Inspect the evidence behind a link count
GET /api2/getBacklinks.php returns a backlinks array and a metrics object. Rows contain anchor text, the source page title, source and target URLs, and nf: 1 for nofollow, 0 otherwise.
The summary includes observed links in and out, pages crawled, referring domains, follow and nofollow counts and their ratio. An optional language_share object appears when language data exists.
Use rows for a candidate review, broken-target investigation or a record of references in your own workflow. The Pretty dataset overview explains the crawl context; this is observed data, not a live exhaustive inventory.
Request a representative domain sample
Authenticate over HTTPS and pass app and domain. Add onePerDomain=1 to return at most one row per referring domain. This Unix-shell example uses a placeholder; a successful real request consumes one credit.
curl -G "https://domdetailer.com/api2/getBacklinks.php" \
-H "apikey: YOUR_KEY" \
--data-urlencode "app=YourResearchTool" \
--data-urlencode "domain=example.com" \
--data-urlencode "onePerDomain=1"To inspect repeated placements, omit onePerDomain. Use subdomain=1 when the target is a specific subdomain and that scope is what you intend to analyze. The backlinks API reference covers the parameters and response fields.
Read anchor, source and target together
Illustrative single row: the domains and values below demonstrate the structure and are not a real backlink observation.
{
"anchor": "Example resource",
"title": "Resource directory",
"nf": 1,
"source": "https://publisher.example/resources",
"target": "https://example.com/guide"
}An empty anchor is still evidence: the link may be an image or have no captured visible text. A page title can help you label a row, but is not a substitute for verifying the source. A nofollow flag describes the returned link attribute, not a guarantee of its effect in a search engine.
Visit a representative set of sources before acting on the export. Check current availability, placement, topic and whether the destination URL still serves the expected resource.
Keep row limits separate from profile totals
The endpoint returns at most 5,000 rows. A summary count larger than the returned array does not mean all those links are included in that array. A one-per-domain option changes the sample, not the real-world number of links.
The API contract does not provide pagination for fetching an unlimited set. Do not infer completeness from a short response or treat absent rows as proof that a backlink does not exist.
Keep request parameters and the lookup date with saved results. That makes a later comparison interpretable, especially when you change from root-domain scope to subdomain scope or enable one-per-domain sampling.
Budget targeted backlink reviews
100 unique successful requests consume 100 credits. That is about $0.08–$0.14 in consumed credit value at the current pack rates. Credits are bought in packs: the entry pack is $34 for 25,000 credits. This is not a $0.08 checkout option. Compare the current DomDetailer credit packs.
Domain summary scores such as DA and TF use the separate domain metrics request. A practical auction workflow checks metrics across the large list, then requests backlink rows only for the shortlist.
Keep completed responses and investigate timeouts before resubmitting. Use saved data within the service terms; access does not grant redistribution rights.
Questions
Does onePerDomain return every link from each website?
No. It returns at most one row per referring domain, useful for breadth. Omit it when repeated placements matter, while retaining the overall 5,000-row limit.
Are these Moz or Majestic backlink rows?
The backlink rows and profile summary come from DomDetailer’s Pretty index. Moz and Majestic scores are supplied by the separate domain metrics endpoint.