Walmart Built a Bridge. Amazon Built a Moat: What Agentic Commerce Means for Every Brand Selling Through Walmart
The Decision That Changes Everything for Walmart Suppliers
For years, more than six in ten US consumers started their online product searches on Amazon, according to NIQ research. That single fact shaped supplier strategy across the industry: Amazon was the front door of commerce, and brands paid heavily to be visible behind it. Retail media spending at Amazon, sponsored products, sponsored brands, sponsored display, became less an option than a cost of doing business for any brand that wanted to be found.
Agentic AI disrupts that starting point. When consumers delegate product research and purchase to an AI assistant, the question of where search begins becomes dramatically more open. Amazon has responded to this threat by blocking 47 AI bots from crawling its site, according to updates in its publicly visible robots.txt file as reported by Forbes in February 2026, including crawlers from OpenAI, Google, Meta, Anthropic, and Perplexity. The logic is straightforward: Amazon’s advertising business generated $56.2 billion in revenue in 2024, per Amazon’s annual earnings report. Every transaction that begins outside Amazon’s platform represents a potential bypass of that toll booth.
Walmart made the opposite bet. Rather than blocking horizontal agents, Walmart partnered with OpenAI in October 2025 to allow customers to shop Walmart directly inside ChatGPT, as announced by both companies at the time. It joined Google’s Universal Commerce Protocol at NRF 2026, co-developed with Shopify, Etsy, Wayfair, and Target, per Google’s January 2026 announcement. It has not blocked a single AI crawler. Walmart CEO Doug McMillon has framed the company’s position plainly: the era of a search bar and a long list of results is ending, and Walmart intends to be on the right side of what replaces it.
For suppliers, this is not background context. It is the strategic environment in which every product listing, every retail media investment, and every data infrastructure decision will play out over the next three years.
Sparky Is Not a Chatbot. It Is the New Shelf.
Walmart’s Sparky launched across all product categories in June 2025. It is described internally as a super agent: not a simple conversational interface but an agentic system trained on retail-specific data that can discover products, synthesize reviews, compare options, and, in its current trajectory, complete transactions autonomously. The smiling icon in the Walmart app is where an increasing share of product consideration begins for Walmart shoppers.
The numbers behind early adoption are notable. An internal Walmart survey conducted in August 2025 found that 81 percent of shoppers had used Sparky to check product availability or review specifications before making a purchase. That figure represents a consumer already in a high-intent moment, actively researching a purchase decision, interacting with a system that surfaces products algorithmically rather than visually. What appears in that conversation depends almost entirely on product data quality, review depth, and structured content, not packaging, shelf positioning, or the visual hierarchy of a product detail page.
Walmart is also actively testing advertising inside Sparky. Beginning with a pilot in fall 2025 and moving to formal testing announced in January 2026, sponsored prompts appear inside Sparky’s conversational responses when shoppers ask for product recommendations. A shopper asking about protein powder for recovery might receive an organic recommendation alongside a sponsored prompt integrated naturally into the conversation, appearing at the precise moment the shopper is deciding. This is a fundamentally different advertising environment from sponsored search, and it requires brands to think differently about what makes a product recommendation compelling to an AI model evaluating on a consumer’s behalf.
Visibility inside an AI conversation is earned differently than visibility in a search result. The optimization levers are product data, review quality, and relevance signals, not just bid prices.
Marty, Walmart’s supplier- and advertiser-facing AI agent introduced in July 2025, completes the picture. Where Sparky faces the consumer, Marty faces the brand. Currently in beta for sponsored search campaigns and rolling out to all brands buying search ads in the first half of 2026, Marty provides conversational guidance on bidding, keywords, billing, and campaign optimization. For smaller suppliers with limited retail media expertise, this meaningfully lowers the barrier to running effective Walmart Connect campaigns. For larger brands, it signals that Walmart expects the advertiser-platform relationship to become increasingly AI-mediated on both sides of the transaction.
Walmart Connect Is Growing Faster Than Amazon Ads. Brands Should Know Why.
Walmart Connect generated $6.4 billion in advertising revenue in 2025, up 37 percent globally and 41 percent in the US year-over-year, according to Walmart’s full-year earnings report published February 2026. That growth rate significantly outpaces Amazon’s advertising segment, which grew approximately 20 percent in 2024. That outperformance is not accidental. It reflects Walmart’s scale of first-party retail data, its expanding omnichannel measurement capabilities, and its willingness to integrate advertising into new surfaces including Sparky and connected television.
The Vizio acquisition, completed for $2.3 billion in December 2024 per Walmart’s official acquisition announcement, extends Walmart’s advertising reach into connected TV in a way that Amazon does not currently replicate at comparable scale. For CPG brands in particular, the ability to connect retail media investment to actual purchase behavior across both digital and physical channels, with measurement that includes in-store transactions at more than 4,700 locations, creates a data asset that is genuinely difficult for platform-only players to match.
EMarketer projects US retail media ad spending at $69.33 billion in 2026, rising to $129.93 billion by 2028. That trajectory assumes continued growth in retail media as a channel. What it does not fully account for is the structural shift in where product discovery begins as horizontal agents gain adoption. For Amazon, that shift represents a threat to search-originated advertising revenue. For Walmart, which has deliberately positioned itself as accessible to those same agents, the dynamic is different: agentic discovery can route traffic toward Walmart rather than away from it, provided brands’ products are accurately and completely represented in the data those agents can access.
This is the crux of what Walmart’s open strategy means for brands in practice. Being findable through Sparky, through ChatGPT’s Walmart integration, and through any agent that connects via the Universal Commerce Protocol is not automatic. It depends on whether Walmart’s product catalog contains the right information in the right structure to be surfaced accurately by an AI making a recommendation on a consumer’s behalf.
The Fulfillment Advantage That Horizontal Agents Cannot Replicate
Horizontal AI agents have a genuine and growing advantage in product discovery: they can search across retailers without the bias that a retailer-owned agent inevitably carries toward its own catalog. When a consumer asks ChatGPT for the best laundry detergent for sensitive skin under twenty dollars, ChatGPT has no structural incentive to favor any retailer’s products. That objectivity is part of what makes horizontal agents compelling to consumers.
What horizontal agents cannot do is deliver the product. Amazon can deliver to most US addresses in hours. Walmart can offer same-day fulfillment from more than 4,700 stores covering a significant share of the US population. When an agent recommends a product and the consumer wants it today, the fulfillment question does not disappear. Walmart’s physical footprint is a structural advantage in that moment, and it is one that no horizontal platform can build quickly.
For suppliers, this is worth internalizing: the categories most likely to shift to agentic commerce first are exactly the replenishment and commodity categories where same-day availability matters most. Household essentials, grocery staples, personal care basics. These are the purchases where a consumer is most willing to delegate, and where Walmart’s store network creates a fulfillment capability that reinforces rather than competes with the agentic discovery model Walmart is building.
What This Requires From Brands Right Now
The strategic implications of Walmart’s agentic commerce posture resolve into a small number of concrete decisions that brand teams and supplier leadership need to make, most of which do not require waiting for the market to mature further.
The first is product data. Bain and Company’s November 2025 research on agentic retail is explicit: retailers and brands whose product catalogs are not structured for machine readability risk becoming invisible in agent-mediated shopping journeys regardless of brand recognition. For Walmart suppliers, this means auditing how products appear in Walmart’s catalog from the perspective of an AI agent reading that data cold: whether attribute fields are complete, whether descriptions are accurate and specific, whether review volume and quality are sufficient to support a confident recommendation. This is not a marketing question. It is an operations and supply chain question that belongs in the same conversation as packaging and pricing.
The second is retail media strategy. Walmart Connect’s advertising inside Sparky represents a new surface that functions differently from sponsored search. A sponsored prompt inside a conversational AI response earns its place through relevance to the consumer’s stated need, not simply through bid price. Brands that understand what signals Sparky uses to surface recommendations, and that optimize their product data and advertising content accordingly, will be better positioned than those treating Sparky as simply another sponsored placement channel.
The third is protocol awareness. Walmart is a co-developer of the Universal Commerce Protocol, the open standard for search-to-buy agentic transactions across the open web. Brands selling through Walmart should understand what UCP integration means for their product surfaceability through horizontal agents that connect to Walmart’s catalog. The question of whether your products are reachable through agent-mediated commerce is not abstract. It is increasingly answerable by checking whether Walmart’s UCP implementation covers your categories and whether your product data meets the requirements for accurate surfacing.
The fourth, and most immediate, is recognizing that Marty makes Walmart Connect more accessible. If your brand has avoided Walmart Connect advertising because the platform required expertise you did not have in-house, the Marty assistant substantially changes that calculus. The conversational interface for campaign management, currently in beta and rolling out to all sponsored search advertisers in the first half of 2026, is designed explicitly to reduce the expertise barrier. That is an opening for smaller and mid-sized suppliers that has not existed before.
The Brands Best Positioned Are Not Necessarily the Largest
One of the more counterintuitive implications of Walmart’s agentic strategy is that it may favor a different kind of brand than traditional retail media has. Sponsored search on Amazon rewards brands with the budget to bid competitively and the scale to sustain that investment continuously. The result, over time, has been a consolidation of visibility around the largest advertisers.
Agentic discovery through Sparky rewards a different set of attributes: accurate product data, genuine consumer reviews, clear and specific product claims, and relevance to the consumer’s stated need. A brand with strong review quality, complete attribute data, and a clearly differentiated product can surface in a Sparky recommendation above a larger competitor whose catalog entry is incomplete or whose review profile is thin. That is a different competitive dynamic, and it is one where mid-sized and emerging suppliers can meaningfully compete.
NIQ’s 2025 research on private label and national brand competition found that national brands carry a 26 percent average price premium over private label alternatives globally, and that 53 percent of consumers report buying more private label than before. An AI agent evaluating product options on a consumer’s behalf does not experience brand familiarity the way a human browsing a shelf does. That premium must be justified in the data: in product attributes, in review signals, in the specificity of claims. For brands that have genuine product differentiation and can articulate it clearly in structured data, this is an advantage. For brands whose premium rests primarily on packaging and advertising recall, it is a challenge worth planning for now.
Walmart made a deliberate strategic choice to open its platform to the agents that are reshaping how consumers discover and purchase products. The brands that benefit from that choice are the ones that understand what openness actually requires: product data that machines can read accurately, retail media investment in surfaces that are growing fastest, and operations flexible enough to serve the fulfillment expectations that agentic commerce will create. The window for getting ahead of this is not permanently open. But it is open now.
Winning With Walmart
Winning With Walmart is an independent platform for suppliers, sellers, solution providers, and industry experts.
Articles are developed by staff editors, contributing experts, and trusted partners.
Some are researched and drafted with the assistance of AI tools and are reviewed, fact-checked, and edited by Winning With Walmart editors before publication.
Editorial judgment, sourcing decisions, and final approval rest with the publication's human editors in every case.
We are committed to accuracy and fairness. If you believe this article contains an error, we welcome your feedback.