zoomrx.com — AI Site Grade

Analysis

ZoomRx: AI Crawlers See Everything, But LLMs Remember a Different Company

The cold LLM knowledge about ZoomRx is stale by roughly 3-5 years — it describes a "ZoomRx survey tool" and "ZoomRx Decode" product that no longer exist on the site, cites a 2013 founding date (the site says 2010), and misses the entire AI-platform transformation (Sagan, Ferma, Ferma Agents) that now defines the brand's positioning.

Crawler Access

All 11 AI crawlers tested (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, Perplexity-User, and a browser baseline) receive identical 200 responses with full 185KB HTML from Cloudflare. No UA-based blocking, no JS shell, no thin content for bots. The robots.txt is a bare HubSpot default with no AI-bot rules whatsoever — no disallow, no allow, no crawl-delay for any AI user-agent. The llms.txt returns 404, meaning there is no AI-friendly content map. The sitemap.xml is present and well-formed with 86 URLs.

Cold-Knowledge Gap

The LLM prior describes ZoomRx as a "mid-tier life sciences consultancy" known for a survey tool and "ZoomRx Decode" analytics. The actual site describes a vertically integrated AI platform company with 450+ employees, two proprietary AI platforms (Sagan for conversational market research, Ferma for competitive intelligence), 700M+ benchmark datapoints, and a 60,000-person HCP panel. The founding date discrepancy (2013 vs 2010) and the complete absence of Sagan/Ferma from the model's memory means AI engines answering "what is ZoomRx" without live retrieval will describe a materially different, smaller, less technologically advanced company.

Schema Posture

The homepage carries a rich Organization + WebSite @graph with founding date, employee count, knowsAbout array, and social profiles. The about-us page adds an FAQPage with five well-structured Q&A pairs. The Ferma and Sagan offerings pages carry Service, FAQPage, HowTo, and comparison-table schema. Blog posts use BlogPosting and Article types with author, date, and publisher. However, the homepage JSON-LD states minValue: 100 employees while the about-us page states 450+ — a contradiction that will confuse knowledge-graph consumers. The Ferma page's Service schema lists the provider founding date as 2007, conflicting with the homepage and about-us pages which say 2010.

External Signals

The newsroom lists press coverage in Fierce Pharma, Business Wire, GlobeNewswire, MM+M, and eMarketer — all from 2025-2026, indicating active PR. The blog contains direct competitor comparisons (Kantar vs ZoomRx, Ipsos vs ZoomRx, ZS Associates vs ZoomRx, Verilogue vs ZoomRx) with detailed comparison tables and FAQ sections — strong answer-format signals that AI search engines can surface as rich results. However, web searches for "ZoomRx" returned zero organic results from DuckDuckGo, suggesting limited off-domain discoverability outside direct press mentions.

Surprising Findings

The site is rich in AI-friendly content (comparison tables, FAQ schema, detailed service pages, 50+ blog posts) but has no llms.txt, no AI-bot-specific robots.txt directives, and contradictory employee/founding-date schema across pages. The cold LLM gap is unusually large for a company that positions itself as an AI-first firm — the model knows nothing about the AI platforms that are the centerpiece of the site's messaging. The Ferma page references "17 years of pharma domain expertise" while the homepage says "15+ Years of Domain Expertise" — another internal inconsistency.

Findings

1. Cold LLM knowledge describes a 3-5 year old version of ZoomRx High

   LLMs recall ZoomRx as a mid-tier consultancy with a survey tool and 'Decode' product, missing the entire AI-platform transformation (Sagan, Ferma) and citing a 2013 founding date instead of 2010.

   What to change: Publish an llms.txt file summarizing the company's current AI platforms and key facts. Ensure all pages use consistent schema with the correct founding date (2010) and employee count (450+).

2. Homepage schema states 100 employees while about-us page says 450+ High

   The homepage JSON-LD Organization schema lists numberOfEmployees with minValue 100, but the about-us page text states 450+ employees. This contradiction will confuse knowledge graphs.

   What to change: Update the homepage schema to reflect the correct employee count (450+) and ensure consistency across all pages.

3. Founding date varies across pages: 2010, 2013, and 2007 High

   The homepage and about-us page state founding year 2010, the Ferma Service schema lists 2007, and the LLM prior recalls 2013. This inconsistency undermines trust in structured data.

   What to change: Standardize the founding date to 2010 across all schema and page content. Remove the 2007 date from the Ferma page.

4. No llms.txt file published Medium

   The site returns a 404 for llms.txt, missing an opportunity to provide AI crawlers with a curated content map and key facts.

   What to change: Create an llms.txt file listing the most important pages (Sagan, Ferma, about-us, blog) and a brief company description.

5. Robots.txt has no AI-bot rules Low

   The robots.txt is a bare HubSpot default with no directives for AI crawlers (GPTBot, ClaudeBot, etc.), leaving their access entirely to default behavior.

   What to change: Add explicit allow rules for AI crawlers to ensure continued access and signal AI-friendliness.

6. Zero organic search results for brand queries on DuckDuckGo Medium

   Web searches for 'ZoomRx' and related terms returned no organic results, indicating limited off-domain discoverability outside direct press mentions.

   What to change: Invest in SEO and backlink strategy to improve organic search presence for brand and product terms.

7. Ferma page claims 17 years of pharma expertise, homepage says 15+ Medium

   The Ferma page states '17 years of pharma domain expertise' while the homepage says '15+ Years of Domain Expertise', creating another internal inconsistency.

   What to change: Harmonize the number of years across all pages to match the founding date (2010) or use a consistent phrasing.

What's working

• All 11 AI crawlers receive full HTML content without blocking — Every tested AI bot gets a 200 response with complete 185KB HTML from Cloudflare, with no UA-based blocking or JavaScript shells.
• Homepage and key pages have rich Organization, FAQPage, and Service schema — The homepage uses a detailed Organization+WebSite @graph with founding date, employee count, and knowsAbout. About-us has FAQPage, and product pages have Service and HowTo schema.
• Blog posts feature detailed competitor comparison tables with FAQ schema — Posts comparing ZoomRx to Kantar, Ipsos, ZS Associates, and Verilogue include structured comparison tables and FAQ sections, strong signals for AI search rich results.
• Newsroom lists recent press coverage from major pharma publications — The newsroom page shows press mentions in Fierce Pharma, Business Wire, GlobeNewswire, MM+M, and eMarketer from 2025-2026, indicating active PR efforts.
• Sitemap.xml is well-formed with 86 URLs — The sitemap is accessible and contains 86 URLs, helping crawlers discover all important pages.
• Blog includes detailed launch post for Sagan AI platform — A blog post announces the Sagan platform with 345 words of content, providing a canonical source for AI crawlers to learn about the product.