Backlink Checker - Referring Domains & Link Gap, Open Data

What does Backlink Checker (Referring Domains & Link Gap) do?

Give it domains, get the websites that link to them (authority score and Chrome traffic rank for each): $1.50 per 1,000 referring domains. Give it your site and 1-10 competitors, get the sites that link to them but not to you: $3.00 per 1,000 link prospects. Domains the web graph does not know, duplicates, invalid lines and empty results are free; there is no start fee. Data: the open Common Crawl web graph (domain level, July-September 2026), not Ahrefs or Semrush. Jump to How to use it.

Two jobs, in bulk and in seconds, for any website on the web:

1. Referring domains: the websites that link to a domain, strongest first, each with a 0-100 authority score and its Chrome traffic rank (Top 1K ... Top 1M websites by real visitors).
2. Link gap ("link intersect"): enter your website and up to 10 competitors and get the websites that link to your competitors but not to you, ranked by how many competitors they link to, then by authority. These are your link-building and PR prospects: directories, review sites, partner pages, blogs and news sites that already cover businesses like yours.

Open data, no scraping of SEO tools, no login. The links come from the public Common Crawl web graph (domain level, July-September 2026: 2.1 billion links between 133 million domains), the traffic ranks from Google's public Chrome UX Report. Nothing is read from Ahrefs, Semrush or Moz.

Real results (3 October 2026)

Referring domains (each domain also gets the list itself, strongest first):

Domain	All referring domains	Among the 5M best-connected sites	Strongest referring domains (authority)
bbc.co.uk	249,224	104,990	top 1,000 listed
zapier.com	48,072	14,804	adobe.com (85), dropbox.com (82), paypal.com (85), stripe.com (85) with filters authority >= 40, real visitors, no big platforms
apify.com	3,556	1,527	google.com (98), wordpress.org (91), amazon.com (88), github.com (87)
scrapingbee.com	956	434	wikipedia.org (91), github.com (87), medium.com (84), hubspot.com (78)
a site nobody links to	0	0	free row, not charged

Link gap: scrapingbee.com against 4 competitors (scraperapi.com, zenrows.com, scrapingant.com, oxylabs.io), at least 2 competitors: 163 prospects in under a second, led by comparison and review sites that link to all four or three of them (e.g. beehiiv.com, flipsnack.com, europeanbusinessreview.com, savedelete.com). plausible.io against usefathom.com, simpleanalytics.com and matomo.org (2+ competitors, real visitors only): 30 prospects such as smashingmagazine.com, yoast.com, cam.ac.uk and buzzsprout.com.

Use cases

• Link building: find the sites that link to two or more competitors but not to you (minCompetitorsLinked: 2), keep the ones with real visitors (onlyWithChromeTraffic), and contact them.
• Competitor research: see who links to a competitor, strongest first: partners, customers, press, directories.
• Backlink audit: list the referring domains of your own site and check which strong sites already link to you.
• PR and outreach lists: news sites and blogs that wrote about companies in your niche.
• Due diligence: check who links to a domain before you buy it or partner with it.

How much does it cost?

Event	Price	When
Referring domain	$1.50 / 1,000 rows	one website linking to a domain on your list
Link prospect	$3.00 / 1,000 rows	link gap mode: one website linking to your competitors but not to you

Free: domains that are not in the web graph, rows removed by your filters, duplicates, invalid lines and link gap runs that find nothing. No start fee. Larger Apify plans get the usual volume discount (Silver 10%, Gold and above 20%).

Examples: the top 100 referring domains of 50 websites = 5,000 rows = $7.50. A link gap against 3 competitors returning 200 prospects = $0.60.

For comparison (Apify Store, checked 8 October 2026): the most used backlink Actors (s-r/backlinks-checker, santhej/backlink-checker, Semrush-based ones) return individual backlinks (page URLs, anchor text) from commercial tools' crawlers and charge per backlink row; one competitor-gap Actor charges $0.15 per domain. This Actor does not have page URLs or anchor texts: it works at the domain level, from open data, with a lower price per row.

Control your cost

• Before it starts, the run logs its worst-case cost and compares it with your maximum charge per run. Rows that would pass the maximum are not returned: SUMMARY.status is LIMIT_REACHED and SUMMARY.notProcessed lists what was left out.
• maxReferringDomainsPerDomain (default 100) and maxResults cap the number of charged rows.
• "Max results per run" must be at least 1; a value of 0 or less stops the run with one free invalid_input row (nothing is looked up or charged).
• Every row has charged: true or false; failed rows have an error that ends with "(not charged)".

How to use it

Referring domains: choose "Referring domains of my domains", paste domains or URLs into Domains (or pick a Dataset ID), set Max referring domains per domain and click Start. Open the Referring domains tab.

Link gap: choose "Link gap", enter Your domain and 1-10 Competitors, optionally Linking to at least N competitors (2 finds the sites that link to businesses like yours on purpose), and click Start. Open the Link gap (prospects) tab.

In "Link gap" mode a "Dataset ID" is not used (it only adds domains in "Referring domains" mode): it is ignored, listed in SUMMARY.ignoredInputs, and nothing is read from it or charged for it.

Filters for both modes: Minimum authority, Only websites with real visitors (in Chrome's top 1 million), Leave out big platforms (about 90 sites where anyone can get a link: google.com, youtube.com, github.com, medium.com, blogspot.com, amazonaws.com, linkedin.com...; on by default for the link gap, off for referring domains; review sites and directories stay in) and Exclude domains.

Subdomains are checked as their registrable domain (blog.example.co.uk becomes example.co.uk): the web graph counts links between registrable domains.

Input example

This is the input the form is prefilled with; a run with it takes a few seconds and costs only cents:

{
    "mode": "referringDomains",
    "domains": [
        "apify.com",
        "scrapingbee.com"
    ],
    "maxReferringDomainsPerDomain": 25
}

Output examples

Referring domains mode (real row, 3 October 2026):

{
    "mode": "referringDomains",
    "domain": "zapier.com",
    "input": "zapier.com",
    "referringDomain": "adobe.com",
    "position": 1,
    "authorityScore": 85,
    "chromeTrafficBucket": 1000,
    "trafficRank": "Top 1K websites (Chrome)",
    "targetReferringDomains": 48072,
    "targetListedReferringDomains": 14804,
    "targetListTruncated": true,
    "found": true,
    "dataRelease": "cc-main-2026-jul-aug-sep",
    "charged": true,
    "checkedAt": "2026-10-03T13:49:45.617Z"
}

Link gap mode (real row):

{
    "mode": "linkGap",
    "yourDomain": "scrapingbee.com",
    "prospect": "beehiiv.com",
    "position": 3,
    "authorityScore": 68,
    "chromeTrafficBucket": 500000,
    "trafficRank": "Top 500K websites (Chrome)",
    "linksToCompetitors": [
        "zenrows.com",
        "scrapingant.com",
        "oxylabs.io"
    ],
    "competitorsLinked": 3,
    "competitorsChecked": [
        "scraperapi.com",
        "zenrows.com",
        "scrapingant.com",
        "oxylabs.io"
    ],
    "linksToYou": false,
    "mayAlreadyLinkToYou": false,
    "prospectUrl": "https://beehiiv.com/",
    "dataRelease": "cc-main-2026-jul-aug-sep",
    "charged": true,
    "checkedAt": "2026-10-03T13:49:46.629Z"
}

What "strongest first" and the scores mean

• Only the 5 million best-connected websites of the web graph (by harmonic centrality, the same rank as our Domain Authority Checker) are listed as referring domains, and at most 1,000 per domain. targetReferringDomains is the full count of linking domains, targetListedReferringDomains how many of them are among the listed websites, and targetListTruncated: true means there are more than 1,000.
• Authority (0-100) is our open-data score (rank in the web graph plus number of referring domains), not Moz DA or Ahrefs DR. It is the same score as our Domain Authority Checker.
• Traffic rank is the Chrome UX Report popularity bucket of the referring website (Top 1K ... Top 1M); null means it is not in Chrome's top 1 million.

Use it from code

curl -X POST "https://api.apify.com/v2/acts/tidytools~backlink-gap/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"mode":"linkGap","yourDomain":"mysite.com","competitors":["competitor1.com","competitor2.com"],"minCompetitorsLinked":2}'

With AI agents (MCP): connect Apify's MCP server (https://mcp.apify.com?tools=tidytools/backlink-gap) to Claude, Cursor or any MCP client and ask, for example, "Which sites link to both of my competitors but not to me?". Minimal input: {"domains": ["example.com"]}.

Limitations

• Domain level, not page level: you get the linking websites, not the individual pages, URLs or anchor texts, and no follow/nofollow.
• Common Crawl sees a large share of the public web, not all of it. Counts are lower than Ahrefs' or Semrush's; a site that links to you may be missing if the crawl did not reach the linking page.
• Quarterly snapshot (currently July-September 2026), not live: a link added last week is not in it yet, and a link removed since then is still counted.
• Only the 5 million best-connected websites can appear as referring domains, and at most 1,000 per domain. Small sites linking to you (personal blogs, spam) are counted in targetReferringDomains but not listed.
• Link gap with a very popular own site: if your own site has more than 1,000 listed referring domains, a weak prospect might already link to you outside the stored list; such rows say mayAlreadyLinkToYou: true.

FAQ

Does the Actor collect usage data? The Actor sends one anonymous usage counter per run (input type, item count range, outcome) to improve the tool; no URLs, content or account data.

What happens when I enter something that is not a domain? Each invalid line becomes a free result row with errorType: "invalid_input" and the run still ends as succeeded. If nothing usable is left (no valid domain, no "Your domain", a bad Dataset ID, "Linking to at least N competitors" higher than the number of competitors), you get one free invalid_input row explaining what to fix; nothing is looked up or charged.

Does it work for URLs, subdomains and e-mail addresses? Yes: https://blog.example.co.uk/page and info@example.co.uk are both checked as example.co.uk. The same domain written differently (Example.com, www.example.com, https://example.com/pricing) is one domain and charged once. Entries can be separated by line breaks, spaces, tabs, commas or semicolons, and pasted quotes or brackets ("example.com", <example.com>) are ignored.

What counts as invalid? Text that is not a domain, IP addresses and localhost, internal names (.local, .internal), unknown top-level domains (foo.invalidtld), labels longer than 63 characters, and objects or numbers in the list. "Your domain" takes one website; a competitor that is the same website as "Your domain" is skipped with a free invalid_input row.

Is the data live? How often is it refreshed? No. It is the quarterly Common Crawl web graph (currently July-September 2026, loaded 3 October 2026) plus the Chrome UX Report; we load a new release when Common Crawl publishes one. Every row carries dataRelease.

What if a link gap finds nothing? One free row says why (none of the competitors is in the web graph, or every referring domain already links to you or is filtered out).

Why is my site "not found"? None of the 5 million best-connected websites links to it in the July-September 2026 web graph (or the crawl did not reach it). The row is free.

Is the link gap the same as Ahrefs' Link Intersect? It answers the same question (who links to them and not to me) at the domain level, from open data.

Can I use the results commercially? Yes. Common Crawl data is free to use under its terms of use; the Chrome UX Report data is CC BY 4.0 (Google).

Support

Open an issue in the Issues tab with your input and the run link. Issues are checked regularly.