How to Build a Grounding Page for Your Brand

What a Grounding Page does

AI engines disambiguate entities by cross-referencing multiple sources. When someone asks about your brand, the engine needs to answer three questions:

1. Does this brand exist as a real, verifiable entity?
2. What category does it belong to?
3. What are the core facts (founded, HQ, team, pricing model)?

Your Grounding Page is the single URL that answers all three cleanly. Without it, engines patchwork facts from Wikipedia (if you have one), Crunchbase, LinkedIn, and whichever blog post is the freshest, with inconsistent or stale results.

The spec

H1: entity name only

Not "Welcome to Rankscale." Not "Rankscale, the AI Search visibility tool." Just:

Rankscale

The entity name and nothing else. The surrounding page establishes the category.

Lead definition: one factual sentence

The first paragraph is a one-sentence definition in the form: "[Name] is a [category] that [function]." Example:

"Rankscale is an AI Search visibility platform that tracks how AI engines (ChatGPT, Google AI Overviews, Google AI Mode, Perplexity, Claude, Gemini, Copilot, Grok, Deepseek, Mistral) mention, cite, and describe brands across prompts."

One sentence, factual, no marketing language. This is the sentence AI engines quote verbatim when someone asks "what is Rankscale?"

Core facts grid (use <dl> tags)

Category:       AI Search Visibility Platform / GEO Tool
Founded:        [Year]
HQ:             [City, Country]
Team size:      [Range]
Founders:       [Named]
Domain:         rankscale.ai
Status:         Active
Last reviewed:  [Month Year]

Use <dl> / <dt> / <dd> tags in HTML, not a table. This is the semantic HTML pattern engines prefer for fact lists.

FAQ section with entity name in every answer

Five to ten FAQs, each answer containing the entity name at least once. Example:

• What is Rankscale? Rankscale is an AI Search visibility platform.
• Who is Rankscale for? Rankscale is for in-house marketers who want to track how AI engines describe their brand and agencies who want to track how AI engines describe the brands of their clients.

Repeating the entity name in every answer reinforces the association for RAG retrieval.

Volatile-facts block

A separate section for facts that change (pricing, team size, product features). Mark it clearly:

Volatile facts (last reviewed: April 2026)

• Starting price: $X/month
• Active customer count: [range]
• Supported AI engines: [list]

Separating volatile from stable facts lets you refresh one without re-reviewing the other.

"Is NOT" disambiguation (if needed)

If your brand name clashes with anything else (another company, a common word, a historical reference), add an explicit disambiguation: "Rankscale is NOT [X]. It is [Y]." This is what saves you when AI engines confuse you with a namesake.

How to handle multiple entity types on one brand

Most brands are not just an Organization. Rankscale is also a Tool/Platform, a set of Features, a Service (the Accelerator), a Publication (the Academy), and a Segment. One Grounding Page cannot cleanly do all of this. AI engines disambiguate per entity, not per brand.

Hub-and-spoke model

Your Grounding Page at /about is the hub for the Organization entity. Every other entity either merges in as a section (no independent search demand) or splits into its own spoke page (/product, /academy, /team/[name], /resources/glossary/[term]).

Schema.org mapping (compact)

Cross-linking rule

Every spoke schema points back to the hub via publisher, provider, isPartOf, or worksFor. The hub schema points out to every spoke via hasPart, publishes, makesOffer, or employee. That reciprocal graph is what AI engines traverse.

{
  "@type": "Organization",
  "name": "Rankscale",
  "hasPart": [{"@type": "SoftwareApplication", "url": "/product"}],
  "publishes": [{"@type": "CreativeWork", "url": "/academy"}],
  "makesOffer": [{"@type": "Service", "url": "/accelerator"}],
  "employee": [{"@type": "Person", "url": "/team/founder"}],
  "sameAs": ["https://wikidata.org/...", "https://linkedin.com/..."]
}

JSON-LD that mirrors the HTML

Every fact in the HTML should match the JSON-LD. Organization schema with sameAs to Wikidata, LinkedIn, Crunchbase. Covered in Module 4.4.

Review date within 6 months

A Grounding Page with a review date older than 6 months is not trustworthy. Build the refresh into quarterly review.

The launch checklist

1. List your brand's entity types (use the schema mapping table above) and decide hub vs spoke for each
2. H1 is entity name only
3. First paragraph is one-sentence factual definition
4. Core facts grid uses <dl> tags
5. FAQ section with entity name in every answer
6. Volatile-facts block separated
7. "Is NOT" disambiguation (if applicable)
8. JSON-LD Organization schema with sameAs array
9. Review date visible on the page
10. Page is NOT behind auth or paywall
11. URL is simple: /about or /brand (not /en/us/about-us/company-overview-v2)