personalized shopping with AI

When Products Look for You – How AI Changes Online Shopping

There’s not much that’s more personal than the clothes that we wear. Depending on the day, clothes might make us feel warm, cozy, cool, stylish, happy, sad, nostalgic, or something different altogether. Clothes also help show our personality to the world at large – a window into our inner workings. And, we pay a pretty penny in our pursuit of the perfect outfit (as in a US projection of $390 billion USD by 2025)1.

As we know, though, a store isn’t always conveniently located around the block. But, a smartphone can usually be found tucked away in a person’s pocket. This instant access to the Internet opens a world of possibilities for shoppers. It sets the stage for artificial intelligence in online shopping to help buyers search and discover products that they’ll love. To have personalized shopping experiences that are more relevant and contextual. It also builds a growing, always-online community for digital retailers, many of whom are expanding their ecommerce platforms.

personalized shopping with AI

One such platform is Walmart Labs. To learn more, we sat down with Sonu Durgia, Group Product Manager of Search and Discovery at Walmart Labs. Our conversation highlighted how visual search, mobile shopping apps, and robust cloud infrastructure are making personalized shopping experiences possible. (Responses have been lightly edited for clarity.)

Interview with Sonu Durgia

1. Today’s shoppers are often looking for highly personalized shopping experiences. How do you and your team use data to make these unique interactions possible, both online and offline?

SONU: At Walmart, we look at a customer’s previous purchases, in store and online, along with any products they have expressed interest in before, to personalize their recommendations and search results. Techniques like collaborative filtering are also popular to show recommendations based on what similar customers like or have bought before. Personalized search results, easy re-ordering of previously purchased items and discovery by association – all are powered by machine learning algorithms that extract intelligence from large amounts of historical transactional data.

2. How has visual search played into online shopping, especially for shoppers from younger generations (i.e. millennials and Gen Z)?

personalized shopping with AI

SONU: With millions of images uploaded online per day2 and several hours spent on social media every day3, the younger generation is discovering products more and more through social media. Visual search or visual commerce is then a natural extension of how this generation would like to shop. In categories like “Furniture” and “Fashion” – where accurately describing one’s search intent is not easy – finding an exact match from many millions of items (especially for large ecommerce platforms like Walmart) that don’t always have great metadata is even harder. China has been a leader in widespread adoption of visual commerce, as its major tech companies are getting a large share of their revenues through visual search and similar applications of this technology4.

3. A couple years ago, Walmart shared its ambitious plans to build a private cloud capable of processing 2.5 petabytes every hour5. Why do you think so many big retailers are going all in on cloud infrastructure?

SONU: A cloud-based architecture offers the flexibility of elastically scaling computing infrastructure and reduced administrative complexity through managed software.

“In the world of Big Data and AI, retailers stand to benefit from real time analytics for better search and discovery systems, pricing and margin management, recommendation platforms and real-time inventory visibility – which is especially important for successful omnichannel retail.”

In an AI-first world, large amounts of unstructured data stored for long periods of time is becoming the norm. The scale of this data is forcing new generation data centers to invest heavily in automation and intelligence for managing infrastructure for both compute and highly accessible storage.

4. Let’s talk about mobile shopping apps. How have shoppers’ new attitude toward these apps changed the shopping experience, especially relating to search and discovery?

SONU: With easy checkout and secure mobile pay options such as Walmart Pay, more than half of online transactions are now happening on mobile6. Mobile apps are becoming more popular than ever before, especially for the most loyal customers. In turn, these customers are demanding unique shopping experiences that are tailored to their preferences. Personalization in search and discovery has now become table stakes for brands to keep their best customers.

5. In 2019, an estimated 1.9 billion people around the world will buy goods and services online7. How do you see data centers enabling these real-time transactions at scale?

personalized shopping with AI

SONU: Every one of those 1.9B purchases will be accompanied by people searching for specific items or getting inspired by products that they might see on a website. The scale and complexity of ecommerce gets amplified several folds because these systems need to understand customers’ intent and match it to millions of available products. At Walmart, we see billions of search queries a year that need to be matched to millions of items that are available on our website – all in a few milliseconds. Also, not all days are equal and a typical retailer’s transaction volume increases several folds during Black Friday8.

“All this is only possible with modern day data centers that are designed for high availability, high throughput and seamless scaling.”

6. In the future, do you see a point where virtual shopping assistants could predict what we need to shop for… before we know it (aka predictive shopping)?

SONU: That IS the future of shopping. As we collect more and more data about people across every device they use around the house and on the go and connect insights from these sources, we will have enough intelligence to predict people’s next needs and wants. Not too far in the future is the time when products will look for us instead of us looking for products.

FORWARD-LOOKING STATEMENTS: This article may contain forward-looking statements, including statements relating to expectations for storage products, the market for storage products, product development efforts, and the capacities, capabilities and applications of Western Digital products. These forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those expressed in the forward-looking statements, including development challenges or delays, supply chain and logistics issues, changes in markets, demand, global economic conditions and other risks and uncertainties listed in Western Digital Corporation’s most recent quarterly and annual reports filed with the Securities and Exchange Commission, to which your attention is directed. Readers are cautioned not to place undue reliance on these forward-looking statements and we undertake no obligation to update these forward-looking statements to reflect subsequent events or circumstances.


  1. U.S. Apparel Market – Statistics & Facts.
  2. How Much Data Do We Create Every Day? The Mind-Blowing Stats Everyone Should Read.
  3. How much time do you spend on social media? Research says 142 minutes per day.
  4. Shopping With Your Camera: Visual Image Search Meets E-Commerce at Alibaba.
  5. Really Big Data At Walmart: Real-Time Insights From Their 40+ Petabyte Data Cloud.
  6. Mobile E-Commerce is up and Poised for Further Growth.
  7. Number of digital buyers worldwide from 2014 to 2021 (in billions).
  8. Every Result You Need To Know From Black Friday, Cyber Monday And The Holiday 2018 Season So Far.

Related Posts