degreed.com — AI Site Grade

Analysis

I have enough data to write the audit. Let me compile the findings.

AI crawlers see a JS shell, not a product page

The homepage at degreed.com returns only 23 words of visible text from a plain GET — a JavaScript shell. Every major AI bot (GPTBot, ClaudeBot, Google-Extended, PerplexityBot, OAI-SearchBot, ChatGPT-User) gets a 200 status with identical byte size to a browser, meaning none are blocked at the edge. But the content they receive is the same empty JS shell. The site runs on WP Engine (Azure DNS, x-powered-by: WP Engine), a WordPress hosting stack that serves a client-rendered React app.

Crawler Access

robots.txt contains only User-agent: * with no disallow rules — no AI-bot-specific directives at all. llms.txt returns 404. sitemap.xml also returns 404. The DNS TXT records include both openai-domain-verification and anthropic-domain-verification tokens, confirming Degreed has proactively verified the domain with both OpenAI and Anthropic for retrieval-augmented generation access. Yet the site delivers no machine-readable content map. Bytespider (ByteDance) gets a 502 error, suggesting a WAF block specific to that crawler.

Cold-Knowledge Gap

The LLM knows Degreed as the company that "popularized the concept of learning experience platforms (LXPs)" and recalls its $400M+ funding, founders David Blake and Mike Palmer, and 2023-2024 layoffs/restructuring. The site itself never mentions funding, founders, or layoffs. The cold knowledge is more historically accurate than the site's current self-presentation. The site heavily pushes "Degreed Maestro" (AI coaching), "Degreed Open Library," and "Degreed Skills+" — product names the LLM does not know cold. The LLM still calls it "Skill Coach" (an older product name), while the site has fully rebranded to "Degreed Maestro."

Schema Posture

Every page carries the same JSON-LD Organization block with sameAs links to Facebook, X, LinkedIn, YouTube, Instagram, and a numberOfEmployees range of 201-500. The WebSite schema includes a SearchAction. However, no page uses Product, SoftwareApplication, or FAQPage schema — despite every product page having an FAQ section with visible Q&A content. The datePublished values are inconsistent: the homepage claims 2026-02-13 (a future date at time of writing), while the LXP page shows 2023-08-24. The WebPage schema on the homepage has about pointing to the Organization ID, which is a structural misuse.

External Signals

The success stories page lists 8 enterprise clients (TEKsystems, 84.51°, Capgemini, BT Group, Exness, Travelers, Cigna, State Street, Ericsson) with specific metrics — TEKsystems ramped sellers "6x faster," Capgemini trained "150,000 employees in 10 weeks," Cigna serves "70,000 employees." The site claims "300+ clients reaching 10 million users in 160+ countries" and "30% of the Fortune 50." These are strong external validation signals, but none are structured as Review, Testimonial, or CaseStudy schema. The blog content referenced in the homepage FAQ section was not found at a discoverable URL path — the site lacks a visible /blog/ directory in the sitemap or navigation.

Content & Answer Signals

Deep pages (LXP, Maestro, Skills+, Open Library, Academies) render 1,100-1,600 words of server-side text — the homepage is the only JS-heavy page. All product pages include FAQ sections with real Q&A content (e.g., "What is a learning experience platform (LXP)?" on the LXP page). The site uses comparison language ("Other systems give you siloed data. We give you more...") on the "Why Degreed" page, but no structured comparison tables. Heading structures are clean with single H1s and logical H2/H3 hierarchies.

Findings

1. Homepage renders as a JavaScript shell for AI crawlers High

   The homepage returns only 23 words of visible text from a plain GET, serving a client-rendered React app. All major AI bots receive the same empty shell despite getting a 200 status.

   What to change: Implement server-side rendering or static generation for the homepage to deliver meaningful content to crawlers.

2. llms.txt returns 404 Medium

   The site does not provide an llms.txt file, which is a recommended standard for AI crawlers to discover content.

   What to change: Create an llms.txt file listing key pages for AI crawlers.

3. Sitemap.xml returns 404 Medium

   The sitemap is missing, making it harder for crawlers to discover all pages.

   What to change: Generate and submit a sitemap.xml to search engines and AI crawlers.

4. Bytespider receives 502 error Low

   ByteDance's crawler Bytespider gets a 502 error, indicating a WAF block that may limit visibility on Chinese platforms.

   What to change: Investigate and allow Bytespider if Chinese market visibility is desired.

5. No AI-bot-specific rules in robots.txt Low

   The robots.txt only has a wildcard rule with no disallows, missing an opportunity to guide AI crawlers to key content.

   What to change: Add specific rules for AI bots to prioritize crawling of product pages.

6. Product pages lack structured schema High

   No page uses Product, SoftwareApplication, or FAQPage schema, despite FAQ sections and product descriptions. This reduces rich result eligibility.

   What to change: Add SoftwareApplication schema to product pages and FAQPage schema to FAQ sections.

7. Homepage datePublished is a future date Medium

   The homepage JSON-LD claims datePublished 2026-02-13, which is incorrect and may confuse crawlers.

   What to change: Correct the datePublished to the actual publication or last modified date.

8. Inconsistent datePublished across pages Low

   The LXP page shows datePublished 2023-08-24 while the homepage shows 2026-02-13, indicating poor schema maintenance.

   What to change: Ensure datePublished is consistent and accurate across all pages.

9. WebPage schema about property misused Low

   The homepage WebPage schema uses about to point to the Organization ID, which is structurally incorrect.

   What to change: Remove the about property from WebPage or point it to the main entity of the page.

10. LLM cold knowledge does not match site content Medium

    The LLM knows Degreed's history and funding but not current product names like Degreed Maestro. The site omits historical context that LLMs recall.

    What to change: Include key historical facts and product evolution on the about or homepage to align with external knowledge.

11. Blog content not discoverable Medium

    The site lacks a visible /blog/ directory, and FAQ references to blog posts could not be verified.

    What to change: Create a blog section with clear URL structure and include it in the sitemap.

12. Success stories lack structured data Medium

    Enterprise client metrics are presented in prose without Review or Testimonial schema, reducing visibility in AI-generated answers.

    What to change: Add Review or Testimonial schema to success story pages.

What's working

• Deep product pages render rich server-side content — Product pages like LXP, Maestro, and Skills+ contain 1,100-1,600 words of server-side text, making them indexable by crawlers.
• Domain verified with OpenAI and Anthropic — DNS TXT records include openai-domain-verification and anthropic-domain-verification tokens, enabling retrieval-augmented generation access.
• Organization schema present on all pages — Every page includes a JSON-LD Organization block with sameAs links and employee count, providing consistent brand identity.
• FAQ sections with real Q&A content — Product pages include FAQ sections with substantive questions and answers, which can be used for FAQ schema.
• Clean heading structure on product pages — Pages have single H1s and logical H2/H3 hierarchies, aiding crawler understanding.
• Strong external validation signals in success stories — The site lists 8 enterprise clients with specific metrics and claims 300+ clients reaching 10 million users, providing social proof.
• No robots.txt blocks for AI bots — The robots.txt does not disallow any AI bots, allowing them to crawl the site freely.
• WebSite schema includes SearchAction — The site's WebSite schema includes a SearchAction, enabling potential rich search results.