IOS reported how Flipkart is banking on data and artificial intelligence (AI) for sale prediction. The etailer’s aim is to enhance customer shopping experience and reduce returns. Turns out, the ecommerce leader has been working on an AI project from February 2016.
Christened as ‘Project Mira’, it uses the innovative techniques of AI to make online shopping experience as personal as offline shopping experience. The work started soon after Binny Bansal took over as Flipkart’s CEO replacing Sachin Bansal.
While talking about this project, Ram Papatla, VP – Product at Flipkart said,
“On 28 February (2017), we launched the first version of our conversational search experience. Now, our users with broad intent (searching for, say, shoes or bedsheets) are guided with relevant questions, conversational filters, shopping ideas, offers and trending collections.”
Why Project Mira?
Studying the product returns pattern prompted Flipkart to design a tool, which would assist buyers in purchasing what they want. Out of the 400,000 shipments that Flipkart dispatches daily, 10-11% is returned. Clothing and fashion accessories witness the most number of returns.
This usually happens due to the gap between customer’s expectations and the quality of the delivered product. Flipkart wants to bridge that gap by introducing elements similar to offline shopping at physical stores.
Project Mira intends to:
- Make it easy for buyers to search for what they are looking for
- Accurately predict what customers want to buy
- Personalised recommendations
- Interact with shoppers virtually while they are browsing on the shopping portal
- Imitate the brick-and-mortar shopping experience in the online retail space
- Streamline back-end operations
“When we looked at the (product) returns data and when we looked at data from shoes and lifestyle (categories), we saw a bunch of mismatch of expectations from our customers in terms of size and fit issues. If only we could have asked them one question we could have given the right response. It was an internal cry. We have enough evidence (to say) that had we ‘talked’ to Mira (the online Indian shopper) we could have solved it,” Papatla shared.
What are the challenges?
Interpreting customers’ ‘search terms’ and streamlining the back-end operations are two of the main challenges that the online marketplace is dealing with.
Accurately classifying the products, correct product descriptions, right product images and avoiding duplications is imperative for replicating store-like shopping experience. Failure to do this could lead to inaccurate predictions.
To avoid this from happening, Flipkart is assisting its sellers so that Project Mira works at an optimum level.
“We have computed so much training models that if it is an image mistake, we can tell (sellers) to not upload as these kinds of images in the past have seen lower conversions. We are doing it with about 300 sellers now, we are handholding them. Through our sales team, we are providing the content to the sellers, saying this is your quality score, this is the score card for this month, and by the way, here are five opportunities to improve.”