AI Shopping Advisor
As a customer, I want to find, based on product properties, in different steps, the best-matching product for me. Maybe combined with AI.
-
Cedric Vanderleenen
commented
Not sure if this is on the roadmap or to be announced at SCD but I was looking for some skin care products and asking Gemini to find the right ones from a specific webshop were I always order so I asked Gemini if it would not be a good move to have this embedded directly in the webshop and this was the result:
You are hitting on the exact nerve center of where e-commerce has moved. Using an LLM as a "use-case to product" translator is a night-and-day upgrade from classical parametric search (clicking checkboxes for sizes, brands, and colors).
To answer your question directly: **Yes, this absolutely exists today—both via ecosystem plugins and natively.** However, the market is currently split between the "Enterprise Giants" rolling out massive native consumer assistants, and "Platform Ecosystems" relying on app stores/plugins to bridge the gap.
Here is a breakdown of how this capability looks across Shopware and its major competitors right now:
---
### 1. Shopware
Since you are working at Shopware, you'll be glad to know the community has already started building this.
* **The Reality:** It currently exists primarily as a plugin/extension layer rather than a core, out-of-the-box native feature.
* **How it works:** Extensions like the **AI Product Advisor** (available in the Shopware Store) integrate directly into the Shopware 6 CMS via API. It connects OpenAI assistants (like GPT-4o-mini) to the store's product catalog. It can read product properties, ask dynamic follow-up questions, and map a user's prose back into valid Shopware filters to display exact product matches with tailored explanations.### 2. Shopify
Shopify has been very aggressive with its AI suite, **Shopify Magic** and its AI sidekick, **Sidekick**.
* **The Reality:** Native focus is mostly merchant-facing (generating copy, editing images, analyzing store data). For the front-end consumer-facing conversational shopping assistant, Shopify still relies heavily on its App Store.
* **How it works:** The Shopify App Store is flooded with highly advanced solutions (like *Shoply AI*, *eesel AI*, or *Prefixbox*). These apps embed a widget directly into the storefront, ingest the live Shopify product feed, track real-time inventory, and offer a "chat-to-checkout" experience where the AI handles the "why this product fits your goal" explanation.### 3. Adobe Commerce (Magento)
Adobe treats AI through a highly enterprise, data-centric lens using **Adobe Sensei / Adobe AI**.
* **The Reality:** Magento’s native AI focuses heavily on behavioral and catalog data processing rather than an open-ended "chat box." It looks at a shopper's real-time intent and automatically injects *Product Recommendation Units* onto the page.
* **How it works:** If you want a fully conversational "explain why this works for your use case" assistant on Magento, merchants typically use third-party **MCP (Model Context Protocol) servers** or enterprise conversational AI integrations. This allows external LLMs to securely "query" the Magento database via natural language to help the shopper.### 4. The Megabrands (The Native Benchmarks)
If you want to look at who is doing this natively at scale without plugins, look to the massive marketplaces:
* **Amazon's Rufus:** Rolled out globally, Rufus lives right in the search bar. You can type, *"I want to start a backyard container garden in a rainy climate, what do I need?"* and Rufus will curate a list of pots, soil, and seeds, explaining exactly why it chose them based on listing data and millions of user reviews.
* **Walmart's Sparky:** A fully agentic assistant embedded natively in their app that handles complex, goal-oriented shopping queries.---
## The Strategic Opportunity for Shopware
As someone shaping the future of Shopware, the massive opportunity here lies in **Native Platform Integration** rather than leaving it to third-party plugins.
When a shopping assistant is a third-party plugin, it frequently suffers from latency, disconnected data silos, and "hallucinated" answers because it doesn't truly *understand* real-time inventory or complex B2B pricing rules.
If Shopware can make a **Semantic Search & Use-Case Advisor** a native, core capability of its core platform, it would give merchants a massive advantage. Imagine a merchant being able to simply check a box in the Shopware Admin that says "Enable AI Conversational Discovery," and the platform natively builds the vector embeddings of the catalog—ensuring 100% accurate product recommendations with context-aware explanations.
-
Jonas Fuchs commented
As a shop owner, I want to give my customers the option to find the right product by offering a step-by-step guide that shows filtered product results based on the answers of the customer.