sales forecasting, and are not solely reliant on past data. Past trends are augmented with variables using hypothesis testing, which enables brands to improve forecast accuracy.
Earlier methods of sales forecasting did not account for the impact of independent factors such as say, the weather or an upcoming sports match. Recently however, with new tools available for demand sensing and heightened awareness, the industry is witnessing increasing attempts to account for such ancillary variables in regression models.
In the next 5-10 years, according to Rishabh, the Indian CPG industry is also expected to incorporate Neural Networks, with function fitting. This will empower brands with improved forecasting, targeted sales, and a significant reduction in waste.
Balaji Venkatraman, Director Consulting, PwC, pointed to channel conflict in the market today, with quick commerce, e-commerce, GT, MT – all of them vying for the attention of the same consumer. CPG organizations will need to handle accurate demand sensing for all these channels.
On a lighter note, Balaji added, the biggest Machine Learning and Artificial Intelligence that has happened has been the Indian retailer. He handles an endlessly complex array of countless SKUs, and he knows exactly when to order what, how much he wants to sell, and at what rate!
The perfect Perfect Store?
A lot of discussion was focused on Perfect Store – a concept that is often debated in generic terms. Today, the proliferation of AI-based solutions has empowered companies to define their own Perfect Store for specific outlets. Tech-enablement has enabled firms to drill down and be as specific as the market demands. Brands can now ask, “What does Perfect Store mean for this particular store?”
The approach of Siladitya Sarangi at Deoleo, where he is country manager, is to set up cohorts of similar stores – stores that satisfy a similar set of conditions – and then define a “Perfect Store” for that cohort.
Another challenge that brands face is that the data used to train models is all internal. How to get external data? Can firms from non-competing categories share data with each other? How to get data from competitors? How to look at data from your competitors’ perspective? Looking at data from the outside often makes the market significantly easier to understand.
The perennial debates: Rural vs Urban, MT vs GT
How are rural consumers different from their urban counterparts? Is the urban market really saturated? In the words of moderator Bibaswan Dutta, Digital RTM Lead, EY Parthenon, “Where does India end and Bharat start?”
The general consensus was that the rural market cannot be underestimated. Every company needs a definite results-oriented strategy for rural consumers.
Khushru Mistry, CTO at GM Modular, mentioned that there is no perfect sales channel today. Mom-and-pop stores are becoming the new “points of interconnect”. These stores are delivering daily groceries to consumers via novel channels such as WhatsApp, without even needing a separate app, and q-commerce companies have an interesting fight ahead of them. Sunil Pudi, CEO, Kores India, argued for the longevity and resilience of GT and the local kirana stores that would co-exist along with q-commerce. Quick commerce is something brands need to seriously look at – and recognize both its strengths and weaknesses.
The importance of Gamification
The proof of the pudding is in the eating, and the proof of the technology is in the usage. How can brands incentivize their staff to drive adoption? Kushal Agrawal, Co-founder, Xoxoday, talked about the importance of gamification-based incentives and innovative nudges. Task- and milestore-based incentivization can provide instant gratification and also motivate users to maximize performance. Khushru Mistry, CTO, GM Modular, talked about providing the right incentives and loyalty programmes for specific segments, and mentioned an interesting anecdote about how GM provides medical insurance for electricians to increase brand loyalty, and thus boost secondary sales.
Siladitya Sarangi, Country Manager, Deoleo stressed the value of empowering the sales staff to embrace volatility. How much tolerance to ambiguity does the salesperson have is the key question, added Manish Vyas, CEO, Wipro Yardley.
The Future of CPG
CPG firms today are moving beyond merely “keeping the lights on” with technology. Cutting-edge technology is necessary, but insufficient. It takes the right business-first strategy and efficient execution to succeed in today’s market. Technology should be a business enabler and a business mover. This is particularly true today, with the dawn of enterprise-ready AI.
The success of any technology lies not in its mere deployment, but its adoption. As firms make the journey from Information to Insight to Impact, FieldAssist is proud as well as humbled to partner with great brands in their journey to unlock the next level of CPG growth.
About Post Author
