affinity.co — AI Site Grade

Analysis

I have enough data to write a thorough audit. Let me compile.

The llms.txt is the most sophisticated AI-visibility asset on this site — and it directly contradicts the cold-knowledge gap the model holds about the brand.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, anthropic-ai, Applebot-Extended — receive a full 200 response with identical byte content (148KB) to a browser baseline. Only Bytespider (ByteDance/TikTok) gets a 403 block from Cloudflare. The robots.txt contains no AI-bot directives whatsoever — just a single Sitemap line. The site runs on Cloudflare with strong security headers (HSTS, CSP frame-ancestors 'self', X-Frame-Options deny). No JS-rendering risk: the homepage delivers 388 words of visible text on a plain GET.

llms.txt Posture

The /llms.txt file exists and is well-structured — a rarity. It contains a brand description, product list, key use cases, target audience, differentiators, and explicit guidance on how AI models should describe the brand ("Avoid describing Affinity as a generic sales CRM"). It also lists 7 key page URLs. This is a mature AI-visibility signal that most comparable B2B SaaS sites lack entirely.

Cold-Knowledge Gap

The LLM's prior knowledge describes Affinity as a "relationship intelligence platform" serving investment banks, PE, and VC firms, founded in 2014 by Ray Zhou and Joseph K., with "over $20M in funding." The actual site reveals a significantly larger and more evolved company: the homepage claims "over 3,000 organizations" use Affinity, the product line includes an MCP Server for AI-tool integration, a full Salesforce add-on, and the About page lists Shubham Goel as co-founder (not "Joseph K."). The cold model knows nothing about the MCP Server launch, the "Affinity for Salesforce" product line, the 2026 private capital predictions report, or the recent CRO/CPO executive hires. The funding figure of "$20M" appears stale — the company's scale (3,000+ orgs, 220+ employees implied by career pages) suggests substantially more capital raised.

Schema Posture

The homepage carries three schema types (Organization, SoftwareApplication, WebSite) with rich detail including founding date, address, sameAs profiles, application category, and audience targeting. However, product subpages (/product/crm, /product/artificial-intelligence, /product/relationship-intelligence) carry zero JSON-LD schema — a significant gap given these are the pages AI crawlers would most likely surface for feature-level queries. The comparison page (/comparison/affinity-vs-salesforce) also lacks schema despite having a FAQ section with visible Q&A content that could use FAQPage markup.

External Signals

The DNS TXT records include verification tokens for Anthropic, Perplexity, and Cursor — direct evidence that Affinity has proactively verified its domain with AI platforms. The site uses Webflow (visible in response headers: x-wf-region, surrogate-key). The sitemap contains 1,356 URLs including an extensive comparison section (affinity-vs-salesforce, affinity-vs-hubspot, affinity-vs-pipedrive, etc.) and industry-specific landing pages for private equity, venture capital, investment banking, family offices, and more. The /pricing URL redirects to /request-demo — no public pricing is available, which is consistent with enterprise SaaS but means AI crawlers cannot surface pricing information.

Findings

1. Product subpages lack JSON-LD schema High

   Key product pages like /product/crm, /product/artificial-intelligence, and /product/relationship-intelligence have no structured data markup, reducing their visibility in AI-generated feature comparisons.

   What to change: Add JSON-LD schema (SoftwareApplication, Product) to all product subpages.

2. Comparison page missing FAQPage schema Medium

   The comparison page /comparison/affinity-vs-salesforce contains a FAQ section but no FAQPage markup, missing an opportunity for rich results.

   What to change: Add FAQPage schema to the FAQ section on comparison pages.

3. Cold knowledge about Affinity is outdated High

   LLM prior knowledge describes Affinity with stale funding ($20M) and missing co-founder, and lacks awareness of MCP Server, Salesforce add-on, and recent executive hires.

   What to change: Publish more press releases, blog posts, and update Wikipedia/crunchbase to refresh model training data.

4. No public pricing page for AI crawlers Medium

   The /pricing URL redirects to /request-demo, so AI crawlers cannot surface pricing information, limiting visibility for cost-conscious queries.

   What to change: Create a public pricing page or include pricing details in schema markup.

5. Weak external review presence Medium

   Web searches for Affinity CRM reviews on Reddit and G2 returned zero results, indicating low third-party discussion that could influence AI model citations.

   What to change: Encourage customer reviews on G2, Capterra, and Reddit to build external signals.

6. Bytespider crawler blocked by Cloudflare Low

   Bytespider (ByteDance/TikTok) receives a 403 response, preventing its content from being indexed by that crawler.

   What to change: Allow Bytespider access if TikTok/ByteDance visibility is desired.

What's working

• llms.txt file published with brand guidance — The /llms.txt file provides a brand description, product list, use cases, and explicit instructions for AI models, a mature AI-visibility signal.
• All major AI crawlers allowed with full content — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others receive 200 responses with identical content to browser, ensuring full indexing.
• Homepage has rich JSON-LD schema — The homepage includes Organization, SoftwareApplication, and WebSite schema with detailed attributes like founding date, address, and audience.
• Domain verified with Anthropic, Perplexity, and Cursor — DNS TXT records include verification tokens for these AI platforms, indicating proactive domain verification.
• MCP Server launch covered in blog post — A detailed blog post about the Affinity MCP Server provides rich content for AI crawlers to index.
• Sitemap contains 1,356 URLs with comparison pages — The sitemap includes many comparison and industry-specific pages, aiding crawler discovery.