Affect Performance Team

|AI|Nov 11, 2026

VTEX Expands Shopping Assistant for AI Product Discovery and Order Tracking

In November 2026, VTEX introduced an update to Shopping Assistant, an AI assistant that helps shoppers find products, receive personalized recommendations, and check their order status directly in an online store's interface.

Shopping Assistant is part of the VTEX CX Platform and supports natural-language conversations. Shoppers do not need to configure search filters or navigate help pages on their own. They can describe what they need and get an answer in chat.

An important part of the update is the native integration with FastStore, VTEX's framework for building online stores. It lets merchants add the assistant to their storefront without developing a separate integration from scratch.

Product Search Through an AI Conversation

Shopping Assistant offers an alternative way to navigate an online store's catalog. Instead of searching for an exact product name, shoppers can describe a task or the specifications they need.

For example, a shopper might say: “I need running shoes for training on pavement. My budget is under $150.” The assistant uses the store's available data to identify suitable products.

Another example is: “Show me compact coffee makers for a small kitchen.”

This approach is particularly useful for catalogs with numerous product categories and attributes. Shoppers do not have to know the exact model or product category. They can explain what they want to buy.

However, the quality of the results depends on catalog structure, the completeness of product descriptions, and the availability of product information. If specifications are missing or incorrect, the assistant may suggest an irrelevant product. Shopping Assistant does not replace catalog management or fix the underlying quality of product data.

Personalized Product Recommendations

Another capability of Shopping Assistant is making recommendations based on needs expressed during a conversation.

For example, an electronics shopper might ask: “Which laptop is best for working with large Excel files and video conferences?” The assistant can identify products that meet the stated requirements.

The shopper can then refine the request: “Show me lighter models,” or “Which of these laptops have screens of at least 15 inches?”

This conversation gradually narrows the range of products under consideration. For online stores, it offers an additional product discovery tool, particularly in categories where customers compare many specifications.

Here, personalization means recommendations based on the available context and the shopper's expressed needs. The official November announcement does not explain in detail which historical customer data is used to rank recommendations for individuals.

It would therefore be incorrect to equate Shopping Assistant with a full recommendation engine based on purchase history, customer lifetime value, and predictive models.

Helping Shoppers After Checkout

Shopping Assistant is not limited to product selection. VTEX also provides the ability to answer questions about orders and deliveries.

A shopper might ask: “Where is my order now?” or “When will it arrive?” The assistant uses available order information to provide an answer in chat.

This could reduce the number of routine inquiries to customer support about order status. For merchants processing large daily order volumes, these questions can account for a meaningful share of support work.

However, the November announcement does not confirm that Shopping Assistant can independently cancel orders, issue refunds, or change delivery terms. These capabilities should not be attributed to the product without separate confirmation. Providing personal order information also requires appropriate shopper identification and data-access controls.

Native FastStore Integration

In October 2026, VTEX separately announced the availability of Shopping Assistant in FastStore, a framework for building high-performance storefronts on the VTEX Commerce Platform.

The native integration connects Shopping Assistant to a store through existing FastStore components and the VTEX CX Platform. This reduces development work compared with building an AI chat interface and integrations with the catalog and order systems independently.

VTEX technical documentation sets out the requirements and configuration process.

How to Connect Shopping Assistant

Native integration requires FastStore version 3.98.0 or later. The main steps are:

Update FastStore to a compatible version.
Synchronize the ShoppingAssistant component with the CMS schema.
Open VTEX Admin.
Navigate to Storefront → Agentic CX → Dashboard.
Start the Shopping Assistant setup.
Connect the online store's data.
Activate the assistant.
Add the Shopping Assistant component to CMS Global Sections.
Save the changes and test how the assistant appears on the storefront.

A separate synchronization option is available for stores using the older Headless CMS. Shopping Assistant can be displayed across store pages through global FastStore components.

The phrase “without additional development” refers to the ready-made integration. Technical work may still be required to configure compatibility, customize presentation, assess data quality, and test the assistant's behavior.

Availability and Pricing

The November Shopping Assistant update is labeled Early Access. VTEX lists these regions:

•United States.
•Brazil.
•Latin America.
•Europe, Middle East, and Africa.
•Asia-Pacific.

A region's inclusion in Early Access does not guarantee automatic activation for every merchant. Activation requirements should be confirmed with VTEX.

The FastStore integration is offered without a separate integration fee, but use of Shopping Assistant itself is billed under the VTEX CX Platform pricing model. A free integration does not mean the AI assistant is free to operate.

The November update page does not provide specific prices, conversation limits, or usage-based billing details.

VTEX also introduced initial AI-agent onboarding features in its CX Platform in April 2026. The November update should not be described as the first release of AI-powered customer assistance in the VTEX ecosystem.

Practical Value for E-commerce

Shopping Assistant is relevant to e-commerce beyond basic product search. A conversational interface can help shoppers who have difficulty identifying the right product, understanding technical attributes, or finding complementary items.

In a conventional store, a shopper uses product categories, search, filters, and product detail pages. To get relevant results, they need some understanding of which category to choose and which specifications matter.

An AI assistant lets them start with a description of their needs and refine their requirements through conversation. This creates different commercial use cases depending on the product category.

Fashion: Finding Similar Items in the Right Color and Size

In fashion e-commerce, Shopping Assistant could help shoppers find alternatives to products they like. Suppose a customer views a dress but their size is out of stock. Rather than returning to the general catalog, they could ask: “Find dresses in a similar style, but in size M and dark green.”

The assistant could use the selected product's attributes and available catalog data to find alternatives.

Another example: “I like this jacket, but I want something similar in beige linen for under $120.”

To work reliably, the catalog must provide structured information about size, color, material, silhouette, and availability. This can be useful for fashion stores with broad assortments where shoppers would otherwise browse dozens of similar product pages.

A related use case is putting together an outfit: “Find pants and shoes to go with this jacket. I need an outfit for the office.” The assistant could serve as a product consultant, moving a shopper from one selected item to a group of complementary products.

The potential commercial value lies in retaining customers when an exact size or color is unavailable, improving product discovery, and potentially increasing average order value through outfit purchases.

The November announcement does not confirm visual similarity search based on product images. These examples assume that the assistant uses the selected product's context and structured catalog attributes, not independent visual recognition.

Electronics: Finding Complex Products by Specifications

Electronics stores offer an especially relevant setting for conversational product guidance. Shoppers often understand what a device needs to accomplish but do not know which technical specifications to compare. Other customers are familiar with the technology but want combinations of parameters that are awkward to specify with standard filters.

Smartphones. A shopper might ask: “I need a smartphone with eSIM, NFC, a good video camera, at least 256 GB of storage, and a battery that lasts all day.” The assistant can narrow the catalog based on those requirements, provided the information is available in product data.

Laptops. A buyer could explain: “I need a lightweight laptop for business trips, working with large spreadsheets, and joining video calls.” A follow-up might be: “Only show models with at least 16 GB of RAM and a 15-inch display.” This makes it easier to refine the choice through several messages rather than rebuilding search filters.

Televisions. A buyer could request a TV for gaming that supports a 120 Hz refresh rate and HDMI 2.1, then narrow the selection by screen size and budget.

Home appliances. A customer might look for a quiet dishwasher with an automatic door-opening feature after the cycle ends. The assistant can help find products based on specific technologies and functions, not only broad categories.

This approach may be particularly useful for computers, televisions, projectors, cameras, audio equipment, appliances, and networking devices. Shoppers can refine their requirements as they learn about the available products.

For electronics merchants, the assistant could reduce cases in which shoppers leave the site to research basic technical specifications elsewhere.

Another opportunity is finding compatible accessories. After choosing a TV, a shopper could request a wall mount. After choosing a laptop, they might ask about a charger, docking station, or monitor. After selecting a camera, they might look for a compatible lens, memory card, or tripod.

These use cases can contribute to average order value, but compatibility data must be accurate. Incorrect accessory recommendations can generate returns and additional service costs.

Sporting Goods: Product Guidance for Beginners

Shopping Assistant could be particularly useful for people starting a new sport. Beginners may not know which equipment they need, how models differ, or which features matter.

For example, a shopper says: “I'm starting to run three times a week. I plan to run 5 km on pavement. Which shoes should I choose?” The assistant can narrow the selection using the intended use, shoe attributes, and budget.

A first-time bicycle buyer might ask: “I want a bike for riding around the city and in parks. I haven't cycled regularly before. What should I consider?” The assistant can explain differences between city, hybrid, and mountain bikes and identify relevant specifications. The shopper can then add: “Show suitable bikes under $700.”

The commercial opportunity also includes the accessories a beginner might need:

•Bicycle helmet.
•Lock.
•Front and rear lights.
•Pump.
•Spare inner tube or repair kit.
•Multitool.
•Water bottle and holder.
•Appropriate riding apparel.

After the bicycle is selected, Shopping Assistant could help identify complementary items from the store's catalog.

Another example is a first overnight hiking trip: “I'm going on a two-day summer hike with an overnight stay. I don't have any gear yet. What will I need?” The assistant could help shoppers navigate backpacks, tents, sleeping bags, sleeping pads, and other equipment.

Similar use cases apply to fitness, tennis, swimming, skiing, and other categories. A beginner tennis player may need help understanding the choice of racket, shoes, and essential accessories.

For sporting-goods stores, these scenarios could increase items per order and average order value by helping customers assemble a coherent equipment set. However, the assistant should distinguish essential items from recommended and optional extras. It should not substitute unverified medical-safety or injury-prevention claims for professional guidance.

Home Improvement and DIY: Finding Materials, Supplies, and Tools

Home-improvement stores may benefit from helping shoppers plan a complete DIY project. Customers often know what they want to accomplish but not the full list of materials, consumables, and tools required.

For example: “I want to paint the walls in a 15-square-meter bedroom myself. What do I need?” The assistant could help shoppers assemble a relevant set of products, potentially including:

•Primer.
•Filler, when surface preparation is needed.
•Paint rollers.
•Brushes.
•Painter's tape.
•Protective sheeting.
•Paint tray.
•Sanding materials.
•Gloves and other protective equipment.

Another shopper might ask: “I want to install laminate flooring in a 20-square-meter room myself. Which materials and tools do I need?” The assistant could help with flooring, underlayment, baseboards, connecting materials, and installation tools.

Related applications include tiling, installing shelving, lighting, gardening, and general home projects. For example: “I need to hang several wall shelves on a concrete wall. Which anchors and tools should I use?” The assistant could identify relevant product categories based on wall material and stated requirements, but should not make unsupported claims about safe load capacity or installation safety.

The commercial opportunity is to shift from selling individual products toward assembling a set of materials for a defined project. A customer buying paint can also find the supplies needed to paint the room. Someone buying flooring can discover compatible installation materials and tools.

This may increase items per order and average order value while reducing the chance that customers need to place an additional order after discovering that materials are missing.

Success depends on product compatibility, accurate material-quantity calculations, and reliable technical catalog attributes.

Shopping Assistant as an Average Order Value Tool

One of the more interesting commercial scenarios is cross-selling based on the shopper's expressed goal.

In a typical online store, complementary product recommendations are often driven by predefined rules, product relationships, or historical co-purchase data. A conversational assistant adds the ability to use the shopper's stated purpose.

A request such as “I want to start cycling” implies potential needs beyond the bicycle itself. “I'm going to paint a room myself” suggests multiple related product categories. In fashion, the equivalent is assembling an outfit; in electronics, selecting compatible accessories.

Potential applications include:

•Cross-Selling: Suggesting products needed or useful alongside the main item.
•Upselling: Comparing products at different price points and explaining differences in specifications or capabilities.
•Product Bundling: Assembling a set of products for a particular use case.
•Assisted Product Discovery: Helping customers who cannot identify the right category or product attributes independently.

However, the November VTEX announcement does not confirm a specialized automated mechanism for creating commercial bundles or optimizing recommendations by margin. These are potential uses of conversational search and recommendations that must be validated in an actual deployment.

Measuring the Commercial Performance of Shopping Assistant

Counting conversations and views of recommended products is not enough to assess business results. Merchants need to know whether the assistant helps shoppers complete purchases and generates incremental revenue.

Relevant measures include:

•Product Discovery Rate: Share of assistant interactions that lead shoppers to recommended products.
•Add-to-Cart Rate: Share of users who add a recommended item to their cart.
•Conversion Rate: Share of purchases among users interacting with the assistant.
•Average Order Value: Average purchase amount.
•Items per Order: Number of items in each order.
•Cross-Sell Attach Rate: Share of orders containing recommended complementary products.
•Revenue per Visitor: Revenue generated per visitor.
•Return Rate: Share of returned orders, particularly relevant for technical products and accessories.
•Support Deflection Rate: Share of inquiries resolved without involving a support employee.

It is particularly useful to analyze shoppers buying from a product category for the first time. In sporting goods, for example, a retailer could compare complete-equipment purchases between new customers who had access to Shopping Assistant and comparable customers who did not.

In home improvement, the merchant could measure additional products in orders built around a specific renovation project. In electronics, it could measure the share of main-device purchases accompanied by compatible accessories.

However, directly comparing visitors who voluntarily open AI chat with those who do not can produce a misleading performance estimate. Chat users may have stronger purchase intent from the outset.

To measure incremental revenue, a controlled A/B test with a group that cannot access Shopping Assistant is preferable. This helps establish whether the assistant actually raises conversion rates and order values rather than simply attracting the most interested shoppers.

Applications in B2B Commerce

VTEX also positions Shopping Assistant as relevant to B2B commerce. An industrial-equipment buyer, for example, could describe the technical requirements of a product and receive help locating matching items in the available catalog.

Another use case is finding consumables and replacement parts according to required specifications.

B2B stores must account for catalog restrictions and contractual terms. Available products, prices, and commercial offers can differ between organizations.

The official November announcement does not explain how Shopping Assistant handles specialized B2B contract terms when generating recommendations. Support for contract pricing and complex procurement rules should therefore be validated separately before deployment.

VTEX: An E-commerce Platform for B2C and B2B

VTEX is an enterprise e-commerce platform for building and managing online stores, B2B portals, marketplaces, and omnichannel sales operations.

VTEX Commerce Platform combines product catalog management, pricing, promotions, checkout, payments, orders, and logistics. It supports headless commerce and storefronts built with FastStore.

For B2B, the platform provides capabilities for managing buying organizations, contractual terms, assortments, and corporate purchasing. Its ecosystem also includes VTEX Ads Platform for retail media and VTEX CX Platform for customer service and conversational commerce.

Shopping Assistant is part of VTEX CX Platform. It provides an AI interface for product discovery, recommendations, and order tracking. The November 2026 update expands its use directly on online storefronts, including through native FastStore integration.

For e-commerce merchants, the principal potential benefits are better product discovery and cross-selling. In fashion, electronics, sporting goods, and home improvement, the assistant can help shoppers navigate product assortments and assemble groups of complementary products.

This creates opportunities to improve conversion rate, items per order, and average order value. Actual commercial impact still needs to be demonstrated with controlled experiments and sales data.

Trust Us Transform Your Performance Marketing.

Turn conversational commerce into measurable growth with reliable product data, customer journey analytics, and controlled experimentation.