# PricePage Extractor PricePage Extractor turns public SaaS pricing pages into structured JSON. Send a pricing page URL or raw HTML and receive normalized pricing data such as plan names, prices, billing periods, features, trial signals, enterprise plan signals, currency, warnings, and a confidence score. The API is deterministic. It does not use AI, language models, or hidden model calls at runtime. ## What It Extracts The API can detect common pricing page elements such as: * Plan names * Prices and price text * Currency * Monthly, annual, weekly, daily, lifetime, and custom billing periods * Feature bullets * Calls to action * Free plans * Free trials * Enterprise or contact sales plans * Pricing changes between two versions of a page ## Confidence Scores Every extraction includes a `confidence` value from `0` to `1`. Suggested interpretation: * `0.00` to `0.30`: unreliable or no useful extraction * `0.31` to `0.60`: review warnings and inspect manually * `0.61` to `1.00`: likely useful, but still heuristic Pricing pages vary a lot, so always check `warnings` when accuracy matters. ## Authentication When calling this API through RapidAPI, use the standard RapidAPI headers shown in the code examples: * `X-RapidAPI-Key` * `X-RapidAPI-Host` Do not send `X-RapidAPI-Proxy-Secret`. That header is added privately by RapidAPI and is not visible to API users. ## Health Check `GET /health` Checks whether the API is running. Example response: ```json { "status": "ok" } ``` ## Extract From URL `POST /extract-url` Fetches a public pricing page and extracts structured pricing data. Example request: ```json { "url": "https://example.com/pricing", "renderMode": "auto", "includeRaw": false } ``` Example response: ```json { "url": "https://example.com/pricing", "currency": "USD", "billingPeriods": ["monthly"], "hasEnterprisePlan": true, "freeTrial": true, "plans": [ { "name": "Pro", "price": 29, "priceText": "$29", "billingPeriod": "monthly", "cta": "Start free trial", "features": ["API access", "Unlimited projects"], "limits": { "users": 5 } } ], "confidence": 0.88, "warnings": [] } ``` ## Extract From HTML `POST /extract-html` Extracts pricing data from raw HTML that you provide. This is useful when you already fetched the page, stored a snapshot, or want to avoid live network access. Example request: ```json { "html": "<html><body>

Pro

$29/mo

API access
Unlimited projects

<button>Start free trial</button>

</body></html>", "baseUrl": "https://example.com/pricing" } ``` ## Compare Pricing Pages `POST /diff` Compares two pricing inputs and reports pricing changes. Each side can be a URL source or an HTML source. Example request: ```json { "before": { "mode": "html", "html": "<html><body>

Pro

$19/mo

API access

</body></html>", "baseUrl": "https://example.com/pricing" }, "after": { "mode": "html", "html": "<html><body>

Pro

$29/mo

API access
Unlimited projects

</body></html>", "baseUrl": "https://example.com/pricing" } } ``` Example response: ```json { "plansAdded": [], "plansRemoved": [], "pricesChanged": [ { "planName": "Pro", "before": { "price": 19, "priceText": "$19", "billingPeriod": "monthly" }, "after": { "price": 29, "priceText": "$29", "billingPeriod": "monthly" } } ], "featuresChanged": [ { "planName": "Pro", "added": ["Unlimited projects"], "removed": [] } ], "billingChanged": [], "summary": [ "Pro price changed from $19 to $29.", "Pro added 1 feature." ] } ``` ## URL Safety For security, the API only fetches public HTTP and HTTPS URLs. The API blocks requests to: * Localhost * Private IP ranges * Link local addresses * Cloud metadata addresses * Internal network destinations * Protocols such as `file` or `ftp` * Redirects to unsafe destinations ## Common Errors Example error response: ```json { "error": { "code": "unsafe_url", "message": "URL points to a private or unsupported destination." } } ``` Common error codes: * `validation_error`: Request body is invalid. * `unsafe_url`: URL is private, internal, unsupported, or otherwise blocked. * `request_too_large`: Request body is too large. * `html_too_large`: HTML content is too large. * `fetch_timeout`: Upstream page fetch timed out. * `extraction_timeout`: Total extraction timed out. * `extract_failed`: The upstream page could not be extracted. * `browser_rendering_disabled`: Browser rendering is not enabled for this deployment. * `browser_unavailable`: Browser rendering is temporarily unavailable. ## Best Practices Use `/extract-html` if you already have the page HTML. Use `/extract-url` for live public pages. Use `renderMode: "auto"` unless you have a specific reason to force another mode. Review `warnings` for ambiguous pages. Treat low confidence results as incomplete. Avoid sending authenticated pages or private HTML. Keep `includeRaw` set to `false` unless debugging. ## Limitations Pricing pages are highly variable. Some pages hide prices behind scripts, geolocation, login walls, experiments, or complex interactive calculators. The API does not log in, bypass access controls, solve CAPTCHAs, infer missing prices with AI, or guess data that is not present in the fetched HTML. Browser rendering may not be available on every deployment. When browser rendering is unavailable, pages that require JavaScript may return lower confidence results.