by Saheli Sen Gupta

Biryani, Kebabs and Big Data – How Petoo is predicting what you want and when you want it

Jun 25, 2015
AnalyticsAndroidBig Data

Food startup Petoo, founded by the JustEat founders, is using predictive analysis to map food patterns in Bangalore, and eventually, all over India.

If you thought technology is still miles away when it came to food, you are wrong. Petoo is bringing predictive analysis and food together into one beautiful plate of biryani with kebabs on the side. Here’s how.

After five years with JustEat, entrepreneurs and foodies, Abhishek Mandal, Ritesh Dwivedy, and Kumar Setu launched another food startup, Petoo (Hindi for ‘gluttons’, pronounced as pay-to).

“We have witnessed the entire era of food-tech industry and this has helped us in understanding order patterns and customer preferences. Our combined experience, along with the data we get helps us to predict real time demand of our customers,” says Mandal, the food expert of the team.

Petoo’s data helps it determine and predict not only sales and profit but to also the most-selling item. For example, Mandal explains how vegetarian dishes are more popular on Tuesdays and Thursdays while customers order non-vegetarian dishes on weekends.

Petoo’s data points, however small they be, have helped the company maintain a wastage between 5 and 10 percent of its sales. The company is saving up on infrastructural costs and follows a hub-and-spoke model with one central kitchen serving 12 outlets, each at an average distance of 3 km.

“We send out food twice, once before lunch and once before dinner,” says Mandal, “Our Petoo central app helps us monitor our inventory from the kitchen itself. While we can predict normal daily sales, one can never predict sudden bulk orders.”

Petoo works with a 20-minute delivery system and is tied up with organized logistic providers. They use optimized routes mainly because there is only one kitchen serving all outlets. A dedicated vehicle for replenishment is always available in case of a sudden bulk order.

India’s vast geography and varied culture means that food is a very hyperlocal phenomenon. “There is no PAN-India player in the food industry. While dosa and poha might be popular breakfast dishes in South India, the northern part of the country might not prefer it,” adds Mandal.

Using Big Data, Petoo determines the ordering pattern of each area individually. According to its data, areas in South Bangalore like J.P. Nagar, Bellandur and Jayanagar are home to traditional South Indian families who prefer vegetarian food than areas like Koramangala, Whitefield and BTM Layout which are more cosmopolitan with people from all over the country and has mostly non-vegetarian eaters.

“People are different in every way and their eating habits are different, so their ordering patterns will also be different,” says Mandal.

The initial data was derived from an array of surveys and from JustEat. The surveys indicated that biryani was the obvious choice to launch with. “Bangalore loves biryani! You can serve biryani here at any time and they will love it,” he laughs.

While most restaurant increase their delivery time during rain, matches and festivals, Petoo’s data reveals that customers are more likely to seek delivery on these days. Mandal also talks about the importance of customer feedback. Every customer is asked to share their food, delivery and overall experience.

Although Petoo is only three months old and its data points are very small, analytics is helping the business grow exponentially. “As long as the data makes sense for you, it doesn’t matter if you have one or 540 restaurants. Big data will work on your system. But, of course, more the restaurants, more the data points and larger your predictions,” says Mandal.

“Technology has become a part of every industry. The food industry is not just food industry, it’s the food-tech industry,” says Mandal.

The company is almost ready to launch a Petoo Boy app which will work on all android phones where delivery boys can report by logging in and the system will automatically assign orders to them. It will also provide an optimized route map and once the delivery is done, the delivery boy can mark the order as complete to receive the next order.

Another app, meant for the customers will be rolled out in a month with the option of being able to track the delivery boy without having to contact him, a model similar to cab companies.