Plain GET requests
No POST task arrays, no polling queues, no Basic Auth dance. Domain in, JSON out, synchronously.
For Developers
SEO data, bought like a developer buys things.
One header, plain GET requests, flat published prices, JSON responses. No suite subscription, no unit archaeology, no sales call. 100 free credits to start, and they never expire.
Integration
The whole integration
Auth is an X-API-Key header. There is no step two.
# Auth is one header. Every endpoint is a plain GET.
curl "https://api.rankparse.com/v1/backlinks?domain=stripe.com&limit=100" \
-H "X-API-Key: rp_your_key"
# Flat, published cost: 2 credits. Failures aren't charged.
# Response: JSON with a flat data array, no task envelopes.
{
"data": [
{ "from_url": "...", "to_url": "...", "anchor_text": "..." }
],
"credits_used": 2,
"credits_remaining": 98
}Principles
Design decisions you'll notice
Flat published prices
Every endpoint's credit cost is on the pricing page: 1 credit for domain authority, 2 for backlinks, 10 for site explorer. You can quote a job's cost before you run it.
No subscription math
Prepaid credit packs from $10. Credits never expire, so a side project that runs quarterly costs the same as one that runs daily.
Failures aren't charged
A 4xx or 5xx costs zero credits, and credit deduction is atomic: concurrent calls can't double-spend.
SDKs and an OpenAPI spec
Official Node and Python SDKs, a full OpenAPI spec, and copy-paste examples per endpoint in the docs.
MCP-native
An MCP server on every tier, so Claude, ChatGPT, Cursor, and Claude Code can call your SEO data as tools with zero glue code.
What you skip
What you're not signing up for
- A $129+/mo suite subscription as the ticket to API access (Ahrefs, Semrush)
- Unit meters where cost depends on rows × fields × plan tier
- POST task queues and nested result envelopes for a simple domain lookup
- Credits or units that expire with the billing month
- Contact-sales pricing for the numbers that should be on the pricing page
The receipts: we keep dated teardowns of the Ahrefs API and Semrush API, and head-to-heads with every major vendor under /vs.
Tradeoff
The honest tradeoff
RankParse is built on aggregated open web data (the public web link graph, HTTP Archive, CrUX, Tranco), rebuilt quarterly, not a continuous proprietary crawl. That's why the prices look the way they do. If your product's value depends on seeing a link that appeared yesterday, use a live-index vendor and pay their premium; if it depends on scoring, enriching, auditing, or powering agents, quarterly data answers the question. Provenance details on the data page.
Build paths
Pick your build path
REST + SDKs
Node and Python SDKs, or plain fetch/curl against 22 GET endpoints.
Getting startedMCP server
Give Claude, ChatGPT, Cursor, or Claude Code the whole API as typed tools.
MCP overviewFramework guides
Server-side patterns for Next.js and FastAPI, with caching strategies.
Next.js guideNo-code / low-code
n8n node, workflow templates, and an HTTP-request recipe for Make and Zapier.
Workflow libraryYour stack
Works with your stack.
FAQ
Common questions
Backlinks, referring domains, anchor text, domain authority and rank, link intersect, competitor gap, top pages, similar domains, tech stack detection, site health, page SEO, crawl history, and batch lookups: 22 endpoints, all returning structured JSON. Rank-tracking reads from Google Search Console, Bing, and Yandex are free (0 credits).
Prepaid credits, no subscription: 100 free credits with no card, then $10 per 1,000 credits ($0.010), $80 per 10,000 credits ($0.008), or $600 per 100,000 credits ($0.006). Each endpoint has a flat published credit cost, so cost estimation is arithmetic, not archaeology.
Those APIs are gated behind suite subscriptions ($129/mo and $549/mo respectively) and metered in units whose per-request cost depends on rows and fields. RankParse has no plan gate and flat per-call prices. The tradeoff: RankParse's index rebuilds quarterly from open web data rather than a continuous proprietary crawl.
Aggregated public data: the public web for the link graph, HTTP Archive for tech stacks, Chrome UX Report for performance, Tranco for popularity, Wayback CDX for domain history, cross-referenced into one queryable index. Full provenance is on the data page.
The free tier is the sandbox: 100 credits, no card, all endpoints, MCP access, and both SDKs. Since failures aren't charged and credits never expire, you can develop against production data without a meter anxiety loop.
60 requests/minute on Starter, 300 burst on Growth, and no rate limit on Scale (with a 99.95% SLA). Insufficient balance returns a clean 402 rather than silently degrading.
Start free
First call in under a minute.
100 free credits. No card. Flat prices on the pricing page, where they belong.