Walmart Says Sparky Users Are Up 70 percent: How Can Brands Optimise for Sparky?

Walmart CEO John Furner says the number of customers using Sparky is up 70% from last year. He also said that customers and members who use Sparky for shopping spend 40% more per order than those who do not.

The figures do not prove that Sparky itself causes customers to spend more. People who choose to use the assistant may already be more engaged or higher-intent shoppers.

But together, the numbers make one thing clear: Sparky is becoming a more important shopping surface. For brands, that raises a practical question: how can you optimise for a shopping assistant that recommends products based on what a customer is trying to buy?

The answer is unlikely to be a single Sparky ranking trick. Walmart has not published a formula for guaranteeing product recommendations. But brands can improve how clearly their products are understood and measure whether they are actually being surfaced and recommended.

How can brands optimise for Walmart Sparky?

Brands can optimise for Walmart Sparky by making product information clear enough for an AI shopping assistant to understand when a product matches a customer's request. Optimisation should also involve measuring how often products are actually appearing and being recommended, rather than assuming that a content change has produced an improvement.

A customer might search Walmart for a specific product name. When shopping through Sparky, they may instead describe a problem, occasion, preference or combination of requirements.

Someone might ask for a gift for a teenager, a compact product for travelling or an option suitable for a particular need. The information available about the product needs to make its relevant characteristics clear.

For brands, the starting point is therefore to reduce ambiguity and measure what happens.

What does Sparky need to understand about a product?

Sparky needs enough clear information to connect a product with the needs a customer describes. The exact information will depend on the category, but brands should generally consider whether the information surrounding a product clearly explains:

• what the product is
• what it does
• key features and specifications
• size and quantity
• ingredients or materials
• compatibility
• who the product is for
• when it is useful
• important limitations
• differences between variants

The objective is not to create the longest possible product listing. It is to make the information that determines relevance easier to understand.

A product description can contain hundreds of words and still leave the most important questions unanswered.

How does Sparky change the way customers discover products?

Sparky changes product discovery because customers can start with a need rather than a product name. Instead of typing a category into a search bar, a shopper can ask for help finding something for a particular situation or explain what they want to achieve.

That creates a broader set of possible starting points for product discovery.

A shopper might describe:

• a problem they need to solve
• an occasion
• a target customer
• a budget
• a preference
• a restriction
• several requirements at once

The product therefore needs to be understandable beyond its exact category or name.

Brands should consider whether a customer could understand why the product is relevant even if they never searched for it by name.

How should brands optimise product information for Sparky?

Product information should clearly explain what a product is and connect relevant characteristics with what they mean for the customer. Marketing language can still have a place, but it should not replace the basic information required to understand the product.

For example, consider the difference between saying that a portable blender is "perfect for life on the move" and explaining that it is compact, rechargeable and intended for making single-serve drinks away from home.

The second description provides clearer information about why the product may be relevant to a particular need.

The goal is not to add every possible use case to every product. Brands should only describe genuine product characteristics and legitimate uses.

But they should not assume that a customer or shopping assistant will automatically infer those uses from vague marketing language.

Why are product attributes important for Sparky?

Product attributes help distinguish one relevant product from another. Size, material, ingredients, compatibility, colour, quantity and technical specifications can all determine whether a product matches a customer's request.

Someone asking for a coffee maker is expressing a broad category need. Someone asking for a compact coffee maker for a small kitchen is describing a narrower set of requirements.

If the product's relevant dimensions or characteristics are unclear, there is less information available to establish whether it is a good match.

That does not mean every attribute determines whether Sparky recommends a product. Walmart has not disclosed a complete recommendation formula.

But complete and consistent product information gives both customers and shopping systems a clearer understanding of where a product fits.

How can brands make products more relevant to conversational shopping queries?

Brands should think about the questions customers ask when choosing a product, not just the keywords they type into a search bar. Conversational shopping queries often contain context that traditional product search terms do not.

A customer might ask for something suitable for:

• travelling
• gifting
• exercising
• a particular room or environment
• a specific age group
• a dietary or material preference
• replacing another product

The important question is whether the relevant information is genuinely available.

If a product is suitable for a particular use, the information surrounding it should explain why. If it is not suitable, that limitation can also be valuable information.

Trying to appear relevant to every possible request is unlikely to help. Clear information about where a product does and does not fit is more useful than broad claims with little supporting detail.

What makes a product easier for Sparky to understand?

Clear, specific and consistent information makes it easier to establish what a product is and how it differs from alternatives. The information should answer the questions that determine whether the item is a match.

The exact framework should vary by category.

A beauty product, grocery item and electronic device all require different information. The principle is that the information should not depend on a customer or system guessing the details that determine whether the product is suitable.

How can brands optimise thousands of products for Sparky?

The biggest challenge is usually not knowing what information a product needs. It is applying a consistent standard across a large catalogue.

Large ecommerce catalogues often contain information created at different times, for different marketplaces and by different teams. One product may have complete attributes and clear descriptions, while another has an outdated listing or missing information.

That creates inconsistency across the catalogue.

Brands need a repeatable way to identify what information each category requires, establish standards for how that information is expressed and monitor whether the resulting products are actually appearing when relevant shopping questions are asked.

Azoma is an agentic commerce optimisation platform. Azoma helps brands measure how products are discovered and recommended across AI-powered shopping experiences, including surfaces such as Walmart Sparky, and identify where optimisation opportunities exist.

For large brands, the challenge is not simply making one product easier for an AI assistant to understand. It is understanding where thousands of products are and are not being surfaced, then identifying what can be improved.

How does Azoma help brands optimise for Walmart Sparky?

Azoma helps brands move from assumptions about AI shopping visibility to measurement. Product information can be improved, but without tracking relevant shopping questions, brands may not know whether those changes are affecting how often products are surfaced or recommended.

Azoma enables brands to monitor how products and brands appear across AI-powered shopping experiences. That gives teams a way to identify gaps, understand where competitors are being recommended instead and prioritise optimisation work.

Agentic commerce optimisation overlaps heavily with what's variously called GEO, or generative engine optimisation, AEO, or answer engine optimisation, and AI search optimisation. The distinctions are mostly emphasis: GEO names the technology producing the answer, AEO names the behaviour being optimised for, and agentic commerce optimisation names the case where the answer ends in a purchase.

For Sparky, that distinction matters because the optimisation target is not simply whether a brand appears in an answer. The commercial question is whether the right products are being understood and recommended when a shopper is trying to buy.

What should brands audit before optimising for Sparky?

Brands should begin with the products where conversational discovery is most likely to matter. There is little value in changing an entire catalogue before identifying where the biggest gaps are.

A useful audit should ask:

1. Is the product type immediately clear?
2. Are key attributes complete?
3. Does the information explain what the product does?
4. Are genuine use cases clear?
5. Are important limitations or compatibility details available?
6. Can customers distinguish between variants?
7. Is relevant information consistent across channels?
8. Is the product appearing when relevant shopping questions are asked?

The first seven questions examine the information available about the product.

The final question examines the outcome. Both matter.

Can better product information guarantee a Sparky recommendation?

No. Better product information cannot guarantee that Sparky will recommend a particular product. Walmart has not published a complete formula for how Sparky selects products in every shopping interaction.

Brands should therefore be cautious of anyone claiming that a particular keyword, title format or information update guarantees visibility.

The more defensible argument is that clearer and more complete information gives an AI shopping assistant a stronger basis for understanding what a product is and when it may be relevant.

That is an important distinction.

Optimising for Sparky means improving the information available about a product and measuring the outcome. It does not mean controlling Sparky's final recommendation.

What should brands do next?

Brands should treat Sparky's growth as a reason to audit both product information and product visibility. The first step is to identify the products where broader, conversational discovery could matter most.

The second is to check whether the information surrounding those products clearly explains what they are, what they do, who they are for and when they are relevant.

The third is to establish a way to measure whether those products are actually being surfaced and recommended when relevant questions are asked.

That turns Sparky optimisation from a one-time content exercise into an ongoing process of measurement, improvement and validation.

What Walmart's Sparky growth means for brands

Sparky's 70% year-on-year user growth means more customers are beginning their shopping journeys through an AI assistant. The reported 40% higher order value among Sparky users does not establish that the assistant caused those customers to spend more, but it does indicate that this is a commercially significant group of shoppers.

That makes product understanding and recommendation visibility increasingly important.

Brands have spent years learning how to win visibility on a Walmart page. AI shopping introduces a different question: can the system understand what your product is and when it is relevant before the customer reaches that page?

There may not yet be a single answer to how Sparky decides what to recommend.

But brands can measure where they are appearing, identify where products are absent and improve the information that may help reduce ambiguity.

Final thoughts

Walmart's latest Sparky figures suggest that AI-assisted shopping is becoming an increasingly important part of the ecommerce journey.

Brands should not respond by chasing a supposed secret Sparky ranking factor. Walmart has not published one.

The more practical response is to make sure product information answers four basic questions:

What is it? What does it do? Who is it for? When is it the right choice?

But understanding a product is only one part of the process.

Brands also need to know whether their products are actually being surfaced and recommended. Azoma is an agentic commerce optimisation platform. Azoma helps brands measure those outcomes across AI-powered shopping experiences and identify where optimisation opportunities exist.

For Sparky, as with other AI shopping experiences, the goal is not to control every recommendation. It is to understand where products are visible, where they are missing and what information may need to change to give the system a clearer basis for deciding when they are relevant.

‍