betterup.com — AI Site Grade

Analysis

BetterUp's AI-visibility posture is best-in-class — but a cold-knowledge gap reveals the site has fully retired its "BetterUp Care" mental-health positioning without the external web catching up.

The site has an unusually sophisticated AI-crawler strategy, yet the brand's external reputation still centers on a product line the site no longer mentions.

Crawler Access

BetterUp runs on Cloudflare behind GoDaddy DNS, serving all 11 tested AI bot UAs (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, anthropic-ai, OAI-SearchBot, Bytespider, Applebot-Extended, Perplexity-User) a full 200 with identical 151,500-byte payloads — no blocks, no JS shells, no UA-based throttling. The robots.txt is explicitly labeled "Optimized for AI Discovery" and grants Allow: / to every major AI crawler while only blocking preview/cache paths and PDFs. An llms.txt exists at a HubSpot-hosted path (/hubfs/llms.txt) with a curated content map covering products, solutions, research, and explicitly excluded sections (careers, legal, legacy D2C). The sitemap.xml contains 2,760 URLs and is referenced in the robots.txt. This is among the most AI-friendly technical setups observed.

Cold-Knowledge Gap

Asked cold, a frontier LLM describes BetterUp as "a coaching and mental health platform" with "BetterUp Care" as a notable product, and recalls a 2023 layoff of ~11% staff plus internal culture concerns. The actual site contains zero mentions of "BetterUp Care" — that product line has been retired. The site now positions itself exclusively as the "Human Transformation Platform" for enterprise workforce performance, with four products (Lead, Manage, Grow, Ready) and a heavy emphasis on AI coaching. The layoff and culture concerns are absent from the site entirely. The cold model also does not know about the August 2025 pure-AI coaching product launch (BetterUp Grow), the acquisitions of Heyday and Practica, or the randomized controlled trial data showing 7-10x outperformance vs. ChatGPT. The Wikipedia page (updated May 2026) does cover these, meaning the knowledge gap is in model training cutoffs, not the public record.

Schema Posture

The homepage carries a rich Organization JSON-LD block with founders, addresses, certifications (SOC 2, ISO 27001), and sameAs links to Wikipedia, G2, and LinkedIn. Product pages (/powered-by-ai, /products/betterup-ai-coaching, /trust-and-security) each embed FAQPage schema with 3-8 Q&A pairs. Blog posts use BlogPosting + BreadcrumbList + FAQPage schema. The /customers and /betterup-labs pages have no JSON-LD at all — a gap for a page that lists 20+ named enterprise case studies and a research lab with 40+ PhDs.

Content Signals

The homepage uses comparison language ("The only platform that..."), lists quantified outcomes (14x ROI, 50% lower turnover, 7.5% revenue increase), and names specific customers (Mercedes-AMG F1, Google, Moderna, USAF). The blog contains substantive, research-backed articles (e.g., "The Workforce Performance Crisis" at 1,678 words with original data from 410,000 employees). FAQ sections appear on product and trust pages but not on the homepage itself. The site has no traditional FAQ page or comparison table page — the FAQ schema is embedded within product pages.

External Signals

Wikipedia lists Meta, Salesforce, Hilton, and the US Air Force as customers — names not all featured on the site's customer page. The Wikipedia article (updated May 2026) covers the 2025 AI coaching launch and acquisitions, but the cold LLM has none of this data. The site's sameAs links to G2 but no G2 reviews were surfaced in search. No Reddit threads appeared in search results. The DNS TXT records show integrations with Google Workspace, SendGrid, Qualtrics, Marketo, Mailchimp, and Greenhouse — a standard enterprise martech stack.

Findings

1. Cold LLM knowledge still references retired "BetterUp Care" product High

   A frontier LLM describes BetterUp as a "coaching and mental health platform" with "BetterUp Care" as a notable product, but the site contains zero mentions of that product line. The brand's external reputation still centers on a retired offering.

   What to change: Proactively publish updated brand positioning and product information via llms.txt, Wikipedia updates, and press releases to align external knowledge with current offerings.

2. Cold LLM recalls 2023 layoffs and culture concerns absent from site Medium

   The cold LLM recalls a 2023 layoff of ~11% staff and internal culture concerns, but the site contains no mention of these events. This negative signal could affect AI-generated summaries.

   What to change: Consider publishing a transparent retrospective or updated company timeline to provide context and counterbalance negative narratives.

3. Cold LLM unaware of 2025 AI coaching launch and acquisitions Medium

   The cold LLM does not know about the August 2025 pure-AI coaching product launch (BetterUp Grow), acquisitions of Heyday and Practica, or randomized controlled trial data showing 7-10x outperformance vs. ChatGPT. Wikipedia covers these, indicating a model training cutoff gap.

   What to change: Ensure llms.txt and sitemap include all recent product pages and press releases; consider submitting updated content to AI training data sources.

4. Customer stories page lacks JSON-LD schema Medium

   The /customers page lists 20+ named enterprise case studies but has no JSON-LD markup. This reduces the chance that AI crawlers extract structured customer evidence.

   What to change: Add Organization or Product schema with customer references, or embed ItemList schema for case studies.

5. Research lab page lacks JSON-LD schema Medium

   The /betterup-labs page describes a research lab with 40+ PhDs but has no JSON-LD markup. This reduces discoverability of research credentials.

   What to change: Add Organization or ResearchOrganization schema with description of the lab and its PhD count.

6. Homepage lacks FAQ schema despite comparison claims Low

   The homepage uses comparison language and lists quantified outcomes but has no FAQPage schema. FAQ schema is only embedded on product pages.

   What to change: Add FAQPage schema to the homepage covering common questions about the platform's differentiators.

7. No dedicated FAQ page exists on the site Low

   The site has no traditional FAQ page; FAQ schema is embedded within product pages. A centralized FAQ page could improve AI crawler extraction of common questions.

   What to change: Create a dedicated FAQ page with comprehensive Q&A pairs and FAQPage schema.

8. No comparison table page for competitive positioning Low

   The site lacks a comparison table page that contrasts BetterUp with alternatives. Such a page could help AI crawlers extract competitive differentiators.

   What to change: Create a comparison page with structured data (e.g., Table schema) comparing BetterUp to competitors.

9. Wikipedia lists customers not featured on site's customer page Medium

   Wikipedia mentions Meta, Salesforce, Hilton as customers, but these names are not all featured on the /customers page. This inconsistency may confuse AI crawlers.

   What to change: Update the customer page to include all major named customers, or ensure Wikipedia reflects only publicly disclosed customers.

10. G2 reviews not surfaced in web search Low

    Despite a sameAs link to G2, no G2 reviews appeared in search results. This may indicate low review volume or poor indexing.

    What to change: Encourage more G2 reviews and ensure the G2 profile is properly linked and indexed.

11. No Reddit discussions found about BetterUp coaching Low

    A search for site:reddit.com BetterUp coaching review returned zero results, indicating low community discussion on Reddit.

    What to change: Consider engaging with relevant subreddits or encouraging user discussions to build external signals.

What's working

• All 11 tested AI bots served full HTML with no blocks — BetterUp serves all major AI crawlers a full 200 response with identical payloads, no UA-based throttling, and no JavaScript shells. This ensures maximum content accessibility for AI training and inference.
• Robots.txt explicitly optimized for AI discovery — The robots.txt is labeled "Optimized for AI Discovery" and grants Allow: / to every major AI crawler while only blocking preview/cache paths and PDFs.
• llms.txt published with curated content map — An llms.txt file exists at a HubSpot-hosted path with a curated content map covering products, solutions, research, and explicitly excluded sections.
• Homepage carries rich Organization JSON-LD with certifications — The homepage includes Organization schema with founders, addresses, SOC 2 and ISO 27001 certifications, and sameAs links to Wikipedia, G2, and LinkedIn.
• Product pages embed FAQPage schema with 3-8 Q&A pairs — Pages like /powered-by-ai, /products/betterup-ai-coaching, and /trust-and-security each embed FAQPage schema, helping AI crawlers extract common questions.
• Blog posts use BlogPosting + BreadcrumbList + FAQPage schema — Blog articles like "The Workforce Performance Crisis" include BlogPosting, BreadcrumbList, and FAQPage schema, enhancing structured data richness.
• Blog contains research-backed articles with original data — The blog features substantive articles like "The Workforce Performance Crisis" at 1,678 words with original data from 410,000 employees, providing high-quality content for AI training.
• Wikipedia page updated May 2026 covers recent developments — The Wikipedia article covers the 2025 AI coaching launch, acquisitions, and other recent developments, providing a reliable external knowledge source.