Your Product Is on the Shelf. Sparky Still Might Not Recommend It.

Roughly half of all Walmart app users have now engaged with Sparky, according to Walmart U.S. President David Guggina on the company’s Q4 FY26 earnings call. Your product may be fully stocked, OTIF-compliant, and on the modular, and Sparky may still pass over it in favor of a competitor whose product data is more complete. That is the operational gap the NIQ data describes, and it is already open.

NielsenIQ’s New Growth Frontier report, released March 3 in collaboration with Kearney, provides the competitive backdrop. Niche brands gained 1.5 percentage points of U.S. market share between 2022 and 2025 while large and mid-size national brands declined by a combined 2.1 points. NIQ characterizes the pattern as structural, not cyclical. The driver is that AI is reorganizing which products get considered in the first place, and niche brands have moved faster to build the kind of structured, attribute-rich, need-state-legible product data that AI systems favor when assembling recommendations.

Seventy-four percent of shoppers now use AI in some form for product discovery, according to NIQ research cited in the report. Twenty percent use it directly for shopping. As Kearney partner Katherine Black noted in coverage of the report: “Brands that ensure their products are legible to AI with structured data, defined need states and credible signals are better positioned to surface in these new discovery pathways.”

At Walmart, that gap runs through Sparky.

Sparky Recommends What It Can Read

Walmart launched Sparky as a customer-facing AI shopping assistant in June 2025. The assistant helps shoppers search, compare items, synthesize reviews, and plan occasion-based purchases from within the Walmart app. According to Walmart CFO John David Rainey on the Q4 FY26 earnings call on February 19, customers who engage with Sparky have an average order value approximately 35% higher than those who do not. Guggina described the tool as helping Walmart evolve “from traditional search to intent-driven commerce.” Walmart also reports that 81% of shoppers who use Sparky do so to check product availability or research product details before purchasing, according to a Walmart survey cited in January 2026 reporting.

What does Sparky pull from when it builds a recommendation? Product titles, attributes, review content, and the structured data that populates a product detail page. The completeness and precision of that data determines whether Sparky describes your product confidently, misrepresents it, or passes over it in favor of a competitor with cleaner attributes. The categories where NIQ found niche brands gaining fastest (pet care, personal care, and health and wellness) are exactly the categories with high attribute specificity and strong need-state searching. A shopper asking Sparky about a protein supplement for recovery, a fragrance-free personal care item, or a grain-free pet food is generating a query that rewards brands with detailed, structured, accurately categorized product data.

Brands that have not audited their Supplier One item setup with AI-mediated discovery in mind have a concrete gap to close. Item Spec 5.0, which became mandatory for Marketplace sellers by August 31, 2025, replaced freeform inputs with structured fields and predefined enumerations. That shift rebuilt how Walmart’s systems parse and match product data. A listing adequate under Spec 4.x may now be functionally incomplete from the perspective of an AI system evaluating it for recommendation relevance, particularly in the ingredient, benefit, and use-case attribute fields that conversational queries most commonly invoke.

The Sponsored Prompt Is a New Competitive Surface, Not Just a New Ad Unit

In January 2026, Walmart confirmed it is actively testing Sponsored Prompt Ads within Sparky, following an initial pilot in fall 2025. The format works differently from traditional Sponsored Products. When a shopper asks Sparky a question in the product recommendation flow, a sponsored prompt can appear directly in the conversational response. Walmart Connect VP Khurrum Malik, announcing a broader AI strategy for the platform on January 6, framed the direction as entering “an era defined by AI” that will transform how brands reach customers and measure impact.

For Walmart Connect advertisers, this creates a competitive surface that does not behave like keyword bidding in Ad Center. Winning a sponsored prompt placement requires more than a bid; it requires a product whose structured data Sparky can synthesize into a coherent recommendation. The Marty advertising assistant (currently in beta for Walmart Connect Sponsored Search campaigns and expected to roll out more broadly in the first half of 2026) adds another dimension. Marty provides share-of-voice analysis, bid optimization recommendations, and campaign troubleshooting through a conversational interface. It is specifically designed to make Walmart Connect more accessible to brands with limited retail media experience, which mirrors NIQ’s finding that AI is lowering the capability barriers that previously favored large brands with deep agency and technology resources.

The implication for 1P brands managing existing Connect campaigns is that the advertising funnel at Walmart is bifurcating. Traditional Sponsored Search remains the volume channel. Conversational prompts within Sparky represent a higher-intent discovery layer that runs on different logic. Brands treating these as equivalent are likely to under-invest in the item content quality (structured attributes, reviewed content, high-quality images) that determines whether organic Sparky placement competes for the same high-value queries they are running in paid search. CEO John Furner acknowledged on the February 19 earnings call that the intersection of agentic commerce and advertising is still being navigated: “How this will work with agentic commerce, I think we’re all learning as we go.” That acknowledged uncertainty is, in the near term, an early-mover window for brands willing to get their item content right now.

Scintilla Digital Landscapes Now Tracks What Sparky Sees

The January 28, 2026 update to Scintilla Digital Landscapes added 15 new metrics covering traffic sources, conversion, digital basket behavior, and search. Suppliers can now see what keywords shoppers use to find their products, where those shoppers begin their journeys (external search, social, Walmart.com, or the Walmart app), and what drives purchase or abandonment at specific funnel stages. Linda Lomelino, group director of product management at Walmart Data Ventures, described the capability as putting pre-purchase behavior data alongside transactional and operational performance for the first time within a single view.

This is the diagnostic layer that connects the NIQ share shift to your specific business. If niche brands are gaining share in your category at Walmart, Digital Landscapes’ search metrics will show whether you are losing keyword share to them, and the conversion funnel data will show whether shoppers who find your product are converting at a rate consistent with your pricing and content quality score. Used together, those two signals close the loop between NIQ’s macro findings and your product-level reality.

The Shopper Behavior module adds a complementary angle. Tracking how loyalists, switchers, and trialists move through your category provides the need-state definition that Kearney’s Black identified as foundational to AI discoverability. If Shopper Behavior data reveals that a meaningful switcher cohort is entering your category from health and wellness or functional supplements, that is an item content and attribute signal, not only a promotional one. It tells you what problem statement your product detail page needs to address in order to surface in the queries that cohort generates through Sparky. Scintilla subscribers saw omni sales grow 15% faster than non-subscribers over the year ending September 2025, according to Walmart Data Ventures at its Inspire 2025 event, and the advantage that comes from knowing what shoppers are actually searching for is likely to widen as AI-mediated discovery captures a larger share of Walmart digital traffic.

Where the Same Shift Lands Differently

For large 1P brands, the NIQ findings are a warning about complacency, not an irreversible verdict. The report is clear that scale still matters; NIQ frames it as “scale remains powerful, but scale is no longer destiny.” The problem is that large brands have historically relied on distribution, trade promotion, and brand equity to maintain share, and all three of those levers operate downstream of discovery. If a shopper never reaches your shelf in an AI-mediated search path because your product attributes are underspecified or your content quality score falls below competitive thresholds in Supplier One, the trade and brand investment upstream of that gap is not pulling through to velocity or share.

The concrete first move for 1P suppliers is an item content audit anchored to Supplier One’s current attribute requirements, cross-referenced against Digital Landscapes search data showing which keywords are actually driving discovery in your category. If Sparky is fielding questions about your category and your product’s attributes do not map to the most common query terms, the gap is measurable and addressable before your next category review cycle or JBP conversation.

Marketplace sellers operating through WFS have a different starting position. Many of the niche and challenger brands gaining share in the NIQ data are precisely the type of sellers building Walmart presence through the 3P channel, with WFS providing the fulfillment speed signals that Walmart’s search algorithm uses as a ranking input. The Marty advertising assistant’s design to make Walmart Connect more accessible for smaller advertisers means these sellers now have an AI-assisted path to competitive Sponsored Search campaigns (and eventually sponsored prompt placement) that previously required significantly more internal capability or outside support. The opportunity is to combine WFS fulfillment with Item Spec 5.0-compliant listing quality and early Marty campaign optimization, building velocity and conversion metrics before larger incumbents fully re-platform their item content strategy.

Three Moves to Make Before Your Next Category Review

Pull Digital Landscapes data in Scintilla for your top items in any category where NIQ identified niche gains. Pet care, personal care, and health and wellness are the named ones, but the same logic applies wherever smaller, attribute-driven brands have been gaining velocity at Walmart. Look at keyword share and conversion rate together. If keyword share is holding but conversion is soft, the problem likely sits in item content quality or pricing. If keyword share is declining, the attribute and title structure warrants an audit against the queries Sparky would plausibly surface.

Run an Item Spec 5.0 compliance check in Supplier One for those same items. The migration deadline has passed, but a significant number of listings meet the minimum threshold without maximizing the structured fields that feed AI recommendation logic. Incomplete enumerations in ingredient, benefit, and use-case fields are the specific gaps that create recommendation blind spots in conversational AI, and they are fixable in Supplier One without waiting for a line review.

Request access to the Marty advertising assistant beta if you are currently running Sponsored Search campaigns through Walmart Connect Ad Center. The share-of-voice reporting Marty provides, combined with Digital Landscapes search data, creates a closed loop between organic discovery signals and paid amplification that was not available to most suppliers twelve months ago. Using that combination before your next Connect budget conversation produces a category-specific view of where paid investment closes actual discovery gaps versus where it reinforces positions you already hold organically. The brands with complete, well-structured item content when Sparky’s sponsored prompt format moves to scaled availability will be the ones appearing in those conversations.

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.