Engagement Foundation Review

Anthropic Audit Foundation

Before we run the audit, we need to make sure we're asking the right questions about the right competitors to the right buyers. This document presents what we've learned about Anthropic's market — your job is to tell us what we got right, what we got wrong, and what we missed.

Prepared July 23, 2026
anthropic.com
Frontier AI / Enterprise LLM Platform
GEO Readiness

Where You Stand Today

Before we measure how often AI answer engines cite Claude in the frontier-AI and enterprise-LLM space, these three signals tell us whether those crawlers can reach, refresh, and cleanly parse your content in the first place.

Technical Readiness
Needs Attention
One high-severity finding and zero critical blockers. Anthropic's most-cited evergreen explainers — "Building effective agents," "Introducing the Model Context Protocol," "Raising the bar on SWE-bench Verified" — carry no visible refresh date. Three medium structural issues follow (off-domain sitemap redirects, a JS-only Economic Index dashboard, non-semantic hub headings).
Content Freshness
At Risk
Critical split: 25 content-marketing pages (news, engineering, research) average 0.24 freshness — 17 carry a date older than 6 months and 12 older than a year, including the evergreen agent/MCP/SWE-bench guides. Product pages are current by contrast (0.81 avg; 6 of 7 updated within 90 days), but AI-cited content runs 25.7% fresher than typical results (Ahrefs, August 2025), and the stale evergreen guides are exactly where citations concentrate. 2 product pages carry no detectable date — verify manually.
Crawl Coverage
Needs Attention
robots.txt confirmed open to every major AI crawler — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and Googlebot are all allowed. But the anthropic.com sitemap still advertises /product/* URLs (Claude Code, Enterprise, Security) that 308-redirect off-domain to claude.com, wasting crawl budget and splitting citation authority across two domains.
Executive Summary

What You Need to Know

AI search is reshaping how buyers in the frontier-AI and enterprise-LLM space discover and evaluate solutions — increasingly, the first shortlist a CTO or CISO sees is assembled by an assistant, not a search page. Companies that establish citation visibility now gain a first-mover advantage that compounds: early citations become self-reinforcing as AI platforms learn which domains to trust. Anthropic enters this shift from an unusually strong position — a widely recognized frontier lab that is, notably, also the vendor behind one of the very crawlers deciding who gets cited.

This Foundation Review presents three things we need you to validate before the audit runs: the competitive landscape that shapes how we construct head-to-head queries, the buyer personas that determine which search intents we test, and the Layer 1 technical baseline that determines whether AI platforms can access and parse your content at all. Think of it as a shared read of your market — here's what we've concluded, and where we need your judgment to confirm or correct it.

The validation call is a working session with real stakes: your answers decide which inputs drive the buyer query set across the AI platforms we select. Two kinds of decisions come out of it — input validation (are the right entities in the right tiers, and is every persona a real buyer?) and engineering triage (which technical fixes can start before results come back?). The Pre-Call Checklist below aggregates every one of those questions into a single printable page.

TL;DR — Action Items
  • 🟡 High: High-value evergreen content sits outside the AI citation freshness window — Content should add a visible "Last updated" date and a quarterly refresh cadence to the ~10 evergreen agent/MCP/context-engineering/SWE-bench explainers that still draw traffic.
  • ✅ Start Now: Regenerate the anthropic.com sitemap — Engineering can drop the /product/* and /engineering/* URLs that 308-redirect off-domain to claude.com without waiting for the validation call.
  • 🟣 Validate at the Call: Price/Cost Efficiency and Multimodal Generation are the only two features rated weak — If premium Opus pricing or the lack of native image/audio/video generation is actually losing deals, we build value-vs-quality query clusters against DeepSeek, Llama, and Nova; if buyers pay for frontier quality, we press differentiation instead.
  • 🟣 Validate at the Call: Sofia Marchetti (Director of AI Enablement) is the only inferred, medium-confidence persona — If enterprise deals are really a CTO-plus-CISO purchase with no separate innovation champion, we drop her adoption/ROI query cluster entirely.
  • 📋 Validation Call: Confirm whether the audit measures anthropic.com, claude.com, or both — The commercial surface (pricing, product, platform docs) has migrated to claude.com, so this single decision determines which domain we measure citation visibility for.
How This Works

Reading This Document

Purpose This is the foundation for your GEO visibility audit in the frontier-AI / enterprise-LLM category. It captures our outside-in read of Anthropic's market — competitors, buyers, capabilities, and frustrations — plus the technical state of your site. Everything here becomes an input to the buyer queries we run against AI platforms. It is deliberately not a content-gap analysis; that requires the audit's response data and comes later.

Your Job Confirm what's right, correct what's wrong, and tell us what's missing. The purple boxes throughout the document are the specific questions where your answer changes what the audit does. You know your deals better than any outside analysis can — treat those as the agenda.

Confidence Badges Every entity carries a confidence badge. High means directly sourced (from your site, reviews, or public landscape data). Medium means inferred and worth a closer look. Scan for medium badges — those are where we most need your validation.

Company Profile

Who We're Auditing

Anthropic

Company name Anthropic High
Domain anthropic.com
Name variants Anthropic PBC · Anthropic AI · Claude · Claude AI · Claude by Anthropic · Anthropic Claude
Category Frontier AI / enterprise LLM platform
Segment Enterprise
Key products Claude (Opus / Sonnet / Haiku) · Claude.ai (Free / Pro / Max) · Claude Team & Enterprise · Claude Developer Platform (API) · Claude Code
Positioning Claude frontier models delivered as an enterprise AI assistant, a developer API, and agentic coding tooling — positioned on safety, reliability, and reasoning quality.

Anthropic's identity splits two ways: the company (Anthropic) versus the product (Claude), and — more consequentially — the commercial surface (pricing, product pages, platform docs) has migrated from anthropic.com to claude.com / platform.claude.com, while this audit crawled anthropic.com. Do buyers and AI engines resolve "Anthropic" and "Claude" to the same entity, and which domain should the audit measure — anthropic.com, claude.com, or both? If citations now accrue to claude.com, auditing anthropic.com alone measures a research-and-news surface, not the commercial one where deals are won.

Buyer Personas

Who Evaluates Claude

5 personas: 3 decision-makers, 1 evaluator, 1 influencer — the buyers whose distinct search intent drives every query the audit generates.

Critical Review Area Personas drive the entire query set — each buyer searches differently, so a wrong role or influence level sends the audit chasing the wrong intent. Scrutinize the influence badges and the one inferred persona below before the call.

Data Sourcing Name, role, department, seniority, influence, veto power, and technical level are pulled directly from the KG. Role descriptions, buying jobs, and query focus areas are synthesized from those fields to show how each persona would actually search — validate the synthesis, not just the labels. Four of five personas are review-mined at high confidence; one (Director of AI Enablement) is LLM-inferred at medium confidence.

Marcus Thorne
CTO / Chief AI Officer
Decision-maker High
Owns the enterprise AI strategy and the platform decision — the executive accountable for whether Claude ships into production and delivers value the board can see.
Veto power: Yes — final approver on the platform choice.
Technical level: High
Primary buying jobs: Sets the AI mandate, evaluates model quality and long-horizon reliability, defends ROI to leadership.
Query focus areas: "best enterprise LLM," "most reliable model for production," "Claude vs OpenAI vs Gemini," reasoning and coding benchmarks.
Source: Review mining (executive titles in reviews & case studies)

Is the CTO and the Chief AI Officer one budget-holder in your deals, or two seats at the table? If AI strategy sits with a separate Chief AI Officer, we split Marcus into two personas with different intents — platform reliability vs. a transformation mandate.

Danielle Osei
CISO / Head of Information Security
Decision-maker High
Guards the security and risk posture — decides whether an AI vendor can be trusted with proprietary data and can satisfy audit before anything reaches production.
Veto power: Yes — can block a deal on security or compliance grounds.
Technical level: High
Primary buying jobs: Runs the security questionnaire, validates data-use and retention terms, brings shadow AI under governance.
Query focus areas: "does Claude train on my data," "enterprise AI governance," "SOC 2 / HIPAA AI," "AI data residency."
Source: Review mining (security/risk reviewer titles)

Does the CISO hold a hard veto in your AI deals, or sign off after IT? If security can independently kill a deal, we weight governance and residency queries as gating criteria, not evaluation factors — which reorders the whole security query cluster.

Ravi Krishnan
VP of Engineering / Head of AI Platform
Evaluator High
Runs the technical evaluation — puts the model against the real codebase and API before anyone signs, and owns the integration afterward.
Veto power: No — a strong evaluator who recommends but does not sign.
Technical level: High
Primary buying jobs: Runs POCs, benchmarks coding-agent performance on the real monorepo, tests API reliability and tool use.
Query focus areas: "best AI coding agent," "Claude Code vs GitHub Copilot," "LLM API reliability," "MCP tool use."
Source: Review mining (engineering leadership titles)

Does the VP of Engineering control the platform budget, or only recommend? If Ravi actually signs the contract, we reclassify him as a decision-maker and add validation-stage queries on API reliability and coding-agent performance.

Sofia Marchetti
Director of AI Enablement / Head of Innovation
Influencer Medium
Drives adoption and ROI across the organization — the champion who runs the rollout and proves the value, but not the one who signs the contract.
Veto power: No — advisory influence on adoption and enablement.
Technical level: Medium
Primary buying jobs: Runs pilots, measures adoption and ROI, builds internal enablement and change management.
Query focus areas: "enterprise AI adoption ROI," "Claude for teams," "AI rollout best practices," change-management framing.
Source: LLM inference (medium confidence — the least-sourced persona in the set)

Is there a distinct AI Enablement / Innovation buyer in your deals, or is this really a CTO-plus-CISO purchase with no separate champion? Sofia is the only inferred persona — if she isn't real, we drop her adoption/ROI query cluster entirely rather than reporting visibility for a buyer who never shows up.

Gregory Vaughn
General Counsel / Head of Compliance
Decision-maker High
The legal and compliance gate — signs off on IP indemnification, data-use terms, and regulatory exposure before deployment, even though day-to-day influence is advisory.
Veto power: Yes — can block on legal, IP, or regulatory terms.
Technical level: Low
Primary buying jobs: Reviews contracts and copyright indemnification, validates regulatory compliance, approves data-handling terms.
Query focus areas: "AI copyright indemnification," "who owns AI output," "AI GDPR compliance," data-residency and retention terms.
Source: Review mining (legal/compliance titles)

Does Legal/Compliance gate the deal on IP indemnification and data-use terms, or only review the paper at the end? If the GC can veto, we promote copyright-indemnification and data-privacy queries into the must-answer set rather than treating them as secondary.

Missing Personas? These roles sometimes appear in enterprise frontier-AI deals — do they show up in yours? A Head of Data / ML Platform (if model-infrastructure decisions sit apart from app-layer AI and warrant their own query cluster); a Procurement / Vendor-Risk lead (if a formal security-and-legal questionnaire gates regulated deals); a line-of-business economic buyer — Head of Support, Sales Ops, or Research (if a specific Claude use case is funded outside central IT). Who else shows up in your enterprise evaluations?

Competitive Landscape

Who You're Compared Against

5 primary + 5 secondary competitors — tier assignments determine which vendors get head-to-head query matchups versus category-awareness coverage.

Why Tiers Matter Tier assignments decide which vendors get direct-comparison queries like "Claude vs OpenAI for enterprise" or "best LLM vs alternatives," versus lighter category-awareness coverage — roughly 6–8 head-to-head queries per primary competitor. Two primaries carry medium confidence: Microsoft Copilot (which may compete on M365 bundling more than model quality) and Cohere (whose presence in your regulated-enterprise deals we can't confirm from public data). Moving either to secondary would shift those queries out of the direct-comparison set.

Primary Competitors

OpenAI

PrimaryHigh
openai.com
The most direct and formidable competitor — broadest consumer and enterprise adoption, largest developer ecosystem, strong coding and multimodal breadth. Anthropic differentiates on safety, lower hallucination, and long-horizon agentic reliability; OpenAI leads on brand ubiquity, feature velocity, and native image/audio/video generation.
Source: Category landscape listing

Google Gemini

PrimaryHigh
gemini.google.com
Google's frontier model family with deep Workspace integration, massive context handling, and strong real-time/search grounding. Wins where buyers are already on Google Cloud/Workspace and need native web access; Anthropic counters on reasoning quality, coding, and a safety-first reputation.
Source: Category landscape listing

Microsoft Copilot

PrimaryMedium
microsoft.com/copilot
The default enterprise AI assistant for Microsoft-shop organizations, bundled into M365 and Azure with existing procurement and identity. Competes hard for the enterprise seat budget Claude Enterprise targets; Anthropic wins on raw model quality and coding, but faces Microsoft's distribution and bundling advantage.
Source: Category landscape listing

Mistral AI

PrimaryHigh
mistral.ai
Europe's leading frontier lab, positioned on open-weight options, data sovereignty, and on-prem/private deployment. Attractive to EU and regulated buyers who need self-hosting — a deployment mode Anthropic does not offer — though generally rated below Claude on frontier reasoning and coding.
Source: Category landscape listing

Cohere

PrimaryMedium
cohere.com
Enterprise-first LLM provider focused on RAG, enterprise search, and private/VPC deployment for security-conscious buyers. Overlaps with Anthropic in regulated enterprise deals; narrower consumer presence and less frontier-benchmark visibility than Claude.
Source: Category landscape listing

Secondary Competitors

DeepSeek

SecondaryHigh
deepseek.com
Chinese open-weight lab delivering near-frontier benchmark scores at a fraction of the token cost. Shows up in cost-sensitive and high-volume evaluations as the value alternative; raises data-residency and governance concerns that push regulated buyers back toward Claude.
Source: Category landscape listing

Meta Llama

SecondaryMedium
llama.com
The leading open-weight model family, chosen by teams that need self-hosting, fine-tuning control, and no per-token cost. Competes with Claude on build-your-own and on-prem use cases but requires in-house ML capability Anthropic's managed offering removes.
Source: Category landscape listing

Amazon Nova

SecondaryMedium
aws.amazon.com/nova
Amazon's in-house model family on Bedrock, the natural default for AWS-committed enterprises with security and compliance baked in. Notably, Bedrock is also a distribution channel for Claude — so the competition is model-vs-model inside the same marketplace rather than platform-vs-platform.
Source: Category landscape listing

xAI Grok

SecondaryMedium
x.ai
Fast-moving frontier entrant with real-time data access via X and aggressive benchmark claims. Appears in AI recommendations and general-purpose assistant comparisons but has limited enterprise governance credibility relative to Claude.
Source: Category landscape listing

Perplexity

SecondaryMedium
perplexity.ai
Answer-engine positioned on real-time, cited web search rather than a general model platform. Overlaps with Claude for knowledge-work and research-assistant use cases where up-to-date sourcing matters; not a foundation-model API competitor.
Source: Category landscape listing

Three tier questions decide the head-to-head set. (1) Does Microsoft Copilot show up as a model-quality rival in your deals, or as M365 bundling pressure at the seat level? (2) Does Cohere actually appear in your regulated-enterprise evaluations, or is it a category listing rather than a real rival? Moving either to secondary shifts ~6–8 head-to-head queries per vendor out of the direct-comparison set. (3) Since Bedrock distributes Claude, should Amazon Nova be tested model-vs-model rather than platform-vs-platform? And are any names here missing entirely — the set skews to public landscape listings rather than your own win/loss data.

Feature Taxonomy

Capabilities Buyers Evaluate

11 buyer-level capabilities mapped — these determine which capability queries the audit tests and where competitive differentiation is pressed.

Reasoning & Response Quality Strong High

Consistently accurate, nuanced answers on complex, high-stakes work without hand-holding — a model I can trust for analysis, not just chat.

Agentic Coding & Software Development Strong High

An AI that can navigate a large codebase, run multi-step engineering tasks end to end, and actually ship working code with an agent like Claude Code.

AI Safety, Alignment & Low Hallucination Strong High

A model that says "I don't know" instead of confidently making things up, and won't go off the rails on sensitive prompts.

Large Context Window & Long-Document Handling Strong High

Feed it entire contracts, codebases, or research corpora at once and have it reason across all of it without losing the thread.

Enterprise Security, Admin & Governance Strong High

SSO/SCIM, audit logs, role-based access, custom data retention, and a compliance API so IT can actually govern how employees use it.

Data Privacy & IP Protection Strong High

Guarantee my business data isn't used to train the model, and stand behind copyright indemnification so legal will sign off.

Developer Platform, API & Tool Use Strong High

A clean, reliable API with tool use, MCP, and SDKs my team can build agents on without fighting the platform.

Deployment & Data-Residency Flexibility Moderate Medium

Run it where our data and compliance rules require — our cloud, our region, ideally on-prem or air-gapped.

Product Ecosystem & Native Integrations Moderate Medium

It should already plug into the tools we live in — Slack, Microsoft 365, our IDE, our data — not sit in a separate tab.

Multimodal Generation (Image / Audio / Video) Weak Medium

Generate and understand images, voice, and video — not just read them — so one vendor covers all our modalities.

Price / Cost Efficiency at Scale Weak Medium

Predictable, competitive token costs for high-volume workloads — the top-tier model shouldn't blow up our budget versus cheaper alternatives.

Prioritization Seven capabilities are rated strong: Reasoning & Response Quality, Agentic Coding & Software Development, AI Safety, Alignment & Low Hallucination, Large Context Window & Long-Document Handling, Enterprise Security, Admin & Governance, Data Privacy & IP Protection, and Developer Platform, API & Tool Use. The audit tests all 11 capabilities, but competitive-differentiation queries will emphasize 3. Which of these seven best represents where Anthropic actually wins deals?

The two weak ratings carry the most consequence: if premium Opus pricing is genuinely losing deals to DeepSeek, Llama, and Nova, we build value-vs-quality queries; if buyers pay the premium for frontier quality, we press differentiation instead. Likewise, is the lack of native image/audio/video generation a real deal-loss factor or an accepted trade-off? Are Deployment Flexibility and Ecosystem Integrations truly only moderate versus Microsoft and Google — and should any strong capabilities merge (e.g., Enterprise Governance + Data Privacy & IP, which buyers often evaluate as one security conversation)?

Pain Points

The Frustrations Behind the Search

10 pain points: 6 high, 4 medium severity — the buyer frustrations whose first-person language becomes the phrasing of the queries we run.

AI produces confident but wrong answers in high-stakes workflows High High

"I can't put this in front of clients or auditors when it makes things up with a straight face — one bad answer and it's a liability."
Personas: CTO / Chief AI Officer, General Counsel, Director of AI Enablement

Fear that proprietary data and code will be retained or used for training High High

"How do I know our source code and customer data aren't quietly training someone else's model? Legal will not sign off on that."
Personas: CISO, General Counsel, CTO / Chief AI Officer

Ungoverned shadow AI leaves security with no visibility or control High High

"Half my company is already pasting confidential docs into random chatbots and I have zero visibility into it."
Personas: CISO, CTO / Chief AI Officer

AI dev tools fail on large, real-world codebases and multi-step tasks High High

"The demos are great, but on our actual monorepo the AI loses the plot after two files — I need it to handle real work."
Personas: VP of Engineering, CTO / Chief AI Officer

Regulated buyers can't prove compliance, residency, and auditability High High

"Before this touches a single record I have to satisfy our auditors on residency, retention, and access logs — most vendors can't even answer the questionnaire."
Personas: CISO, General Counsel

Premium token costs become unpredictable and expensive at scale High High

"Our AI bill tripled last quarter and finance is asking why we're paying premium rates when a model at a tenth the price scores almost as well."
Personas: CTO / Chief AI Officer, Director of AI Enablement

Managed API-only vendors can't meet on-prem / air-gapped requirements Medium Medium

"Our security posture says the model has to run inside our walls — if you can't self-host, we can't even start the conversation."
Personas: CISO, CTO / Chief AI Officer

Tools that don't connect to existing systems create friction and low adoption Medium Medium

"If people have to copy-paste between a chatbot and the tools they actually work in, they just stop using it."
Personas: Director of AI Enablement, VP of Engineering

Champions can't prove measurable ROI or drive durable adoption Medium Medium

"We bought seats for the whole team and six months later I can't show leadership a number that says it was worth it."
Personas: Director of AI Enablement, CTO / Chief AI Officer

Over-refusals and unannounced behavior changes break production workflows Medium Medium

"The model suddenly refuses a request it handled fine last week, and now a production pipeline is broken with no warning."
Personas: VP of Engineering, Director of AI Enablement

Is "runaway token cost" really high-severity for your enterprise segment, or does frontier quality neutralize price objections at the deals you actually win? Severity ordering decides which frustrations we test first. And are we missing category-specific pains — model-deprecation / version-migration churn (models retired on a schedule, breaking pipelines), evaluation difficulty (buyers can't objectively benchmark models before purchase), or vendor lock-in / portability fear? Which of these actually stalls your deals?

Layer 1 · Site Findings

Technical Baseline

What our crawl found on anthropic.com — the technical fixes engineering can act on now, independent of the audit results.

For Engineering Good news first: robots.txt is confirmed open to every major AI crawler — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and Googlebot are all allowed, so nothing is blocked at the door. No critical blockers were found. Engineering should focus on one high-severity item — the stale evergreen explainers with no refresh date — and three medium structural fixes: the sitemap URLs that 308-redirect off-domain to claude.com, the JS-only Economic Index dashboard, and the non-semantic hub-page headings. All are actionable now.

🟡 High-value evergreen content sits well outside the AI citation freshness window

What we found: Across the 25 inventoried content-marketing pages (news, engineering, research), the category freshness average is 0.25 — 17 of 25 carry a visible date older than 180 days and 12 are older than 365 days. This includes Anthropic's most-linked evergreen technical assets — "Building effective agents" (Dec 19, 2024), "Introducing the Model Context Protocol" (Nov 25, 2024), "Raising the bar on SWE-bench Verified" (Jan 6, 2025) and "How we built our multi-agent research system" (Jun 13, 2025) — none of which show any "last updated" refresh signal even though they remain live, frequently-cited reference pages.

Why it matters: AI answer engines concentrate citations on recently-updated pages. ConvertMate found 76.4% of ChatGPT's most-cited pages were updated within the prior 30 days, and Ahrefs measured AI-cited content as 25.7% fresher on average than typical results. Anthropic's evergreen agent/MCP/coding explainers are exactly the pages that surface for "how to build AI agents," "what is MCP," and coding-evaluation queries; without a visible refresh date they lose freshness credit to newer third-party write-ups even when Anthropic is the primary source.

Business consequence: In the frontier-AI category, queries like "how to build AI agents," "what is the Model Context Protocol," and "best model for SWE-bench coding" surface exactly these explainers — when Anthropic's own primary-source guides show no refresh date, newer third-party write-ups collect the freshness credit and the citation even though Anthropic authored the underlying work.

Recommended fix: Add a visible "Last updated" date and a lightweight quarterly refresh cadence to the ~10 evergreen engineering/research explainers that still attract traffic (agents, MCP, context engineering, SWE-bench, multi-agent). Update model/version references inline and re-timestamp. Leave dated launch announcements as-is, but keep the evergreen guides current.

Impact: High Effort: 1–2 weeks Owner: Content Affected: ~25 content-marketing pages under /engineering, /research, /news

🔵 Sitemap lists commercial URLs that 308-redirect off-domain to claude.com

What we found: www.anthropic.com/sitemap.xml still lists product URLs — /product/claude-code, /product/enterprise, /product/security, /product/claude-cowork — and at least one engineering URL (/engineering/claude-code-best-practices) that now return HTTP 308 Permanent Redirect to a different registrable domain (claude.com, code.claude.com). The bulk of Anthropic's commercial surface (pricing, solutions, API/platform docs, customers) has migrated to claude.com / platform.claude.com and is no longer served on the audited anthropic.com domain, yet the sitemap continues to advertise the moved paths.

Why it matters: Sitemap entries that redirect cross-domain waste crawl budget and split link/citation authority: inbound links and prior citations pointing at anthropic.com/product/* bounce to a second domain, so neither domain accumulates the full signal. AI crawlers that resolve the sitemap encounter redirect chains instead of canonical 200 responses, reducing the share of Anthropic's own commercial content that is cleanly indexable under the brand's primary domain.

Business consequence: When a buyer asks "Claude Enterprise pricing" or "Claude Code vs Copilot," crawlers following the anthropic.com sitemap hit 308 redirects instead of a clean 200 — so neither domain accumulates the full citation authority for Anthropic's commercial answers, and competitors with a single consolidated domain gain the edge.

Recommended fix: Regenerate the anthropic.com sitemap to drop paths that permanently redirect off-domain, and either (a) keep canonical commercial content reachable under anthropic.com, or (b) if the claude.com split is intentional, publish and cross-submit a claude.com sitemap and add rel=canonical / consistent internal linking so authority consolidates on the intended domain.

Impact: Medium Effort: 1–3 days Owner: Engineering Affected: sitemap.xml; /product/* and moved /engineering/* paths

🔵 Hub and index pages use non-semantic or malformed heading structures

What we found: Several commercially relevant hub/index pages have heading structures that don't function as passage labels: /learn/claude-for-work repeats a generic "Resources" heading roughly eight times with no descriptive nesting; /system-cards exposes no article H2s at all (content is a bare table, and nav labels surface as the only H3s); /economic-index emits a single run-on H1 and no section headings; and the homepage uses full-sentence strings ("Anthropic is built on hard questions.") as H2s. A minor rendering artifact also duplicates the model name in the H1 on /claude/fable and /claude/mythos ("Claude Claude Fable 5").

Why it matters: LLMs use heading text as passage boundaries and standalone labels when selecting what to quote. Repeated "Resources" headings, missing section headings, and sentence-as-heading patterns give extraction engines no clean, self-describing anchor to cite, lowering the odds these pages are quoted even when their underlying content is relevant.

Business consequence: For comparison queries like "Claude for enterprise security" or "Claude economic impact data," extraction engines look for a clean, self-describing heading to quote — repeated "Resources" labels and sentence-as-heading patterns leave them no anchor, so relevant Anthropic content goes unquoted in favor of competitors with cleaner structure.

Recommended fix: Give hub/index pages a single descriptive H1 and unique, noun-phrase H2/H3 section headings; replace repeated "Resources" labels with specific ones; ensure /system-cards and /economic-index expose real section headings around their tables/dashboards; and fix the duplicated-word H1 on the /claude/fable and /claude/mythos model pages.

Impact: Medium Effort: 1–3 days Owner: Engineering Affected: /learn/*, /system-cards, /economic-index, homepage, /claude/fable & /claude/mythos

Manual Verification Checklist

The following items could not be assessed through our analysis method (rendered markdown). We recommend your engineering team verify these manually before the validation call.

Data-driven pages may render only a loading placeholder to non-JS fetchers

What to check: The /economic-index page returned only its H1 plus a "Loading data" placeholder in rendered output — the actual economic dashboard content did not appear in the fetched HTML/markdown, consistent with client-side rendering of the main data payload. Because our method reads rendered markdown rather than executing a full browser, we flag this for confirmation rather than asserting it definitively. If the substantive content is injected only after JS executes, non-JS crawlers and some AI ingestion pipelines will index an empty shell — making an otherwise citable, data-rich asset invisible for economy / AI-labor-impact queries where the Economic Index is a differentiated primary source.

Recommended action: Verify /economic-index (and any other dashboard/interactive pages) with JavaScript disabled or via a rendering crawler such as Screaming Frog in JS-rendering mode. If the data is client-side only, add server-side rendering or a static, crawlable summary of the key figures within the initial HTML response.

Effort: 1–2 weeks Owner: Engineering

Structured data (JSON-LD schema) could not be assessed and needs manual verification

What to check: Our analysis reads rendered markdown, so JSON-LD / microdata blocks in the raw HTML head are not visible to us. We could not confirm whether model pages carry Product schema, whether news/engineering posts carry Article schema, or whether FAQ-bearing model pages carry FAQPage schema. Schema presence is recorded as null for every inventoried page rather than asserted absent. Appropriate schema helps search and AI systems parse entities, dates, and Q&A pairs reliably — a low-cost structural win, but it must be verified against raw HTML before any claim is made.

Recommended action: Run the model pages, a sample of /news and /engineering posts, and the FAQ-bearing pages through Google's Rich Results Test or a structured-data validator (or view-source). Add Article, Product, and FAQPage schema where absent, prioritizing model pages and evergreen guides.

Effort: < 1 day Owner: Engineering

Meta descriptions and Open Graph tags could not be assessed and need manual verification

What to check: Meta description and Open Graph / social-card tags live in the raw HTML head, which is not present in the rendered markdown our method reads. We cannot confirm their presence or quality on any inventoried page, so meta_description is recorded as null and has_og_tags as false-by-default (unverified) across the inventory. These tags shape how pages are summarized and previewed in search and when shared into the tools where buyers discuss vendors.

Recommended action: Spot-check meta descriptions and OG/Twitter card tags across the model pages, top /news posts, and /learn hub pages using a social-preview debugger or view-source. Add or de-duplicate where needed.

Effort: < 1 day Owner: Marketing

Site Analysis Summary

Pages analyzed 41 (all commercially relevant)
Heading hierarchy (avg) 0.74
Content depth (avg) 0.68
Passage extractability (avg) 0.65
Freshness (weighted avg) 0.43  (content mktg 0.24 · product 0.81 · structural n/a)
Schema coverage (avg) Unable to assess (41 pages unscored)
Findings logged 6 total — 1 high, 3 medium, 2 low (0 critical)

Partial Sample This analysis covered 41 pages on anthropic.com. Two data limits to note: schema coverage is null across all 41 pages (our method can't see raw-HTML JSON-LD — see the verification checklist) and 9 freshness scores are unscored (7 structural/reference pages plus 2 undated product pages). Just as importantly, the bulk of Anthropic's commercial surface now lives on claude.com / platform.claude.com and was outside this crawl — which is why the audit-domain question in the Company Profile matters before we finalize scope.

Next Steps

What Happens Next

Why Now The window to establish GEO visibility is open but closing:

• 94% of B2B buyers now use LLMs during the buying process (6sense, November 2025) — for a frontier-AI vendor, discovery has already moved into the assistant.
• Early citations compound: domains AI platforms learn to trust now get cited more often as retrieval and training data accumulate.
• Gartner predicts 90% of B2B buying will be AI-agent-intermediated by 2028 (Gartner, October 2025) — increasingly, the "buyer" reading your pages is an agent, not a human.
• Frontier-AI vendor comparison is still early-innings in GEO — acting now means competing against inaction, not against entrenched strategies.

The full audit will measure how often Claude surfaces — and how often it doesn't — across the buyer queries that define this category: "best enterprise LLM for regulated industries," "most accurate model for high-stakes analysis," "Claude vs OpenAI vs Gemini for coding," "AI with copyright indemnification." You'll see exactly which of these return your competitors but not Anthropic across the AI platforms your buyers use. Because Layer 1 already shows your crawlers are open and your product pages are current, you're starting from a stronger technical baseline than most — so the fixes in this document compound your advantage rather than just closing a gap.

01

Validation Call

45–60 minutes. We walk through this document together, confirm the inputs, and resolve the open questions — starting with which domain the audit measures.

02

Query Generation & Execution

We generate buyer queries from the validated personas, competitors, features, and pain points, then run them across the selected AI platforms.

03

Full Audit Delivery

Visibility analysis, competitive positioning, and a prioritized three-layer action plan — including the content recommendations this document deliberately holds back.

Start Now — Engineering Three technical fixes your engineering team can start before we even meet — none depend on the audit results: (1) regenerate the anthropic.com sitemap to drop the /product/* and /engineering/* URLs that 308-redirect off-domain to claude.com; (2) fix the non-semantic hub headings (the repeated "Resources" labels on /learn/claude-for-work, missing section headings on /system-cards and /economic-index) and the duplicated-word H1 on /claude/fable and /claude/mythos; and (3) verify /economic-index renders server-side (or add a static summary) and confirm JSON-LD schema + meta/OG tags via Google's Rich Results Test. Your robots.txt is already confirmed open to GPTBot, ClaudeBot, PerplexityBot, and Google-Extended — no action needed there. These improve your baseline visibility before we measure it.

Before the Call

Your Pre-Call Checklist

Two jobs before we meet. The questions on the left require your judgment — no one knows your business better than you. The engineering tasks on the right don't require the call at all.

Questions for You
Which domain does the audit measure — anthropic.com, claude.com, or both?
If wrong: we measure a research/news surface while buyers' citations resolve to the claude.com commercial domain we didn't test.
Is Sofia Marchetti (Director of AI Enablement) a real distinct buyer, or is this a CTO-plus-CISO purchase?
If wrong: we generate and report a full persona's adoption/ROI query cluster that maps to no real buyer.
Are Price/Cost Efficiency and Multimodal Generation (both rated weak) actually losing you deals?
If wrong: we frame the competitive audit around price/modality when frontier quality is what wins — or miss a real value objection.
Are Microsoft Copilot and Cohere true primary rivals, or bundling pressure / a category listing?
If wrong: ~6–8 head-to-head queries per vendor test the wrong competitive axis (seat economics vs. model quality).
Does the VP of Engineering (Ravi Krishnan) control the platform budget, or only recommend?
If wrong: we mis-weight validation-stage vs. evaluation-stage queries on API and coding-agent performance.
Does the CISO (Danielle Osei) hold a hard veto, or sign off after IT?
If wrong: governance and residency queries are mis-weighted as gating criteria rather than evaluation factors.
Does the General Counsel (Gregory Vaughn) gate deals on IP/data terms, or only review the paper?
If wrong: copyright-indemnification and data-privacy queries are mis-prioritized in the must-answer set.
Is the CTO and Chief AI Officer (Marcus Thorne) one buyer, or two seats at the table?
If wrong: one persona should split into two with different intents (platform reliability vs. transformation mandate).
Do a Head of ML Platform, Procurement/Vendor-Risk lead, or LOB economic buyer show up in your deals?
If wrong: a real buyer gets no dedicated query cluster and goes unmeasured.
Is "runaway token cost" really high-severity, and are we missing model-deprecation, benchmarking, or lock-in pains?
If wrong: severity ordering mis-ranks which frustrations we test first.
For Engineering — Start Now
Regenerate the anthropic.com sitemap
Drop the /product/* and /engineering/* URLs that 308-redirect off-domain to claude.com — recovers crawl budget and consolidates citation authority.
Add visible "Last updated" dates to the ~10 evergreen explainers
Re-timestamp the agent / MCP / context-engineering / SWE-bench guides so they regain freshness credit in AI citations.
Fix non-semantic hub headings + duplicated-word H1s
Give /learn hubs, /system-cards, and /economic-index unique noun-phrase headings; fix "Claude Claude" on /claude/fable & /claude/mythos.
Verify /economic-index server-side rendering
Fetch with JS disabled; if the dashboard is client-side only, add SSR or a static crawlable summary of the key figures.
Confirm JSON-LD schema + meta/OG tags via Rich Results Test
View-source on model, /news, and FAQ-bearing pages; add Article / Product / FAQPage schema where absent.
Alignment

We're Aligned On

This isn't a contract — it's a shared understanding. The audit runs against what's below. If something changes between now and the call, we adjust. The goal is to make sure we're asking the right questions for the right buyers against the right competitors.
Already Confirmed
Competitive set — 5 primary (OpenAI, Google Gemini, Microsoft Copilot, Mistral AI, Cohere) + 5 secondary (DeepSeek, Meta Llama, Amazon Nova, xAI Grok, Perplexity)
Persona set — 5 personas: 3 decision-makers (CTO / Chief AI Officer, CISO, General Counsel), 1 evaluator (VP Engineering), 1 influencer (Director of AI Enablement)
Feature taxonomy — 11 capabilities with outside-in strength ratings (7 strong, 2 moderate, 2 weak)
Pain point set — 10 buyer frustrations with severity ratings (6 high, 4 medium)
Layer 1 technical audit — 6 findings logged (1 high, 3 medium, 2 low; 0 critical), engineering notified; robots.txt confirmed open to all AI crawlers
Decided at the Call
Audit domain scope — anthropic.com vs. claude.com vs. both; the commercial surface has migrated off the audited domain, and this decision sets everything downstream
Feature overweighting — which 3 of the 7 strong capabilities to emphasize; Enterprise Governance and Data Privacy & IP each map to three high-severity pains (data leakage, shadow AI, regulatory compliance), making them the strongest defensive picks — the third slot (Reasoning, Agentic Coding, or Safety) is the call's to make
Weak-feature framing — whether Price/Cost Efficiency and Multimodal Generation are deal-losers (value-vs-quality queries) or accepted trade-offs
AI Enablement persona — keep, reclassify, or drop Sofia Marchetti (the only inferred, medium-confidence persona)
Competitor tier adjustments — Microsoft Copilot & Cohere (primary, medium confidence); Amazon Nova model-vs-model treatment given Bedrock distributes Claude
Client
Date