Info-tech

Flipkart uses AI-led innovations to burnish its record

Sangeetha Chengappa Bengaluru | Updated on January 12, 2018

Has equipped e-tailer to deliver contextual solutions for customers



A dedicated team of 25 data scientists at India’s largest e-tailer Flipkart is using artificial intelligence to track customer behaviour which has helped the company to reduce several pain points related to its customers.

This has resulted in its NPS (Net Promoter Score) growing 14 per cent over the last 12 months, while customer pain has reduced 25 per cent in the last six months and customer query resolution has gone up 14 per cent since the launch of chatbots in April.

Net promoter score is an index which measures the willingness of customers to recommend a company’s products or services to others.

Ram Papatla, Vice-President, Product Management at Flipkart, told BusinessLine, that sharp focus on AI-led innovations over the past 18 months, enabled by a large data pool of Indian shopping behaviour that Flipkart has built over 10 years, has equipped the e-tailer to deliver highly contextual solutions for customers in the form of personalised recommendations.

“If a customer keys in a query for running shoes, we show only the category landing pages of the particular brand the customer wants to see, in the price point and styles that is preferred, as gauged by previous buying behaviour, therefore, ensuring a faster, smoother checkout process” he explained.

Deep customer insights

Since its inception, a decade ago, Flipkart has amassed a base of 100,000 sellers and over 100 million shoppers whose shopping behaviour has been mined to come up with 40 insights for each customer. The e-tailer has built a machine learning model for each customer which provides insights such as, whether the shopper is a male or female, their brand affinity, store affinity, price preference, frequency, volume of purchases etc, which gets stronger with time as the machine learns more and more about the customer. “We know that customer X is looking for chiffon sarees, even if the query just says “sarees,” and that customer Y is looking for kurtas from a particular store, even if the query doesn’t mention the store name,” said Papatla.

As the stand-off between Flipkart and Amazon is no closer to being resolved yet, victory in the war for dominance in the Indian e-commerce market which is estimated to quadruple to $60-70 billion by 2019, will go to the e-tailer who delivers superior customer experience and deep personalisation, say experts.

Customer experience

“The difference between brands like Apple, Uber and Airbnb with competing brands, boils down to superior customer experience and personalisation - the only two ways to garner brand loyalty” observed Rajeev Banduni, CEO of GrowthEnabler, a London-based provider of insights and intelligence on disruptive start-ups. For e-tailers to deliver customer experience and personalisation, they must automate business acumen using AI and Cognitive technologies that allows them to imitate human intelligence, he said.

Others in the e-commerce rat-race have also upped the ante in using AI. While Ola recently appointed a Chief Data Scientist, Badri Raghavan with expertise in AI and machine learning to work across business units to gain efficiencies in demand/supply prediction, location intelligence, driver behaviour and performance, Myntra has launched Moda Rapido, an in-house fast fashion brand powered by AI, three months ago, becoming the only fashion brand in the country to offer a fully-automated design collection without any human intervention.

Papatla claims that Flipkart has the strongest sense of consumer awareness, consideration and purchase data for any brand in India, surpassing the likes of Google, which he says, only has search and shopping trends but no data on how aware consumers are of a brand, how many of them actually considered buying that brand and how many purchased that brand. “Our strategy is clear – we will continue to invest in AI, bots and automation to be relevant to our customers over the next decade” he said.

Published on June 26, 2017

Follow us on Telegram, Facebook, Twitter, Instagram, YouTube and Linkedin. You can also download our Android App or IOS App.

This article is closed for comments.
Please Email the Editor