Express Computer recently conducted an insightful interview with Rahul Bharde, Senior Vice President and Head of Analytics & Insights at Jubilant FoodWorks Ltd., shedding light on the strategic utilisation of AI and ML technologies, Rahul Bharde discusses how Jubilant FoodWorks leverages advanced analytics to drive operational efficiency and enhance customer engagement.
How is Jubilant Foodworks leveraging AI and ML technologies within its analytics and insights strategy to drive operational efficiency and customer engagement?
As the head of analytics at Jubilant Foodworks, our vision is to enable our customers, employees, and machines to make better decisions using data and advanced analytics. This theme is pervasive throughout our operations.
We leverage AI and ML technologies to drive operational efficiency and enhance customer engagement/
Our customer science engine & insights capabilities help understand every customer uniquely, predict customers’ future actions and prescribe Next Best Action that helps in personalising customer experience across every touchpoint.
Each year, as a JFL group, we open 200 to 250 stores, which necessitates a deep understanding of our catchments. Advanced analytics help us identify optimal locations for new stores and determine which cities to expand into next. Additionally, we use data and technology to manage our delivery network efficiently.
Our operations, spanning over 400 cities in India, also benefit significantly from advanced analytics. We manage inventory for our stores on a daily basis, with predictive models ensuring the right amount of fresh ingredients for pizza making while optimising inventory investment.
Can you provide some specific examples of how AI algorithms have been deployed to optimise various aspects of the business, such as menu optimisation, demand forecasting, or personalised marketing campaigns?
We are personalising ordering experience on our Domino’s App with our personalisation engine providing meaningful recommendations across the customer journey. Similarly, with our customer lifecycle management communication, the ML models determine the right target audience, the optimal timing for targeting, and the kind of offers to provide the best value to customers. All of this is driven by the AI and ML at the core of our customer science engine.
Each year, as a JFL group, we open about 200 to 250 stores, which necessitates a deep understanding of our catchments. Advanced analytics help us identify optimal locations for new stores and determine which cities to expand into next. Additionally, we use data and technology to manage our delivery network efficiently.
Our auto-indenting tool utilises advanced algorithms to forecast daily ingredient requirements at each store. This allows us to optimise ingredient orders, ensuring maximum availability while minimising sales loss and wastage. With this tool, we strike the perfect balance between efficiency and customer satisfaction
Can you provide specific examples of how AI algorithms have been deployed to optimise various aspects of the business, such as menu optimisation, demand forecasting, or personalised marketing campaigns?
When it comes to our customer-facing customer science engine, we heavily rely on first-party data, which is a key part of our data strategy. We focus extensively on customer behaviour during our model-building process and deliberately avoid using demographic data that might introduce bias.
We have a rigorous process for validating our models at various stages—training, testing, and out-of-time validation. This thorough process ensures that our models are both accurate and fair. We continuously monitor the health of our models to prevent biases and avoid using variables unrelated to consumer behaviour. We maintain a robust stage-gate process with multiple reviews before any model is deployed into production.
How do you utilise and protect data responsibly, and what measures are in place to ensure data security?
We generate tons of data. If you look at it, we have a very strong food app, and we’re a large operations company with 2000+ restaurants. We’re also a supply chain and manufacturing company with our own factories and distribution centres supplying material to all our stores. So, we generate tons of data from our manufacturing units, distribution centres, stores, energy and environment systems, bikes, customers, our restaurants as well as our app. We continually collect and expand our data platform.
What we’ve focused on is democratising data, making it easily accessible to everyone in the organisation while ensuring access control based on purpose and role. Everyone has a very simple way to access the data relevant to them, usually just a few clicks away. This data spans across 50+KPIs and 50 plus reporting levels, allowing users to track trends from an hour ago, yesterday, or even five years back. Because these KPIs are centrally defined, the information remains consistent regardless of whether someone is at the corporate office or in a regional team in the northeast, south, west, or north. All they need to do is access the platform with their field link.
The access is fit-for-purpose, ensuring everyone gets the data they need to enhance their roles, but nothing more. This access control is strictly managed. We also follow stringent guidelines on data encryption and have numerous controls in place to ensure that the data we hold is user-consented and used for the intended purpose. Additionally, we have policies for archiving data to manage its lifecycle properly.
As we continue to build our practice, the amount of data used and the advanced accessibility we’ve developed ensures we provide value across the entire value chain of the organisation.
How do you see leveraging GenAI in your work with data, analysis, and insights?
There are tons of opportunities to use generative AI. We’re building a number of different capabilities, covering use cases across customers, operations and employee productivity. We have a problem-first approach where technology comes second. Whatever is the best fit for the purpose and can scale, we’ll go with that. Whenever we’re confident to go live, we’ll obviously come back with that update.
In the next 6 to 12 months, how do you plan to better utilise or leverage the potential of technology in your operations?
We’ll continue to build and leverage data and technology to provide better value for our customers, enhancing their experience and that of our employees. This strategy applies across all the brands we work with.
We’re continuously evolving our app, an ongoing workstream that encompasses everything from the signup process to ordering and payment methods. These areas are constantly being improved, both in our digital commerce platform and our personalisation engine. Similarly, we’re focusing on several aspects of marketing, particularly customer lifecycle management and pricing, which are driven by scientific insights. We also do substantial work in location science to guide our expansion strategies, and in operations and supply chain planning through our forecasting engine. We’re exploring various generative applications to enhance productivity and customer experience.
These initiatives are aimed at helping our employees, customers, and systems make better decisions using data.