glassette.com — AI Site Grade

Analysis

Glassette: A content-rich marketplace invisible to AI knowledge systems

The homepage canonical points to /index rather than /, creating a self-inflicted duplicate-content problem for every crawler that respects canonical tags.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended — receive a full 200 response with the same 710KB HTML payload as a browser. No UA-based blocking exists. The site runs on Next.js hosted on Vercel behind Cloudflare, with X-Powered-By: Next.js and Cloudflare ray headers present. The robots.txt contains no AI-bot-specific directives; the catch-all * rule allows / and disallows only cart, checkout, account, and search paths. The /llms.txt URL returns a full HTML page (710KB, text/html) rather than a plain-text LLM content map — a misconfiguration that defeats the purpose entirely.

Cold-Knowledge Gap

A frontier LLM queried cold about glassette.com returned: "I don't have specific, verifiable information about glassette.com in my training data." The model cannot confirm what the brand does, who it serves, or any reputational signals. This is a complete knowledge vacuum despite the site being a launched marketplace since November 2021 with 8,357 products across 200+ brands. The site describes itself as "a content-led lifestyle platform" curating interiors, food, travel, and culture — a positioning that exists nowhere in the model's prior.

Schema Posture

The homepage, About page, Shop page, category pages, and brand directory all contain zero JSON-LD structured data. No Organization, WebSite, WebPage, ItemList, Product, or BreadcrumbList schema exists anywhere on the pages sampled. The only schema found on the entire domain is a single Article type on individual discover articles (e.g., the barbecue article), which includes author, datePublished, and headline. For a marketplace selling 8,357 products across hundreds of brands, the absence of Product and Offer schema is a critical gap for AI-driven search engines and knowledge panels.

Content Architecture

The homepage and /discover page are near-identical — both display the same 26-item carousel of editorial articles with dates ranging from April to May 2026. The site is editorially rich (articles on BBQ hosting, Cannes guides, ceramicists to follow, Met Gala interiors) but the content is buried under a carousel UI that repeats the same navigation sidebar on every article page, inflating page weight. The /studio-g subpage (a creative agency arm) contains lorem ipsum placeholder text alongside real client logos (Google Pixel, IKEA, NEXT, Wahaca, Starling Bank), suggesting an unfinished page that is nonetheless indexed and in the sitemap.

External Signals

Web searches for "glassette" combined with "homeware marketplace," "reviews," and "Laura Jackson" returned zero results across multiple queries. No Reddit threads, no press mentions, no review sites surfaced. The brand has Pinterest (pinterest-site-verification), Klaviyo, and multiple Google Search Console verification TXT records, indicating active marketing operations — but the external citation footprint is effectively nil, which compounds the cold-knowledge gap. The Wayback Machine returned no snapshots for the domain, suggesting either a recent domain or blocking of archiving.

Findings

1. Zero JSON-LD structured data across all sampled pages High

   No Organization, WebSite, WebPage, ItemList, Product, or BreadcrumbList schema exists on the homepage, About, Shop, category, or brand pages. Only a single Article schema appears on discover articles.

   What to change: Add Organization, WebSite, WebPage, ItemList, Product, and BreadcrumbList JSON-LD schema to all relevant pages.

2. /llms.txt returns full HTML page instead of plain-text LLM content map High

   The /llms.txt URL returns a 710KB HTML page with text/html content type, defeating its purpose as a lightweight LLM content map.

   What to change: Replace the /llms.txt endpoint with a plain-text file listing key pages and summaries for LLM consumption.

3. Frontier LLM has no knowledge of Glassette as a brand or marketplace High

   A cold query to a frontier LLM returned no verifiable information about glassette.com, indicating the brand is absent from training data despite being live since November 2021.

   What to change: Increase external citations through press, backlinks, and structured data to improve inclusion in LLM training data.

4. Zero external citations found across web searches High

   Searches for 'glassette.com' combined with 'homeware marketplace', 'reviews', and 'Laura Jackson' returned no results. No Reddit threads, press mentions, or review sites were found.

   What to change: Build a PR and backlink strategy to generate external mentions and reviews.

5. Homepage canonical points to /index creating duplicate content Medium

   The homepage canonical URL points to /index instead of /, causing a self-inflicted duplicate content issue for all crawlers that respect canonical tags.

   What to change: Change the homepage canonical to point to https://www.glassette.com/.

6. Studio G page contains lorem ipsum placeholder text Medium

   The /studio-g page includes lorem ipsum placeholder text alongside real client logos, suggesting an unfinished page that is indexed and in the sitemap.

   What to change: Complete the Studio G page with real content or remove it from the sitemap until finished.

7. No Wayback Machine snapshots for the domain Low

   The Wayback Machine returned no snapshots for glassette.com, suggesting either a recent domain or blocking of archiving.

   What to change: Ensure the site is not blocking archive.org crawlers and allow archiving.

What's working

• All major AI crawlers receive full 200 responses — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, and Applebot-Extended all get full HTML content with no UA-based blocking.
• Site features rich editorial content across multiple categories — The site publishes detailed articles on topics like BBQ hosting, Cannes guides, and interior design, providing valuable content for AI indexing.
• Article schema present on discover articles — Individual discover articles include Article schema with author, datePublished, and headline, aiding AI understanding of editorial content.
• robots.txt allows all key content paths — The robots.txt allows crawling of /, /shop, /discover, /brands, and other content sections, only blocking cart, checkout, account, and search.
• Sitemap contains 80 URLs covering key pages — The sitemap includes 80 URLs, covering homepage, shop, discover, brands, and category pages, aiding crawler discovery.