AI at the Top: Walmart

If you live in a Tier 1 city in the countries Walmart operates, there is a 70% chance that you have visited Walmart or its subsidiary at least once - online or offline.

That’s how huge Walmart is.

They operate in 24 countries with over 10500 stores, serving over 270 million customers a week.

With millions of customer interactions, data often determines consumer behaviour, interests, and what merchants should focus on.

In fact, Walmart’s CEO Doug McMillon mentioned, “The quality of data in our catalogue affects nearly everything we do.”

New normal is AI answering most of the customer questions unless a human is really needed. But Walmart takes it a step ahead with some agentic nature.

It is designed to take actions like ‘cancel order’, ‘initiate refund’, etc., on behalf of customers.

It does sound peaceful to not take 4-5 extra steps to cancel the order you didn’t like.

Customers’ issues are resolved instantly. Humans perform complex tasks.

Each user gets a personalised home page based on their interests, search and purchase history.

In fact, the customer satisfaction score was up by 38% since the brand introduced AI.

PS: The actions customers take help the brand optimise even more, then use insights from the data in their marketing campaigns.

Not only this, Walmart is testing solution-based searches.

Instead of balloons or candles, you can search “help me plan a birthday party for my niece.” It shows all the items you need to host a happy party.

Providing no-brainer CX (Customer experience) on the app gives Walmart the competitive edge against the online giants like Amazon.

Imagine rushing to Walmart after a workout to grab a protein bar, and the store says, “Sorry, we don’t have your favourite flavour.”

Hurts, right?

It doesn’t happen anymore.

Here’s what used to happen: Walmart merchants are responsible for every single item you find on shelves/online. They spend hours understanding data and making decisions on how they manage inventory, send shipments, supplies, etc.

The data analysis and reporting are time-intensive, and Walmart wanted to turn its Merchants into Flash.

So they developed an AI assistant, Wally (cute name), to process all complex data and answer merchants’ questions in natural language. No complex sheets or graphs.

Wally enters and analyses data. Identifies root causes behind why certain products are over/underperforming. Raises tickets, resolves issues and does all the complex calculations to predict customers’ interests.

Wally ensures the right products are at the right place at the right time.

Walmart uses robots in its warehouses to move and sort goods. In stores, robots scan shelves to find missing or misplaced items.

Suresh Kumar, CTO, Walmart, mentioned they have created multiple agents for different purposes.

While they are critical and improve efficiency, multiple agents can quickly become overwhelming and often confusing.

So the development team decided to go beyond individual tools and build a unified, company-wide framework.

Every time Walmart rolls out a new agent, it makes life simpler for customers, associates, and partners.

The heart of this framework includes four super agents.

In the CTO’s words,

“Over the next year, super agents will become a more visible part of the Walmart ecosystem.”

Walmart has proven to use speed to its advantage, especially when you’re running a company that handles large data.

The time you save is insane!

With analysis and prediction happening in seconds, you have all the resources to act swiftly.

Speed is the biggest advantage of the AI era.

YOUR TURN