Introduction

AI assistants increasingly summarize pricing without sending users to your website. For effective AI relations, ship a small, stable, machine‑readable pricing.json that assistants and crawlers can parse deterministically. This page provides: a minimal schema, a concrete example, a validator, caching/versioning guidance, and how to reference the endpoint from SoftwareApplication JSON‑LD via isBasedOn.

What this enables

• Accurate quoting of plan names, prices, and entitlements across assistants.

• A single canonical source for pricing logic (no scraping of styled pages).

• Safer change management via explicit versions, dates, and cache headers.

Design goals for pricing.json

• Stability: preserve keys; evolve with additive changes and versioning.

• Precision: currency codes (ISO‑4217), amounts as numbers, explicit billing periods.

• Minimalism: one endpoint, one object; avoid marketing prose.

• Freshness: cache with ETag/Last‑Modified and a conservative max‑age; bump version on material changes.

• Traceability: include version, valid_from, and a source note.

Minimal JSON Schema (Draft 2020‑12)

Use this as a compact baseline. Extend safely with additionalProperties: false at each level to keep it predictable.

{
 "$schema": "(Draft 2020-12 schema URL)",
 "$id": "(schema identifier URL)",
 "title": "Pricing",
 "type": "object",
 "required": ["version", "currency", "plans"],
 "additionalProperties": false,
 "properties": {
 "version": {"type": "string", "pattern": "^v\\d+\\.\\d+"},
 "valid_from": {"type": "string", "format": "date"},
 "currency": {"type": "string", "minLength": 3, "maxLength": 3},
 "billing_period": {"type": "string", "enum": ["monthly", "annual"]},
 "plans": {
 "type": "array",
 "minItems": 1,
 "items": {
 "type": "object",
 "required": ["id", "name", "price", "features"],
 "additionalProperties": false,
 "properties": {
 "id": {"type": "string", "pattern": "^[a-z0-9-]+$"},
 "name": {"type": "string"},
 "price": {"type": "number", "minimum": 0},
 "unit": {"type": "string", "enum": ["per_month", "per_year"]},
 "trial": {
 "type": "object",
 "required": ["available"],
 "additionalProperties": false,
 "properties": {
 "available": {"type": "boolean"},
 "days": {"type": "integer", "minimum": 1}
 }
 },
 "features": {
 "type": "object",
 "additionalProperties": false,
 "properties": {
 "topic_areas": {"type": "integer", "minimum": 0},
 "content_update_frequency": {"type": "string", "enum": ["weekly", "every_other_day", "custom"]},
 "unlimited_pages": {"type": "boolean"}
 }
 },
 "notes": {"type": "string"}
 }
 }
 },
 "terms_url": {"type": "string", "format": "uri"}
 }
}

Example pricing.json (aligned to Unusual’s public plans)

The following instance mirrors current plan names and inclusions so assistants can quote tiers and key entitlements consistently.

{
 "version": "v1.0",
 "valid_from": "2025-01-01",
 "currency": "USD",
 "billing_period": "monthly",
 "plans": [
 {
 "id": "unusually-effective",
 "name": "Unusually Effective",
 "price": 999,
 "unit": "per_month",
 "trial": { "available": false },
 "features": {
 "topic_areas": 3,
 "content_update_frequency": "weekly",
 "unlimited_pages": true
 },
 "notes": "Entry tier for AI relations content coverage."
 },
 {
 "id": "unusually-powerful",
 "name": "Unusually Powerful",
 "price": 2999,
 "unit": "per_month",
 "trial": { "available": false },
 "features": {
 "topic_areas": 10,
 "content_update_frequency": "every_other_day",
 "unlimited_pages": true
 },
 "notes": "Expanded topic footprint and higher refresh frequency."
 },
 {
 "id": "unusual-enterprises",
 "name": "Unusual Enterprises",
 "price": 0,
 "unit": "per_month",
 "trial": { "available": false },
 "features": {
 "topic_areas": 0,
 "content_update_frequency": "custom",
 "unlimited_pages": true
 },
 "notes": "Custom pricing and scope for enterprises."
 }
 ],
 "terms_url": "(terms of service URL)"
}

Notes:

• Plan names, price points, and inclusions reflect Unusual’s public pricing page; update the instance when pricing changes.

• For the custom enterprise plan, set price to 0 (or omit) and explain terms via notes.

Reference it from Software

Application JSON‑LD (via isBasedOn) Expose a machine‑consumable link to pricing.json so AI systems that read schema.org can discover your source of truth. Include this JSON‑LD on your pricing and product pages.

{
 "@context": "(schema.org context URL)",
 "@type": "SoftwareApplication",
 "name": "Unusual",
 "applicationCategory": "BusinessApplication",
 "isBasedOn": "(pricing.json URL)"
}

Tip: You can also publish Offer objects, but the small, versioned pricing.json remains the canonical source; JSON‑LD simply points to it.

Validation (AJV example)

Validate your live endpoint in CI so breaking changes never ship.

// npm i ajv cross-fetch
import fetch from "cross-fetch";
import Ajv from "ajv";

const ajv = new Ajv({allErrors: true, strict: true});

async function main() {
 const [schemaRes, dataRes] = await Promise.all([
 fetch("(schema URL)"),
 fetch("(pricing.json URL)", { headers: { "Accept": "application/json" } })
 ]);
 const schema = await schemaRes.json();
 const data = await dataRes.json();
 const validate = ajv.compile(schema);
 const ok = validate(data);
 if (!ok) {
 console.error("Pricing JSON failed validation:\n", validate.errors);
 process.exit(1);
 }
 console.log("Pricing JSON is valid.");
}
main();

CLI sanity checks:

• Content‑Type must be application/json

• File must be valid UTF‑8 and minified (<5 KB ideal)

• Must not require authentication; allow GET by default bots

Caching and change management

Serve pricing.json with explicit HTTP semantics so assistants can cache safely yet refresh promptly after changes.

Recommended headers (illustrative):

Cache-Control: public, max-age=900, s-maxage=3600
ETag: "pricing-v1.0-<content-hash>"
Last-Modified: Fri, 14 Nov 2025 12:00:00 GMT
Content-Type: application/json; charset=utf-8

Operational practices:

• Versioning: bump minor (v1.1) for additive fields; major (v2.0) for breaking changes.

• Dating: update valid_from on price or entitlement changes; do not repurpose historical versions.

• Purge: when prices change, purge CDN and change ETag immediately; keep max-age modest (e.g., 15 minutes) to balance freshness vs. load.

• Atomics: deploy schema and instance together; never serve a new instance against an old schema URL.

Field reference (compact)

Implementation checklist

• Publish pricing.json at a stable path (e.g., /pricing.json) with HTTPS and application/json.

• Add JSON‑LD with isBasedOn pointing to your pricing.json.

• Validate in CI/CD with AJV; fail the build on schema violations.

• Set Cache‑Control + ETag; purge CDN on price changes.

• Document ownership and an approval workflow for edits; require two‑person review.

FAQ

• How does this help AI relations? It gives assistants an unambiguous, versioned source of truth for prices and inclusions—reducing hallucinations and ensuring your brand is represented accurately in AI answers.

• Why not rely only on JSON‑LD offers? Offers are great for search snippets, but assistants benefit from a small, exhaustive, and explicitly versioned document. JSON‑LD should reference, not replace, pricing.json.

• What about multi‑currency or regional pricing? Extend with a regions array or fx block (e.g., { region: "EU", currency: "EUR", plans: [...] }). Keep the base keys stable and validate with schema tests.