Sample report:
salesforce.com
A real AI Visibility Audit — not a mockup. We crawled salesforce.com and analyzed 10 representative pages across Classical SEO, Schema.org JSON-LD, AI-crawler access, answer-readiness, and the site-wide llms.txt index. This is the engineering half of the loop; the measurementhalf — Share of AI Voice, citations, and gap detection — lives inside the app.
salesforce.com
10 pages
6 layers
llms.txt
This is a limited sample.
Salesforce.com has thousands of pages. To keep this preview snappy, we analyzed 10 representative pages spanning their homepage, AI products, industries, customer stories, and pricing. When you run a report on your own site, every page you select is included — typically 20-100+ pages for most websites.
Key findings from this sample audit
AgenticSEO audited 10 representative pages on salesforce.com across six discoverability layers. The pattern below is typical of large enterprise sites: classical SEO is strong, and the AI-visibility layers are where the gaps sit.
- Classical SEO is solid. Titles, meta descriptions, canonicals, and Open Graph tags are present and well-formed on every page analyzed.
- Structured data is uneven. Schema.org JSON-LD exists on some templates but does not consistently declare the organization, product, and FAQ entities that answer engines use to identify who the page is about.
- Answer-readiness is the weakest layer. Most pages are written as marketing narrative rather than self-contained passages an assistant can quote directly as an answer.
- No site-wide llms.txt. There is no index file telling language models which pages matter or how to describe the brand — so AgenticSEO generated one, shown below.
- AI-crawler access needs an explicit policy. Assistant crawlers such as GPTBot, Google-Extended, PerplexityBot, and ClaudeBot should be allowed or denied deliberately, not by omission.
In the full app, each finding becomes a work item: AgenticSEO writes the missing metadata, JSON-LD, llms.txt, and FAQ passages, then publishes them to WordPress, Webflow, HubSpot, or a Git repository after you approve them.
SEO findings ready
10 pages analyzed for Salesforce
unavailable on older reports
fixes written across 10 pages
Layer Coverage
- 🔍 Classical SEO10 · 33%
- 🤖 AI / AIO Tags10 · 33%
- 🧬 JSON-LD10 · 33%
Older report — bands reflect recommendation coverage. Re-run to grade the live site.
Preview only — 7 more pages ready to unlock
Unlock the full report for $29
Pay once. Get the branded Word report, full JSON-LD, your site-wide llms.txt, a deploy-ready sitemap.xml, and up to 25 pages analyzed. No subscription required.
- ✓Branded .docx with TGP analysis
- ✓Up to 25 pages (vs. 3 in preview)
- ✓Site-wide llms.txt included
- ✓Updated sitemap.xml included
- ✓TGF coverage scoring per page
<title>Salesforce: The #1 AI CRM - Welcome to the Agentic Enterprise</title><meta name="description" content="Unite humans and AI agents with Salesforce, the world's #1 AI CRM. Explore Agentforce, Customer 360, and Data Cloud to drive customer success and growth."><meta name="keywords" content="Salesforce, AI CRM, Agentforce, Customer 360, Data Cloud, CRM software, AI agents, Slack, business automation"><meta name="author" content="Salesforce"><link rel="canonical" href="https://www.salesforce.com"><meta property="og:title" content="Salesforce: The #1 AI CRM"><meta property="og:description" content="Drive customer success with the world's leading AI CRM. Discover how humans and agents work together on one trusted platform."><meta property="og:image" content="https://wp.sfdcdigital.com/en-us/wp-content/uploads/sites/4/2026/02/agentforce-dark.png"><meta property="og:type" content="website"><meta name="twitter:card" content="summary_large_image"><meta name="twitter:title" content="Salesforce: The #1 AI CRM"><meta name="twitter:description" content="Experience the Agentic Enterprise where humans and AI agents drive success together.">
The $29 report is just the starting point.
Subscribers also get a continuous closed loop — measure how AI systems actually answer questions about your category, find the gaps, and track your share of voice over time. Below is illustrative output from a Salesforce run.
AI Visibility Scan
We use AI to generate the prompts your buyers actually use, then measure who gets cited. Here's what we saw for Salesforce when we asked "What is the best AI CRM for enterprise?"
How well your pages answer the prompts your buyers ask AI.
Your share of citations vs HubSpot, Microsoft Dynamics, Zoho, Pipedrive.
- Salesforce48 · 48%
- HubSpot27 · 27%
- Microsoft14 · 14%
- Others11 · 11%
ChatGPT + Gemini, weekly
Across 6 weeks of runs
When cited, top of answer
Content Gaps
Prompts where you weren't cited, ranked by buyer-intent value. Each gap suggests the page to update — or a new page to publish.
"What's the best AI agent platform for sales teams?"
Cited instead: HubSpot · Missing comparative content
"Salesforce vs HubSpot pricing for SMB"
Cited instead: HubSpot · No SMB-specific pricing page
"Open-source alternatives to Salesforce CRM"
Cited instead: Zoho, SuiteCRM · No defensive comparison content
"How does Agentforce compare to Microsoft Copilot?"
Cited instead: Microsoft · JSON-LD lacks ComparisonAction
Competitive Benchmarks
Side-by-side AI visibility vs the competitors AI actually mentions in your category.
Numbers shown are from a sample Salesforce run. When you subscribe, these visuals populate with your brand, your competitors, and your tracked prompts — refreshed automatically every week.
Everything in the full app
The complete AI visibility platform.
Subscriptions unlock the rest of the closed loop — scheduling, multi-site management, dashboards, and continuous monitoring.
6-layer discoverability audit
Classical SEO, JSON-LD, AI-crawler access, answer-readiness, llms.txt, and sitemap.xml on every page. Experimental ai:* annotations available as Annex A (opt-in).
Site-wide llms.txt
A single index file telling LLMs how to read your site.
AI Visibility Scan
Test ChatGPT and Gemini on your category prompts.
Citation tracking
Who got cited, where you ranked, what changed week over week.
Content Gaps
Prompts where you lost — with the page to fix or create.
Competitive Benchmarks
Share of voice vs the brands AI mentions in your category.
Coverage trend chart
Watch your answer-readiness score climb over time.
Scheduled re-crawls
Weekly or monthly cadence per site, with email reminders.
Action queue
Prioritized to-do list pulled from gaps and stale pages.
Multi-site management
Manage the parent brand and its product websites from one place.
Finn AI assistant
Built-in copilot that explains findings and drafts updates.
Team & client access
Invite teammates or share read-only views with clients.
Branded .docx reports
Client-ready Word reports your team can ship as-is.
TGF-framed AI prompts
Traction Gap Framework baked into every analysis.
API + integrations
Coming: WordPress and Webflow auto-publish.
Methodology
How this report is produced, so the numbers can be checked rather than taken on trust.
- Page selection. Pages are discovered from the site's sitemap, an llms.txt index when present, and a live crawl, then ranked by buyer intent. The free check scores 5; the one-time report scores up to 500 in a single pass on one website (scheduled re-audits are on the monthly plans).
- Fetching. Each page is fetched as a rendered document, escalating through four crawl tiers when a site blocks or throttles the first attempt, so scores reflect the markup an assistant would receive.
- Six scored layers. Every page is graded on classical metadata, Schema.org JSON-LD validity and entity clarity, AI-crawler access (GPTBot, OAI-SearchBot, Google-Extended, ClaudeBot, PerplexityBot), answer-readiness of its passages, llms.txt coverage, and sitemap coverage.
- Prompt measurement. Buyer prompts are run against live assistants — ChatGPT and Google Gemini on every plan — and each response is stored with the brands named and the domains cited. Share of AI Voice is your named-mention count divided by all competitor-set mentions across the run.
- Google Search Console. Where a site connects GSC, clicks, impressions, click-through rate and impression-weighted average position are reported as measured Google values, never third-party estimates.
- Limitations. Assistant answers are non-deterministic, so single-run results are indicative and trends over scheduled runs are the reliable signal. llms.txt adoption by major assistants is unproven and is treated as the least-proven layer. This sample uses a public website not affiliated with Traction Gap Partners and reflects its markup on the date shown.
Cite this report
AgenticSEO (Traction Gap Partners). "AI Visibility Sample Report." agenticseo.live, April 28, 2026. https://www.agenticseo.live/sample-report
Reuse permitted with attribution and a link to the source URL. Methodology is described above; questions about the scoring layers can be directed to Traction Gap Partners.
Like what you see? Run one on your own site.
Sign up free and run a 5-page preview on your own site — no credit card required. Unlock the full report for a one-time $29.
AI visibility resources
The guides, comparisons, and free tools behind AgenticSEO.