Outbound Links
Get the links a domain points out to, with anchor text and a concentration summary.
https://api.rankparse.com/v1/outbound-links2 creditsAPI keyReturns links going out from a domain to other websites, ranked by how often each link appears, plus a small summary of how concentrated those destinations are. Use it to audit who a website links to or to find websites that a competitor endorses. Data comes from the pre-processed link graph snapshot (see Response format).
Query parameters
domainstringrequiredDomain whose outgoing links you want, for example smartify.in. No scheme or path. Must contain a dot. Matched exactly: links from subdomains are not included.
limitintegerdefault 100Number of rows to return. Clamped to 1 to 1000. A non-numeric value falls back to 100. There is no offset: raise limit to see more rows.
Response
domainstringThe normalized (lowercased) domain you queried.
dataarrayOutbound links sorted by edge_count descending. Empty when the domain has no data.
Show propertiesHide properties
to_domainstringDomain being linked to.
to_urlstringFull URL being linked to.
anchor_textstring | nullAnchor text, or null when there is none.
link_typestringhyperlink, image, script, stylesheet, iframe or other.
edge_countintegerHow many times this exact link was observed.
totalintegerNumber of rows returned in data.
limitintegerThe limit value you sent, echoed back as received.
credits_usedintegerAlways 2.
credits_remainingintegerYour balance after this call.
summaryobjectConcentration summary computed over the returned rows only.
Show propertiesHide properties
unique_destinationsintegerNumber of rows in data (not distinct domains).
top_domainstring | nullDestination domain of the first row, or null when there are no rows.
top_domain_concentrationnumberedge_count of the first row divided by the sum of edge_count across the returned rows (0 to 1). 0 when there are no rows.
Behavior and limits
- Row grain: a row is one distinct combination of destination domain, URL, anchor text and link type, so the same domain can appear many times.
- Fallback: if the grouped query fails, the endpoint retries with a simpler query that returns stored rows (with exact duplicates removed) instead of summed ones. The response looks the same and carries no flag, so
edge_countcan then undercount. - Credits: 2 per call, charged before the lookup. Server errors are refunded. A request rejected with
invalid_domainis refused before any charge. - Caching: results are cached for 30 days per crawl release and per combination of
domainandlimit. A cache hit is billed the same as a fresh lookup. - Empty results: a domain with no data returns
200withdata: []and asummarywithunique_destinations: 0, never404. - Rate limit: 60 requests per minute per key. See Rate limiting.
Errors
| HTTP | Code | When |
|---|---|---|
| 400 | invalid_domain | domain is missing or not a valid domain. Checked before the charge: nothing is charged. |
| 401 | invalid_api_key | Missing, invalid or revoked API key. |
| 402 | insufficient_credits | Balance is lower than the call cost. |
| 402 | spend_cap_reached | Your monthly spend cap is reached. |
| 429 | rate_limited | More than 60 requests in a minute. |
| 503 | query_timeout, upstream_error | Transient data store problem. Retry after the Retry-After header. Credits are refunded. |
See Errors for the error body format.
Related
MCP tool: get_outbound_links (accepts domain, limit).