AI-powered evaluation using the Model Context Optimization BS Detection Framework, based solely on publicly available website content.
Based on 2381 businesses audited.
Unclear / Mixed / Unclassifiable Industry BS: Savills (savills.com)
Savills is a substance-first enterprise that uses a thin layer of corporate polish to wrap a massive, high-performance data engine. This is the gold standard for how global advisory firms should handle the signal-to-substance ratio. The BS is superficial; the data is foundational.
Replace sentiment-based H1 headings with results-oriented alternatives such as ‘Data-Driven Property Advisory Across 70 Countries’. Implement Person schema for lead analysts to verify the ‘42,000 experts’ claim at the individual level. Link the ‘9 reviews’ to a third-party verification source to eliminate the minor trust-theatre vulnerability. Expand the ‘People’ page from a simple heading to a directory of verifiable professional credentials.
The site exhibits a dual nature where primary H1 headings are often fluff-heavy, such as ‘Here to help you thrive’ and ‘Making your progress our priority’. However, the body text and H3 research abstracts are remarkably dense with substance, citing ‘vacancy stable at 9.4%’ and ‘138% increase in SCPI investment volumes’. The transition from vague H1s to hyper-specific data points creates a strong substance-to-signal ratio once the user moves past the hero sections.
When your heading hierarchy collapses, AI cannot determine where one idea ends and the next begins. Run a Semantic HTML Machine Readability Audit to see how your structure is actually chunked by LLMs.
There is minimal semantic drift between the high-level promises and the sub-page evidence; the homepage promise of ‘world-class research’ is directly supported by the Research sub-page which indexes 1,787 results. Unlike most ‘advisory’ sites that offer generic blogs, Savills provides granular, dated reports like the ‘2026 Global Talent Cities Index’. The messaging is consistent across the 42,000-expert claim and the global office distribution list.
Move beyond vague agency reporting and visualize your surgical implementation plan. Order an Executive SEO Strategy and stop relying on superficial keyword tracking.
The site reports a review_count of 9 but provides no direct proof_links_count to external platforms like Trustpilot or Google Reviews for these specific ratings, creating a minor trust theatre flag. While the ‘Impacts 2025’ and ‘Q1 2026’ reports act as massive intellectual proof, the social proof layer is under-developed and unverified. The trust_theatre_flag is false, but the lack of third-party validation for customer satisfaction scores is a measurable gap.
The proof density is exceptionally high for the industry, with a significant ratio of specific data points to vague assertions. Across the homepage and research pages, we see specific mentions of ‘138% increase in SCPI investment’ and ‘YOY increase to 5%’ for prime office costs. This forensic level of detail provides a roughly 8:1 ratio of substance to unsubstantiated marketing fluff across the sampled pages.
For a demonstration of entity driven retail architecture, open the Walmart Structured Data audit. View the Walmart Structured Data Audit to see how product, brand, and service entities are reconstructed for AI systems.
The site utilizes standard high-tier real estate jargon such as ‘bespoke solutions’, ‘world-class research’, and ‘pioneering data’. However, these are anchored to a verifiable founding date of 1855 and a physical footprint of 700+ offices, which prevents the content from feeling like a copy-paste template. Proprietary frameworks like the ‘Impacts’ publication provide a unique value proposition that differentiates the brand from generic competitors.
Authority is strongly established at the Organization level through detailed schema_json including founder Alfred Savill and a London headquarters. There is a minor gap at the individual expert level; while 200+ researchers are mentioned, the absence of individual Person schema or sameAs links for key analysts in the crawl limits the verification of their specific professional footprints. The brand’s historical authority is clear, but its current human capital is less transparent.
The marketing tone is surprisingly restrained, relying more on market reporting than ‘guaranteed results’ or aggressive growth claims. The site demonstrates performance by showing the breadth of its data (e.g., ‘1787 results’) rather than just asserting superiority. The disconnect is low because the claim of being a ‘leading property advisor’ is functionally proven by the sheer scale of the 70-country office list provided.
Unclear / Mixed / Unclassifiable Industry BS: Savills (savills.com)
The site perfectly matches the Global Real Estate and Property Advisory industry, providing exactly the type of macroeconomic and sector-specific data expected from a market leader. The presence of specific vacancy rates (9.4%) and investment volume growth (138%) confirms the professional classification and supports the claimed expertise.
When links fail to express hierarchy, the model cannot form clusters or identify primary entities. Examine the Internal Linking Technical Guide and understand how structural signals—not navigation—define your semantic map.
“The BS score of 22 reflects a high-authority site that occasionally relies on corporate cliches in its primary H1 messaging. The score was significantly suppressed by the presence of 1,787 specific research publications and current Q1 2026 reporting data. The primary points lost were in the Information Density and Trust and Proof pillars due to unverified review counts and power-word-heavy headers.”