SaaS Pricing Intelligence Extractor

Deterministic extraction of SaaS pricing tiers from any public pricing-page URL.

What it does

Give it a pricing page; it returns structured plans — plan name, price, billing period and feature bullets — plus every price string detected on the page.

Why it is safe & cheap

• Pure code, no AI by default. Pricing is parsed with deterministic regex + structural HTML heuristics. No LLM and no paid API are called on a normal run, so there is $0 idle cost and $0 uncovered per-run cost beyond Apify compute.
• SSRF-guarded. Reuses OMEGA's proven DataPulse fetch core: private / loopback / link-local / reserved IPs are blocked (fail-closed), a domain blocklist is enforced, and every redirect hop is re-validated.
• Bounded. Hard caps on connect/read timeouts, response size (2 MB) and redirect count — it never hangs.

Input

{
  "url": "https://www.example.com/pricing",
  "urls": ["https://other.com/pricing"],
  "llm_api_key": "<optional: YOUR OpenRouter key>",
  "llm_model": "deepseek/deepseek-chat"
}

url / urls are public pricing-page URLs. llm_api_key is optional and, if supplied, enables AI enrichment using your own key — the actor never uses any built-in AI key.

Output (one dataset item per URL)

{
  "url": "https://www.example.com/pricing",
  "status": "completed",
  "plans": [
    {"name": "Starter", "price": "$19", "period": "month", "features": ["10 projects", "Email support"]},
    {"name": "Pro", "price": "$49", "period": "month", "features": ["Unlimited projects", "Priority support"]}
  ],
  "raw_prices": ["$19", "$49", "Custom"],
  "parse_confidence": "high",
  "method": "deterministic_code",
  "extracted_at": "2026-06-23T00:00:00+00:00"
}

status is completed, blocked (security gate / SSRF / blocklist / auth-walled) or failed. parse_confidence is a code-owned heuristic (high ≥2 plans, low 1 plan, none 0). When you supply your own key an extra llm_plans field is added; the deterministic plans always stand on their own.

Use cases

• Competitive pricing monitoring — track competitor plans and price changes.
• Build a pricing-intelligence dataset across many SaaS vendors.
• Normalize messy pricing pages into structured plans for analysis or dashboards.

How it works (deterministic, code-only)

The page is fetched through the SSRF-guarded DataPulse core, then plan cards, price strings and feature bullets are extracted with deterministic regex + structural HTML heuristics. A code-owned parse_confidence reflects how many clean plans were found. AI enrichment is optional and only runs with your key.

Limitations (honest)

Pricing pages that render entirely client-side (heavy JS, no server-side HTML) may expose fewer tiers; the actor still returns all detected raw_prices and an honest low/none confidence rather than fabricating plans.