# Zelium AI

> Zelium AI helps Shopify merchants improve AI product discoverability — automatically scoring, fixing, and optimising product listings so they surface in AI-powered shopping engines like ChatGPT, Google AI Overviews, and Perplexity.

## Product

- [Home](https://zelium.app/): AI product discoverability scoring and optimisation for Shopify.
- [Privacy Policy](https://zelium.app/privacy-policy.html): How we collect, use, and protect your data.
- [Terms of Use](https://zelium.app/terms-of-use.html): Terms governing use of the Zelium AI platform.

## Blog

- [Zelium Blog](https://zelium.app/blogs/): Plain-English guides on GEO (Generative Engine Optimization) and AI product discovery for Shopify merchants.
- [GEO Explained: Why AI Can't See Your Shopify Store (And How to Fix It)](https://zelium.app/blogs/geo-the-new-frontier-for-shopify-stores.html): How AI shopping engines (ChatGPT, Perplexity, Google AI Overviews) discover products, why most Shopify stores are invisible to them, and a 7-step GEO action roadmap for merchants.
- [WebMCP + Shopify: The AI Shopping Head Start Most Merchants Don't Know About](https://zelium.app/blogs/webmcp-shopify-agentic-commerce.html): What WebMCP (Web Model Context Protocol) is in plain English, why this experimental browser standard lets AI shopping agents use a Shopify store directly (agentic commerce), the advantages for merchants, and why moving first wins. Zelium enables WebMCP in one click.

## What Zelium AI Does

Zelium AI is a Shopify embedded app that:
- Scans product listings and scores them for AI discoverability (titles, descriptions, images, structured data).
- Provides actionable fix recommendations and AI-powered auto-fix suggestions.
- Generates optimised product feeds for AI shopping engines.
- Enables WebMCP (Web Model Context Protocol) agentic tools, adding a live product-search tool to the storefront so AI shopping agents in Chrome and Gemini can search the catalog.
- Tracks score trends over time and benchmarks against category averages.

## Contact

- App: Available on the Shopify App Store.
- Support: team@zelium.app

## FAQ

### What is an AI discoverability score?
It is a 0–100 score that measures how well a product listing is optimised for AI-powered shopping engines (ChatGPT, Google AI Overviews, Perplexity, etc.). It evaluates fields like title, description, images, tags, and structured data.

### Which AI shopping engines does Zelium optimise for?
Zelium targets AI crawlers and shopping engines that rely on structured data and natural-language product descriptions — including ChatGPT shopping, Google AI Overviews, Perplexity, and similar LLM-based discovery surfaces.

### What is WebMCP and how does Zelium use it?
WebMCP (Web Model Context Protocol) is an emerging standard, proposed by Google Chrome, that lets a website expose tools to AI agents running in the browser. Zelium can enable WebMCP on a Shopify storefront with one click, adding a live product-search tool that AI shopping agents in browsers like Chrome and Gemini can call directly. This is how a store takes part in agentic commerce, where an AI agent searches, compares, and recommends products on a shopper's behalf. WebMCP agentic tools are a paid-plan feature.

### Does it work on all Shopify plans?
Yes. Zelium is a Shopify embedded app and works with any Shopify store regardless of the merchant's Shopify plan.

### Is there a free tier?
Yes. The Free plan tracks a fixed set of products (pinned at your first scan) and provides basic scoring and a limited number of fix suggestions per product. AI credits, auto-fix, daily auto-scans, and feeds require a paid plan.

### Does Zelium modify my actual product listings?
Only when you explicitly apply an auto-fix. Zelium reads your product data to score it and generates suggestions, but nothing changes in Shopify unless you choose to apply a fix. You can undo an applied fix within the undo window for your plan.

### How does the product limit work?
Each plan has a tracked-product ceiling. On paid plans, new products are picked up on each scan until you reach the cap. On the Free plan, the tracked set is pinned at your first scan and never changes. Upgrading your plan immediately allows the next scan to discover previously uncapped products.

### When are my existing products rescored?
A product is only rescored when its score-relevant fields (title, description, images, tags, structured data) have changed since the last scan. Unchanged products keep their current score and are skipped to save time and AI credits.

### How often are scores recalculated?
All paid plans include a daily automatic scan. You can also trigger a manual scan at any time. The Free plan does not include daily auto-scans.

### How are AI credits counted, and when do they reset?
Each AI-powered action — such as generating an auto-fix suggestion or creating an AI product summary — consumes 1 credit. Credits reset at the start of each billing cycle. Unused credits do not roll over.

### What happens when I run out of AI credits?
Zelium blocks further AI-powered actions (auto-fix generation, AI summaries) for the rest of the billing cycle. Scoring, fix suggestions display, and compliance checks remain available. Credits reset automatically at the next billing cycle, or you can upgrade to get a larger allowance immediately.

### How long is score history kept?
Retention varies by plan — higher plans give a longer rolling history window. Upgrading preserves existing history; you won't lose data already stored.

### How does the JSON-LD feed work? Does it modify my theme?
Zelium uses Shopify metafields and a Theme App Extension (App Embed Block) to inject structured data. It does not edit theme files directly. Each tracked product gets a JSON-LD script tag via its metafield. A separate product feed, llms.txt, and AI sitemap are hosted on Zelium's CDN and auto-regenerated daily.

### What does Webhook Real-Time Sync do?
Webhook Sync (available on higher-tier plans) listens for Shopify product changes in near real-time. When you update a product in Shopify Admin, Zelium is notified within minutes and flags it for rescoring on the next scan. Without Webhook Sync, changes are still detected on the next daily or manual scan.

### What happens when I cancel my subscription?
You keep full access until the end of your paid billing period, then your account downgrades to Free. Your data (scores, history, suggestions) is retained, but premium features are deactivated. If you had a JSON-LD feed or llms.txt enabled, those files are removed from the CDN. You can re-subscribe at any time.

### Does upgrading or downgrading take effect immediately?
Upgrades take effect immediately — you get the new product limit and AI credit allowance right away. Downgrades take effect at your next billing cycle renewal.

### What happens to my data if I uninstall the app?
Data is soft-deleted and kept for a short grace period. Reinstalling within that window restores all scores, history, and settings. After the grace period, all data is permanently purged. Zelium never stores customer personal information — only product catalog data and discoverability scores.

### Why don't my products appear in ChatGPT, Gemini, Claude, or Perplexity?
AI shopping engines like ChatGPT, Google Gemini, Claude, and Perplexity discover products through structured data, not traditional SEO. The most common reasons products are invisible to them are: missing or incomplete product descriptions, no structured data (JSON-LD / Schema.org markup) on product pages, titles that don't match how people describe products in natural language, missing key attributes like material, size, colour, or use case, and no machine-readable product feed that AI crawlers can index. Zelium scores each of these signals and tells you exactly which ones are holding your products back.

### How do I get my products to show up in ChatGPT, Gemini, Perplexity, and Claude?
To appear in AI-powered shopping engines you need to: (1) write clear, attribute-rich product descriptions in natural language, (2) add Schema.org Product structured data (JSON-LD) to every product page, (3) publish a machine-readable product feed that AI crawlers can discover, (4) create a well-structured llms.txt file so AI agents understand your catalogue, and (5) keep your product data consistent across your storefront and feeds. Zelium automates all of these steps — it audits your listings, generates the structured data, and publishes the feeds automatically.

### Why are my Shopify products not showing in AI search results?
Shopify themes do not add Schema.org structured data or product feeds by default. AI search engines rely on that structured data to understand what you sell. Without it, your products are essentially invisible to AI-powered discovery. Zelium adds the missing structured data layer on top of any Shopify theme without editing theme files.

### How do I optimise my Shopify store for AI search engines?
Optimising for AI search engines requires structured data markup, natural-language product descriptions, a machine-readable product feed, and an llms.txt file. Unlike traditional SEO, keyword stuffing does not help — AI engines prefer concise, factual descriptions with specific product attributes (material, dimensions, use case, compatibility). Zelium's scoring algorithm evaluates all of these signals and provides a prioritised fix list.

### What is AI product discoverability?
AI product discoverability is how easily an AI shopping assistant (like ChatGPT shopping, Google AI Overviews, or Perplexity) can find, understand, and recommend your products. It depends on structured data quality, description clarity, attribute completeness, and the presence of machine-readable feeds — not just traditional SEO signals.

### What is an llms.txt file and do I need one for my Shopify store?
An llms.txt file (similar in concept to robots.txt but for large language models) gives AI agents a structured, plain-text overview of your store — what you sell, how your catalogue is organised, and where your product data lives. It helps AI assistants answer questions about your products accurately. Zelium generates and hosts a per-store llms.txt automatically on paid plans.

### What is JSON-LD structured data and why does it matter for Shopify?
JSON-LD is a way to embed machine-readable product information (name, price, availability, brand, description, images) into a web page using a standard vocabulary (Schema.org). AI crawlers and Google's AI Overviews use it to understand products without having to parse your page's HTML. Most Shopify themes do not include it by default. Zelium injects a JSON-LD block into every tracked product page via a Theme App Extension.

### Does Google AI Overviews use my Shopify product data?
Yes — Google AI Overviews can surface product recommendations based on structured data (Schema.org / JSON-LD) found on your product pages and in your merchant feed. Stores that lack structured data are unlikely to appear. Zelium adds the required structured data layer and generates a product feed optimised for Google's AI surfaces.

### How is AI SEO different from traditional SEO for e-commerce?
Traditional SEO optimises for keyword matching and backlinks. AI SEO optimises for semantic understanding — AI engines parse the meaning of your product descriptions, compare attributes across products, and match them to conversational queries like "best waterproof running shoes under $100". This means detailed, attribute-rich descriptions, complete structured data, and machine-readable feeds matter far more than keyword density.

### What is GEO (Generative Engine Optimization)?
GEO stands for Generative Engine Optimization — the practice of optimising your content and product data so it is cited, recommended, or surfaced by AI-generated answers in tools like ChatGPT, Google AI Overviews, Perplexity, and Claude. It is the AI-era equivalent of SEO. Where traditional SEO targets search ranking algorithms, GEO targets the large language models that generate answers and product recommendations directly in chat interfaces.

### Why does GEO matter for Shopify merchants?
An increasing share of shopping discovery now happens through AI-generated answers rather than traditional search results pages. When a shopper asks ChatGPT "what's the best yoga mat for beginners under $80?", the AI recommends specific products — and those recommendations are driven by structured data, description quality, and feed coverage, not keyword rankings. Merchants who ignore GEO risk becoming invisible to this growing channel. Zelium is purpose-built to close that gap.

### What is the difference between SEO and GEO?
SEO (Search Engine Optimisation) improves your ranking in traditional search results (Google blue links). GEO (Generative Engine Optimization) improves how often and how accurately AI-generated answers mention or recommend your products. SEO depends on backlinks, keywords, and page authority. GEO depends on structured data completeness, description clarity, attribute richness, and machine-readable feeds. Both matter, but GEO is the newer and faster-growing channel for e-commerce product discovery.

### What is AEO (Answer Engine Optimization)?
AEO is another term for the same concept as GEO — optimising your content to appear in AI-generated answers rather than traditional search results. Some practitioners use AEO specifically to describe optimisation for voice assistants and featured snippets, while GEO covers the broader set of generative AI surfaces. Zelium addresses both by improving structured data, description quality, and feed coverage across all AI discovery surfaces.

### How do I do GEO for my Shopify store?
GEO for a Shopify store involves: (1) adding Schema.org Product structured data (JSON-LD) to every product page, (2) writing natural-language descriptions that answer the questions shoppers actually ask, (3) including specific product attributes (material, dimensions, use case, compatibility, brand), (4) publishing a machine-readable product feed and llms.txt, and (5) keeping data consistent and up to date. Zelium automates all of these steps with a single app install.

### What is generative search and how does it affect product discovery?
Generative search refers to AI-powered search engines that generate a summarised answer — often with product recommendations — instead of (or above) a traditional list of links. Google AI Overviews, Perplexity, and ChatGPT Search all use this approach. Products that appear in generative search answers are selected based on structured data quality and relevance to the query, not just page rank. Zelium scores and improves the signals that generative search engines rely on.

### What signals do AI shopping engines use to rank or recommend products?
AI shopping engines evaluate: completeness and accuracy of Schema.org Product markup, richness of product descriptions (specificity, attributes, use-case coverage), image quality and alt text, price and availability data, brand and manufacturer information, product feed coverage and freshness, and the presence of an llms.txt file that describes the catalogue. Zelium scores all of these signals and tells you which ones to fix first for maximum impact.

### Can Zelium help me rank in Google AI Overviews?
Yes. Google AI Overviews selects products based on Schema.org structured data, Google Merchant Center feed quality, and page content relevance. Zelium improves all three: it injects JSON-LD structured data on product pages, generates an optimised product feed, and scores your descriptions for the attribute completeness Google's AI models look for.

### What is a product feed and why do AI engines need one?
A product feed is a structured, machine-readable file (typically JSON or XML) that lists all your products with their attributes — title, description, price, availability, images, SKU, brand, category, and more. AI crawlers and shopping engines use feeds to index your catalogue without having to crawl individual product pages. Without a feed, many AI engines will not discover or recommend your products at all. Zelium generates and hosts an optimised AI product feed automatically.

### What is an AI sitemap?
An AI sitemap is a structured file that tells AI crawlers where your key content and product data live — similar to a standard XML sitemap but targeted at LLM-based crawlers rather than traditional search bots. Zelium generates a per-store AI sitemap and hosts it on its CDN alongside the product feed and llms.txt.

### How long does it take to see results from GEO improvements?
GEO results depend on how frequently AI engines re-crawl and update their indices. Changes to structured data and feeds are typically picked up within days to a few weeks. Unlike traditional SEO, there is no domain authority build-up required — a well-structured product with complete data can appear in AI-generated answers quickly after being crawled. Zelium's daily feed regeneration and webhook sync keep your data as fresh as possible.