Generate Organization + LocalBusiness Schema (JSON-LD)

AI Visibility & GEO Claude intermediate

Produce valid JSON-LD structured data so AI and search engines can parse your business facts without guessing.

When to use it: When you want to add schema markup to your homepage so AI can confirm what you are, where you operate and how to reach you.

You are a technical GEO specialist generating JSON-LD structured data for an Australian small business. Structured data states your business facts in a format AI and search engines parse directly. You produce only markup that reflects confirmed facts — invented schema misleads the engines and can be flagged.

<context>
[BUSINESS NAME]: exact legal/trading name.
[TYPE]: closest business type (e.g. plumber, cafe, law firm, retailer) — I'll map it to the right schema type.
[URL]: homepage URL.
[ADDRESS]: street, suburb, state, postcode — or 'service-area only, no public address'.
[SERVICE AREA]: suburbs/regions served.
[PHONE / EMAIL]: public contact details.
[HOURS]: opening hours, if applicable.
[SOCIALS / PROFILES]: URLs of your Google Business Profile, LinkedIn, Facebook, directory listings (these become sameAs links).
[PRICE RANGE]: e.g. $$ — optional.
</context>

<task>
First, pick the most specific schema @type that fits [TYPE] (e.g. Plumber, CafeOrCoffeeShop, LegalService, Store) and note it. If the business is service-area only, use areaServed and omit a street address rather than inventing one.
Then produce one JSON-LD block that:
1. Uses @type Organization or the specific LocalBusiness subtype, with name, url, description.
2. Includes address (PostalAddress) OR areaServed, telephone, email, openingHours where provided.
3. Adds sameAs with every profile URL given.
4. Adds priceRange only if provided.
Validate mentally against schema.org — no empty required fields, correct nesting, AU phone/postcode formats.
</task>

<output_format>
- The complete JSON-LD in a single code block, wrapped in a <script type="application/ld+json"> tag, ready to paste into the homepage <head>
- A plain-English list of every field you populated and where the value came from
- Any [NEEDED: ...] fields you left out because the fact wasn't provided
- One line on how to test it (Google Rich Results Test)
en-AU conventions.
</output_format>

Grounding: populate fields only from the context. Never invent an address, phone number, hours, price range or profile URL. If a fact is missing, omit the field and list it as [NEEDED: ...] — do not use placeholder data that could ship live.

Copy the block above straight into Claude — anything in [BRACKETS] is yours to fill in.

Want it tuned to your business? Bring it to the free weekly call and we'll adapt it live.

Join the free call

More ai visibility & geo prompts

Audit Your Website's AI Visibility in 15 Minutes

Get a structured, scored read on whether ChatGPT, Perplexity and Google's AI can find, read and recommend your business.

Write an llms.txt for Your Business

Generate a clean llms.txt that tells AI crawlers what your business is, what you offer and which pages matter most.

Turn a Service Page Into AI-Quotable Answers

Rewrite a vague service page into question-led sections an AI can lift and cite verbatim.