Retail Spotlight: How Fli...

Retail Spotlight: How Flipkart Led India’s E-Commerce Market with AI

Nov 12, 2021
Retail Spotlight: How Flipkart Led India’s E-Commerce Market with AI

Flipkart, India’s largest e-commerce player, has witnessed a dramatic transformation from an online bookseller to India’s first billion-dollar e-commerce company. 

Founded in 2007 by Sachin Bansal and Binny Bansal, Flipkart has the advantages of both the home team and brick-and-mortar expertise. In 2020 alone, Flipkart reported a revenue amounting to around 346 billion Indian rupees, making it Amazon’s top competitor in the global e-commerce market. 

Today, Flipkart online shopping features 8 million products across more than 80 categories. The Walmart-acquired platform has a total of 100 million registered users, 100,000 sellers, 21 warehouses, and 10 million daily page visits. Considering its strong growth, it is not surprising that the company is planning for an IPO by the end of 2021.

Over the course of its development process, Flipkart has continuously applied Artificial Intelligence (AI) and Machine Learning (ML) techniques to stay ahead of its competition. Let’s take a look at some of their outstanding AI & ML solutions!

Machine Learning in Addressing Problems

It is of common knowledge that finding the right delivery address poses a huge problem in India. This is because the address network in India is diversified at the regional level and does not follow an organized system. Moreover, Postal Index Number (PIN) codes in India are neither consistent nor do they precisely represent geolocation, thus causing difficulties in processing last-mile delivery logistics

To address this, Flipkart’s data scientists have created an intelligent machine learning solution. They visited the company’s warehouses and delivery hubs where they observed how personnel at the sorting facility understood and classified addresses. Working with the field executives and supervisors at the hub, the data scientists then collected and validated the data sets for labeling the addresses. 

Flipkart’s machine learning system can classify Indian addresses and resolve inconsistencies with a 98% accuracy rate. The company is now working to solve the challenges of its ML solution, one of which is to identify representative features and pre-processing stages.

(Image Source: Flipkart)

Interactive E-Commerce with Flipkart Camera

Flipkart has developed an interactive camera feature that helps create immersive, AR-enabled shopping experiences for its millions of users. Customers no longer have to imagine the products in their heads as they can now have the Flipkart online shopping experience in the comfort of their homes.

In November 2020, Flipkart acquired Scapic in November 2020, expanding the platform’s capabilities across multiple use cases to increase the accessibility of this cutting-edge technology.

“The goal was to create a platform that allowed users, especially those that weren’t tech-savvy, to create AR and 3D content without writing code, and to also be able to distribute that content all over the web,” says Sai Krishna V K, Senior Product Manager at Flipkart Camera.

(Image Source: Flipkart)

AI-Driven Voice Assist Shopping

Not long ago, Flipkart partnered with Liv.ai to develop a voice shopping assistant powered by AI, deep neural networks, and advanced natural language processing technology. The aim is to help enhance the experience of all users across the country, making it as comfortable and as easy as it can be.

Flipkart camera development team adopted ML-based approaches to support free-flowing user inputs without any constraints of being syntactically correct or adhering strictly to just one language. Flipkart’s NLP system was designed to handle mixed Hindi and English input without the need to specify the input language.

Flipkart customers can expect the voice assistant to be like a tech-savvy friend, helping them find the best benefits, deals, and offers. At this rate, Flipkart's AI-driven voice assistant will democratize the field of e-commerce in India soon enough.

AI-Driven Voice Assist Shopping

(Image Source: Flipkart)

Warehouse Robots

The AI-powered bot system at warehouses enables Flipkart to shape the future of India’s e-commerce. With the help of AI-powered bots known as Automated Guided Vehicles (AGVs), the Flipkart sortation center can process more than 4,500 shipments an hour at twice the speed and with 99.9% accuracy.

The robots are cuboidal, small in size, move quickly, and are loaded with sensors. The AGV setup operates on three levels — the first, on the ground level, is where the packages are sent to the AGV area; level two is the bagging operation; level three is where the bots sort the packages according to the addresses. The combination of humans and bots gears toward speed, safety, and customer satisfaction.

“We customized and designed the AGV to be the right the solve for scale and the upcoming growth. We essentially took a long-term view by creating a future-proof automation platform, which allows easy on-boarding of new automation technology,” says Sonali Thite, Senior Product Manager, Flipkart.

(Image Source: Flipkart)

Indeed, Flipkart has shown great advancement in implementing AI and ML technologies within its platform. With these advantages, we expect its great development potential across the global marketplace in the future.

For more exclusive reports, insights, and interviews on the latest updates in e-commerce and logistics, follow us on LinkedIn or join our community as a member.

Share this with your network

Also worth your time