faire.com — AI Site Grade

Analysis

Faire.com — AI-Visibility Audit

The site is a fully client-side-rendered Next.js application that delivers zero visible text content to any crawler that does not execute JavaScript — every key page (about, how-it-works, blog, newsroom, careers, supplier listings) returns 0 words of extractable text from a plain GET, including to GPTBot, ClaudeBot, Google-Extended, and PerplexityBot. This is the single most consequential finding: Faire is functionally invisible to the vast majority of AI crawlers that do not run JS.

Crawler Access

All AI bots receive HTTP 200 with no blocking — Cloudflare does not challenge GPTBot, ClaudeBot, PerplexityBot, Google-Extended, or any other tested agent. The robots.txt is permissive for the * catch-all (no AI-specific disallows except a Crawl-delay: 5 for Bytespider). However, every page served to every bot is a JavaScript shell: the HTML contains only a <div id="__next"> root, CSS/JS bundles, and font preloads. The body snippet for GPTBot fetching /how-faire-works is 671 KB of CSS-in-JS and framework code — zero words of actual content. The robots.txt explicitly blocks paginated category/supplier pages beyond page 11 and blocks all search result pages (/search?*), which means AI crawlers cannot discover the product catalog through standard crawl paths.

Schema Posture

The live site has zero JSON-LD schema on any page tested — homepage, about, how-it-works, blog, support, careers, newsroom. A Wayback Machine snapshot from January 2025 shows the homepage previously included Organization schema (with address, founding date, sameAs) and WebSite schema (with SearchAction). These have been removed in a subsequent deployment. The llms.txt endpoint returns a 404 (redirected to /pagenotfound). No FAQ, HowTo, Product, or BreadcrumbList schema exists anywhere.

Cold-Knowledge Gap

The LLM prior knows Faire as a $12B-valued wholesale marketplace founded in 2017 by Max Rhodes, operating across the US, Canada, Europe, and Australia, with net-60 payment terms and free returns. The live site's homepage meta description says only "Faire is the platform connecting thousands of brands and retailers worldwide" — no mention of valuation, founding story, geographic scope, or the net-60/free-returns differentiator that the model already knows. The site's thin 293-word homepage excerpt contains no structured data about the company's scale, funding, or unique value proposition.

External Signals

The DNS TXT records reveal an anthropic-domain-verification token and an openai-domain-verification token, confirming Faire has proactively registered with both Anthropic and OpenAI for crawler verification. Despite this, neither ClaudeBot nor GPTBot receives renderable content. The site uses Google Workspace for email, Cloudflare for CDN/WAF, and AWS DNS — a mature infrastructure stack that makes the JS-rendering gap more surprising. No external press or Reddit threads were surfaced through search, suggesting limited off-domain citation density that AI engines could draw on for retrieval-augmented generation.

Findings

1. All key pages render as empty JavaScript shells to AI crawlers High

   Every tested page (about, how-it-works, blog, newsroom, careers, supplier listings) returns 0 words of extractable text to GPTBot, ClaudeBot, Google-Extended, and PerplexityBot. The HTML contains only a root div and CSS/JS bundles.

   What to change: Implement server-side rendering (SSR) or static generation for key pages so that AI crawlers receive meaningful HTML content without requiring JavaScript execution.

2. Zero JSON-LD schema on any tested page High

   No Organization, WebSite, FAQ, HowTo, Product, or BreadcrumbList schema exists on the homepage, about, how-it-works, blog, support, careers, or newsroom pages. A January 2025 snapshot shows Organization and WebSite schema were previously present but have been removed.

   What to change: Re-add Organization and WebSite schema to the homepage, and add relevant schema types (e.g., Product, FAQ, BreadcrumbList) to appropriate pages.

3. llms.txt endpoint returns 404 Medium

   The llms.txt file at /llms.txt redirects to a 404 page, providing no guidance to AI crawlers about which pages to prioritize or how to interpret the site.

   What to change: Create an llms.txt file that lists key pages (about, how-it-works, blog, support) and provides a brief summary of the site's purpose.

4. Homepage meta description omits key differentiators Medium

   The homepage meta description is only 'Faire is the platform connecting thousands of brands and retailers worldwide' and does not mention valuation, founding story, geographic scope, net-60 payment terms, or free returns — information that LLMs already know about Faire.

   What to change: Expand the homepage meta description to include key differentiators such as valuation, founding year, geographic reach, and unique payment/return policies.

5. Robots.txt blocks paginated supplier pages and search results Medium

   The robots.txt disallows paginated category/supplier pages beyond page 11 and all search result pages (/search?*), preventing AI crawlers from discovering the full product catalog through standard crawl paths.

   What to change: Allow AI crawlers to access paginated supplier pages and search results, or provide an alternative crawl path (e.g., sitemap) to ensure catalog content is discoverable.

6. Limited external citations and press coverage surfaced Low

   Web searches for Faire reviews, news, and Reddit discussions returned zero results, indicating low off-domain citation density that AI engines could use for retrieval-augmented generation.

7. Proactive crawler verification does not yield renderable content High

   DNS TXT records show anthropic-domain-verification and openai-domain-verification tokens, confirming Faire registered with Anthropic and OpenAI. Despite this, both ClaudeBot and GPTBot receive empty JavaScript shells.

   What to change: Ensure that verified crawlers receive server-rendered HTML content, not just JavaScript shells.

What's working

• Permissive robots.txt for AI bots — The robots.txt does not disallow any AI-specific bots (GPTBot, ClaudeBot, Google-Extended, etc.) from crawling the site, allowing access to all pages that are not explicitly blocked.
• Proactive crawler verification with Anthropic and OpenAI — DNS TXT records include anthropic-domain-verification and openai-domain-verification tokens, indicating Faire has proactively registered with both AI companies for crawler verification.
• Mature infrastructure stack with Cloudflare and AWS — The site uses Cloudflare for CDN/WAF, AWS DNS, and Google Workspace for email, indicating a robust and scalable infrastructure.
• Historical presence of Organization and WebSite schema — A January 2025 Wayback Machine snapshot shows the homepage previously included Organization and WebSite schema, indicating the team has prior experience with structured data.