kustomer.com — AI Site Grade

Analysis

I have enough data now. Let me compile the audit.

Kustomer's Meta-shaped blind spot

The site's most consequential AI-visibility gap is not technical — every AI crawler gets a 200 with full content — but narrative: the cold-knowledge model prominently recalls the Meta acquisition ($1B, 2020) and spinout (2022), yet the site's homepage, about page, and product pages never mention Meta at all. The only references are buried in two blog posts from 2022 and 2023, one of which carries a note saying "Kustomer is no longer part of Meta." This creates a dangerous split between what AI engines know about the brand (a company with a complicated Meta history) and what the brand says about itself (a pure AI-native CX platform with no prior corporate entanglements).

Crawler Access

All eleven tested AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, Perplexity-User — receive a 200 with identical 114KB payload as a browser baseline. No UA-based blocking, no JS shell, no Cloudflare challenge. The site runs on Vercel with Google Cloud DNS and Google Workspace mail. The robots.txt is a single User-Agent: * Allow: / rule with no AI-bot-specific directives — a permissive posture that is technically correct but misses the opportunity to signal crawl priority to AI engines.

Cold-Knowledge Gap

The frontier model queried cold describes Kustomer as "acquired by Meta in 2020 for an estimated $1 billion; later spun out as an independent company in 2022." It names legacy customers (Glossier, Ring, Away) and flags "mixed signals post-spinout" and "lingering concerns about data privacy given the prior ownership." The site itself mentions none of this on its primary pages. The about page tells a founding story that jumps from Assistly/Desk.com (2015) directly to "today" — skipping the Meta years entirely. The homepage claims "Trusted by 600+ companies" but names none of the well-known brands the model associates with Kustomer.

Schema Posture

Every page carries the same minimal Organization + WebSite schema with sameAs links to LinkedIn, Twitter, Facebook, and YouTube. No Product, SoftwareApplication, FAQPage, HowTo, or WebPage schema exists anywhere on the site. The blog uses BlogPosting with datePublished and author. The comparison pages (/zendesk/, /intercom/, etc.) have no comparison-specific schema. The pricing page has no Product or Offer schema. This is a significant missed signal for AI engines trying to understand what Kustomer *is* as a product category.

External Signals

The DNS TXT records reveal an unusually broad verification footprint: OpenAI, Anthropic, Cursor, Notion (x2), Slack, HubSpot, Pendo, LaunchDarkly, MongoDB, Loom, Docker, Postman, Atlassian, Apple, Zoom — indicating deep integration with the AI/developer ecosystem. The llms.txt is well-structured with 50+ links across platform, industry, competitor comparison, and resource sections — one of the most complete llms.txt files observed. The security page redirects to a JS-rendered trust portal (trust.kustomer.com) that returns zero visible text to a plain GET — a thin-content risk for crawlers seeking compliance/security documentation.

Findings

1. Meta acquisition and spinout omitted from primary pages High

   Cold-knowledge models recall Kustomer's $1B Meta acquisition (2020) and spinout (2022), but the homepage, about page, and product pages never mention Meta. The only references are in two blog posts from 2022 and 2023, creating a dangerous split between AI knowledge and site narrative.

   What to change: Add a brief mention of the Meta chapter on the about page and homepage, framing it as a period of accelerated AI investment that led to the current independent product.

2. No Product or SoftwareApplication schema on any page High

   Every page carries only minimal Organization and WebSite schema. No Product, SoftwareApplication, FAQPage, HowTo, or WebPage schema exists, missing a significant signal for AI engines to understand Kustomer's product category.

   What to change: Add SoftwareApplication schema with applicationCategory, operatingSystem, and offers to product and homepage. Add Product and Offer schema to the pricing page.

3. Comparison pages lack FAQ or comparison schema Medium

   Pages like /zendesk/ compare Kustomer to competitors but have no FAQPage or comparison-specific schema, reducing their ability to appear in AI-generated comparison answers.

   What to change: Add FAQPage schema with common comparison questions and answers on each competitor page.

4. Security page returns zero visible text to crawlers Medium

   The /security/ page redirects to a JS-rendered trust portal (trust.kustomer.com) that returns zero visible text on a plain GET, creating a thin-content risk for crawlers seeking compliance documentation.

   What to change: Ensure the security page serves static HTML with key compliance certifications and data handling summaries, or add a server-side rendered fallback.

5. Homepage does not name well-known customers Medium

   The homepage claims 'Trusted by 600+ companies' but names none of the brands (Glossier, Ring, Away) that cold-knowledge models associate with Kustomer, missing a trust signal.

   What to change: Add a customer logo bar or testimonial section on the homepage featuring recognizable brands.

6. Robots.txt lacks AI-bot-specific directives Low

   The robots.txt has a single Allow: / rule with no AI-bot-specific directives, missing the opportunity to signal crawl priority or restrict certain paths for AI crawlers.

   What to change: Add explicit Crawl-Delay or separate rules for AI bots to prioritize key pages like product and pricing.

7. About page omits the Meta chapter entirely High

   The about page jumps from Assistly/Desk.com (2015) directly to 'today', skipping the Meta acquisition and spinout period, which creates a narrative gap for AI models that know this history.

   What to change: Add a paragraph about the Meta era, positioning it as a period of accelerated AI investment that led to the current independent product.

8. Pricing page lacks Offer schema Medium

   The pricing page has no Product or Offer schema, missing a key signal for AI engines to understand pricing tiers and plans.

   What to change: Add Product and Offer schema for each pricing tier with price, currency, and description.

What's working

• All AI crawlers receive full content with 200 status — All eleven tested AI crawlers receive a 200 with identical 114KB payload as a browser baseline, with no UA-based blocking or JS shells.
• Well-structured llms.txt with 50+ links — The llms.txt is one of the most complete observed, with 50+ links across platform, industry, competitor comparison, and resource sections.
• Broad integration ecosystem verified via DNS TXT records — DNS TXT records show verification with OpenAI, Anthropic, Cursor, Notion, Slack, HubSpot, and many others, indicating deep integration with the AI/developer ecosystem.
• Permissive robots.txt allows all crawlers — The robots.txt has a single Allow: / rule with no AI-bot-specific restrictions, ensuring all crawlers can access the site.
• Blog posts use BlogPosting schema with dates and author — The blog uses BlogPosting schema with datePublished and author, providing structured metadata for AI crawlers.
• Consistent Organization schema with social links — Every page carries Organization and WebSite schema with sameAs links to LinkedIn, Twitter, Facebook, and YouTube.