On May 22, Walmart released Sparky AI shopping assistant data during its Q1 FY27 earnings call, offering the first clear look at whether AI is truly driving retail results or just PR spin.
CEO John Furner stated:
'Weekly active users are up over 100% just in the last quarter. And our investments in AI have increased Sparky intelligence and response quality by 40% this year.'
Four numbers in one sentence: weekly active users doubled in a single quarter, and Sparky's 'intelligence' improved 40% this year. These are hard metrics tied to earnings, not vague enthusiasm.
35% vs 4x: Which number matters more
Walmart disclosed several Sparky core metrics:
- Weekly active users: 100%+ quarterly growth (Q1 FY27)
- Unit purchases: 4x quarter-over-quarter (Q1 FY27 vs Q4 FY26)
- User AOV vs non-users: 35% higher (current)
- Sparky intelligence/response quality: 40% improvement YTD 2026
The 35% AOV lift is widely discussed, but the 4x unit sales growth truly reflects changing shopping behavior. Higher AOV could be explained by selection bias—Sparky users may already be big spenders. But 4x unit sales means these users are adding items they previously wouldn't buy from Walmart: paper towels, laundry detergent, snacks, fresh produce—full cart replenishment. Walmart internally calls this 'intent-driven commerce'—AI not just finding products but proactively suggesting what to restock.
Specific features align with this. Walmart US CEO David Guggina noted new Sparky capabilities:
- Personalized replenishment: Proactive reminders based on past purchase rhythm
- Meal planning: Users specify weekly meals, Sparky generates a shopping list
- Smart recommendations based on inventory and delivery speed: Suggests alternatives for out-of-stock or slow-delivery items
These tasks previously required user effort; now AI pushes them proactively.
Not a single feature, but taking over the entire shopping process
Sparky now runs across website, app, and in-store. Each entry point handles different tasks:
- App: Handles 'what to buy'—checks fridge, weather, history, then compiles a shopping list
- Website: Handles 'how to choose'—smart filtering, comparison, alternative recommendations
- In-store: Handles 'where it is'—shelf location, scan-to-cart
Spanish language support was recently added. Over 20% of Walmart US sales come from Spanish-speaking regions, opening access to tens of millions of users.
The real money is in advertising
Looking only at AOV lift misses the bigger picture. Another key Q1 number: global advertising revenue up 37%, US up 36%.
The logic: AI drives more browsing, creating more ad slots; AI drives more purchases, making sellers willing to pay higher CPC for that traffic. Walmart's ad business, Walmart Connect, is the company's fastest-growing segment. Sparky acts as an accelerator—not selling ads directly, but widening the shopping funnel to create new ad inventory.
A playbook Amazon never had
Amazon's Rufus launched last year with little impact and was folded into Alexa in April. One reason: Rufus was stuck as a 'shopping assistant' without real integration with Amazon's downstream businesses (ads, Prime, AWS).
Sparky takes the opposite approach: integrated from the start with inventory, advertising, and delivery systems. When a user asks 'what's for dinner,' Sparky checks local store real-time inventory, delivery windows, and promotions—not just a recipe database.
The challenge isn't the AI model; it's connecting all company systems. Walmart spent tens of billions on digital transformation over the past decade, and Sparky now reaps those dividends.
What's next
Furner described Sparky as 'becoming AI native'—Walmart transforming into an AI-native company. It sounds like PR, but the numbers give it more substance than typical CEO rhetoric.
Two things to watch:
First, when Sparky expands beyond the US. Walmart has large operations in Mexico, Canada, and China (Sam's Club and Walmart China), but Sparky is US-only. International expansion could further boost AOV growth.
Second, Amazon's response. The Alexa + Rufus integration should show results by mid-year, and Bezos is personally overseeing AI. Walmart's earnings are a challenge Amazon must answer.
Retail hasn't seen a compelling story in a while. AI agents might truly start rewriting the narrative from Walmart's checkout counters.
Sources: Walmart credits Sparky AI agent with lifting AOV, unit sales growth (Digital Commerce 360); Walmart: 'Sparky' AI Chatbot Is Lifting Spend, Engagement (eWeek); CocoLoop; Walmart's Sparky AI agent increases order value (Constellation Research)