beqom.com — AI Site Grade

Analysis

beqom.com publishes blog articles dated in the future (May 2026, March 2026) — content that AI crawlers index today as if it were published ahead of time, creating a temporal credibility gap for any LLM that checks the datePublished field.

Crawler Access

All 11 AI crawlers tested — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, Bytespider, Applebot-Extended, ChatGPT-User, Perplexity-User, anthropic-ai — receive identical 200 responses with full HTML content (171KB) from Vercel's edge. No UA-based blocking, no JS shell, no rate-limiting. The robots.txt contains a single User-agent: * rule with no AI-bot-specific directives; llms.txt returns 404. DNS TXT records confirm beqom has proactively verified its domain with Anthropic (anthropic-domain-verification) and Cursor (cursor-domain-verification), indicating awareness of AI-product integrations, yet no llms.txt exists to guide those same crawlers.

Cold-Knowledge Gap

The LLM's prior knowledge describes beqom as a "cloud-based total compensation management platform" founded in 2010, headquartered in Switzerland, with a Gartner Magic Quadrant Leader recognition. The actual site tells a different story: beqom was founded in 2009 (not 2010), is now branded as "beqom PaySuite" — a term absent from the model's knowledge — and centers its positioning on "Intentional AI" (explainable, collaborative, controllable AI agents). The Gartner MQ mention, which the model cites as a key reputational signal, is nowhere on the site — not on the homepage, about page, or awards page. The model also knows nothing about the PayAnalytics acquisition (2023), the CompComplete rebranding (2024), or the CEO transition to Lars Pedersen (2025).

Schema Posture

The homepage and about page carry Organization and WebSite schema with rich address data (4 offices), multiple contact points, and social profiles. However, product pages (/products/beqom-ai, /products/compensation-management, /products/pay-equity-by-payanalytics) contain zero JSON-LD — no Product, SoftwareApplication, or FAQPage schema despite having FAQ sections with visible Q&A content. Blog articles do carry Article schema with datePublished and author, but the dates are problematic: one article is dated May 28, 2026 and another March 20, 2026 — future dates that will confuse any LLM or search engine that trusts the timestamp.

External Signals

External search returns zero results for beqom on Reddit, Gartner reviews, or third-party review sites — a vacuum that leaves AI models dependent entirely on the site's own content and whatever press mentions exist in training data. The site claims "trusted by 40+ S&P 500 companies" and "5 million employees" but provides no verifiable customer logos beyond the 3-4 success stories (Puma, Allianz, Total Energies, Breitling, Lowe's). The trust center references ISO 27001/27017/27018 and SOC2 Type II certifications but links to an external portal (trustcentre.beqom.com) that was not tested.

Findings

1. Blog articles dated in 2026 create temporal credibility gap High

   Two blog articles carry datePublished values of May 28, 2026 and March 20, 2026. AI crawlers and LLMs that trust the timestamp will treat these as future-published content, undermining trust in the site's temporal accuracy.

   What to change: Update the datePublished values on all blog articles to reflect actual publication dates. Implement a content management system that prevents future dates from being set.

2. Product pages lack JSON-LD schema High

   Key product pages for beqom AI, compensation management, and pay equity contain no Product, SoftwareApplication, or FAQPage schema, despite having FAQ sections with visible Q&A content. This limits AI crawlers' ability to understand and surface product information.

   What to change: Add JSON-LD schema of type SoftwareApplication and FAQPage to each product page, including product name, description, application category, offers, and FAQ entries.

3. No llms.txt file to guide AI crawlers Medium

   The site returns a 404 for /llms.txt, despite having verified its domain with Anthropic and Cursor for AI integrations. An llms.txt file would help direct AI crawlers to key pages and provide a concise summary of the site's content.

   What to change: Create an llms.txt file at the root that lists key pages (homepage, product pages, blog, about, trust center) and provides a brief summary of the site's purpose.

4. LLM knowledge lacks PaySuite rebranding and recent milestones Medium

   The LLM's prior knowledge describes beqom as a 'cloud-based total compensation management platform' founded in 2010, but the site now brands as 'beqom PaySuite' founded in 2009. The model is unaware of the PayAnalytics acquisition, CompComplete rebranding, and CEO transition to Lars Pedersen. The Gartner Magic Quadrant mention in the model's knowledge is absent from the site.

   What to change: Prominently feature the PaySuite brand, founding year, and key milestones (acquisitions, leadership changes) on the homepage and about page. Consider adding a press or news section to reinforce recent developments.

5. No external reviews or community presence found Medium

   Web searches for beqom on Reddit, Gartner reviews, and third-party review sites returned zero results. This absence of external signals leaves AI models dependent solely on the site's own content and any press mentions in training data.

   What to change: Encourage customers to leave reviews on Gartner Peer Insights, G2, and other platforms. Engage in relevant Reddit communities and forums to build organic external signals.

6. Customer claims lack verifiable evidence Low

   The site claims 'trusted by 40+ S&P 500 companies' and '5 million employees' but only lists 5 customer logos (Puma, Allianz, Total Energies, Breitling, Lowe's). No case studies or detailed references are provided for the broader claim.

   What to change: Add a customer page with logos, quotes, and case studies for all claimed customers. Provide a verifiable list or link to press releases that substantiate the 40+ S&P 500 claim.

What's working

• All AI crawlers receive full HTML content — All 11 tested AI crawlers receive identical 200 responses with full HTML content from Vercel's edge. No UA-based blocking, JS shells, or rate-limiting is applied.
• Domain verified with Anthropic and Cursor — DNS TXT records show proactive domain verification with Anthropic and Cursor, indicating awareness of AI-product integrations and readiness for AI crawler access.
• Homepage and about page carry rich Organization schema — The homepage and about page include Organization and WebSite schema with detailed address data for 4 offices, multiple contact points, and social profiles, helping AI crawlers understand the entity.
• Blog articles include Article schema with author and date — Blog articles carry Article schema with datePublished and author fields, which helps AI crawlers understand content structure and attribution.
• Trust center references ISO and SOC2 certifications — The trust center page mentions ISO 27001/27017/27018 and SOC2 Type II certifications, which are strong trust signals for enterprise buyers and AI crawlers evaluating credibility.
• Sitemap contains 80 URLs covering key sections — The sitemap lists 80 URLs, including product pages, blog articles, and other key sections, ensuring good discoverability for crawlers.