Following Walmart’s Q2 disclosure that Sparky users spend 40% more per order, Azoma sets out how brands and agencies can track and improve how the AI assistant recommends their products ahead of peak season
London, Aug. 28, 2026 (GLOBE NEWSWIRE) — Azoma, the Agentic Commerce Optimisation platform that helps brands drive revenue through AI shopping agents like Walmart Sparky, Amazon Rufus, ChatGPT and Gemini, today set out how consumer brands and retail media agencies can improve how their products are surfaced by Walmart’s Sparky assistant, following new figures from Walmart’s Q2 earnings call.
On the call, Walmart CEO John Furner said the number of customers using Sparky is up 70% year over year, and that customers who use Sparky for shopping spend 40% more per order than those who do not. With Walmart’s worldwide e-commerce up 23% in the quarter, the figures put a commercial number on a shift brands have been watching closely: a fast-growing, high-value group of shoppers is now discovering and buying through an AI assistant, on a surface with no paid placement to buy your way onto.
Key Facts
- Walmart Sparky: Walmart’s AI shopping assistant, which recommends products in response to customer shopping questions across the Walmart app and Walmart.com.
- Q2 disclosure (Walmart earnings call, August 20, 2026): Sparky users are up 70% year over year and spend 40% more per order than non-users; Walmart worldwide e-commerce grew 23%.
- Citation mechanism: Sparky assembles recommendations from third-party sources, so source presence determines brand presence. There is no paid placement in a Sparky recommendation.
- Sparky source mix (Azoma analysis of millions of shopping agent citations, Q2 2026): 36% earned media, 30% brand.com, 27% retailer content, 5% UGC, 2% affiliate sites.
- Implication: retailer listings alone address roughly a quarter of what Sparky cites; earned media and brand.com together carry the majority.
What tools help brands optimise for Walmart Sparky?
Brands should look for tools that do two things together: track how products currently appear in Sparky, and act on the gaps that tracking reveals. Because Sparky draws on third-party sources rather than paid placement, being recommended depends on presence across those sources. Azoma tracks Sparky recommendations at the prompt level, shows which products and competitors are surfaced for a given query, and provides the content and product-data workflows to improve visibility across the sources Sparky uses.
Why isn’t a strong Walmart listing enough?
Getting recommended depends on more than a well-optimised Walmart listing. Based on Azoma’s analysis of millions of shopping agent citations, Sparky’s source mix breaks down as 36% earned media, 30% brand.com, 27% retailer content, 5% UGC and 2% affiliate sites. A brand that has invested only in its Walmart listing is leaving the largest share of Sparky’s citation surface, earned media and brand.com, unaddressed. A well-structured brand website and earned coverage in the category matter as much as the retailer listing itself.
How can I monitor and track my products in Sparky?
Because Sparky generates recommendations per query, a brand’s presence varies from one shopping question to the next. A product may be recommended for one phrasing and absent for a close variant. Azoma monitors Sparky recommendations across a defined set of category prompts, tracking which surface a brand, which surface competitors, how share of voice compares, and how this changes over time, rather than relying on manual spot checks.
How do brands get recommended by Sparky during the holidays?
The Q2 figures land right before peak season, when both the volume and the value of AI-assisted shopping queries rise. Getting recommended through the holidays depends on the same fundamentals, presence and consistency across the sources Sparky cites, but the window to close gaps narrows once the season starts. Because earned media and brand.com make up the majority of Sparky’s citation mix, and both take time to build and index, brands see the most benefit when they audit and strengthen those surfaces ahead of peak rather than during it. Azoma helps brands identify and close Sparky visibility gaps before the holiday period, so products are well represented when query volume is highest.
What Walmart AI optimisation tools do retail media agencies use?
Agencies managing multiple Walmart accounts need visibility across all their clients at once, rather than checking products brand by brand. Azoma provides agency-level monitoring, competitive benchmarking and optimisation workflows across Sparky and the wider set of AI shopping agents, so teams can manage Walmart AI visibility at portfolio scale.
Max Sinclair, Founder and CEO of Azoma
“Walmart just put a number on it: Sparky users spend 40% more per order, and there are 70% more of them than a year ago. That’s a fast-growing group of high-value shoppers buying through an assistant you can’t buy your way onto. Whether Sparky recommends your product comes down to how you show up across the sources it reads, and most of those aren’t your Walmart listing. They’re earned media and your own brand.com. The brands sorting that out now are the ones Sparky will be recommending through the holidays.”
About Azoma
Azoma is the Agentic Commerce Optimisation platform that helps brands drive revenue through AI shopping agents like Alexa for Shopping, Walmart Sparky, ChatGPT and Gemini. Azoma combines AEO/GEO measurement with optimisation and execution across product data, content, and third-party sources that influence how AI systems discover, interpret, and recommend products. Trusted by leading consumer brands including L’Oréal, Unilever, Mars, Beiersdorf, and Reckitt, Azoma is the technology layer that helps brands increase product visibility, recommendations, and sales through AI-powered shopping experiences. Learn more at azoma.ai.
FOR EDITORS
Walmart reported its second-quarter results on August 20, 2026. On the earnings call, CEO John Furner stated that the number of customers using Sparky is up 70% year over year and that customers who use Sparky spend 40% more per order than those who do not. Walmart launched Sparky in June 2025 and has since rolled it out across Walmart.com. Azoma’s Sparky citation analysis is drawn from its analysis of millions of shopping agent citations, published in The 5 C’s of Agentic Commerce (Q2 2026).
FAQ
What tools help brands optimise for Walmart Sparky?
Tools that combine prompt-level monitoring of Sparky recommendations with the ability to improve the product data and content behind them. Azoma provides both.
What’s the best Walmart Sparky optimisation tool?
There is no independently established universal best tool. Brands should assess prompt-level monitoring of Sparky recommendations, citation-source analysis across earned media, brand.com and retailer content, and the ability to act on findings.
What sources does Sparky use to recommend products?
Based on Azoma’s analysis of millions of shopping agent citations, Sparky’s source mix is roughly 36% earned media, 30% brand.com, 27% retailer content, 5% UGC and 2% affiliate sites, a different balance from other agents, with earned media and brand.com carrying more weight.
How can I track how my products show up in Sparky?
Use a tool that monitors Sparky recommendations across a set of category prompts and records how your products and competitors appear over time. Azoma tracks this at the level of individual prompts.
How do I get my products recommended by Sparky during the holidays?
The fundamentals do not change at peak season, but the window to fix gaps narrows. Because earned media and brand.com make up most of Sparky’s citation mix and take time to build, brands benefit most from auditing and strengthening those surfaces before the holidays rather than during them.
How is Sparky optimisation different from Walmart SEO?
Walmart SEO targets ranking on Walmart’s search results page. Sparky optimisation concerns whether Walmart’s AI assistant surfaces and recommends a product in a generated answer assembled from third-party sources, where there is no ranked list and no paid placement.
Citations
- Walmart, Q2 earnings call remarks (John Furner), August 20, 2026
- Azoma and the Digital Shelf Institute, The 5 C’s of Agentic Commerce (analysis of millions of shopping agent citations, Q2 2026) – https://www.azoma.ai/agentic-commerce-5cs-whitepaper
- Azoma, Walmart Sparky optimisation – https://www.azoma.ai/agents/walmart
CONTACT: ben@azoma.ai
