Link gap analysis: comparing your profile with the sites that rank

A link gap analysis answers two questions. Which sites link to the pages that outrank you and not to you? And how many referring domains stand between your page and theirs? The first gives you a prospect list. The second gives you a budget.

It is also called link intersect analysis, after the report in the tools that produce it.

The two kinds of gap

Gap	What you compare	What it gives you
Domain gap	All referring domains of two or more competitors against your own	A list of sites that have shown they link in your niche
Page gap	Referring domains of the pages ranking for one keyword against your page	The number of links that keyword is likely to need

The domain gap is the classic version. A site that links to two or three of your competitors has already made the editorial decision to cover your topic, or has already shown that it sells placements in it. Either way it is a warmer prospect than a cold list.

The page gap is the one that sets money. If the pages on page one for a keyword have far more referring domains than yours, that difference, multiplied by what a link costs in your market, is the link budget for that keyword.

How to run one

This is a working method, not a formula from a search engine.

1. Pick the keyword group and the page on your site that should rank for it.
2. Take the pages that rank at the top for that group. Benchmarking against the top five for the same page type is a sensible default.
3. Run the intersect report in Ahrefs Link Intersect, Semrush Backlink Gap or Moz Link Intersect, with the competitors as inputs and your site as the exclusion.
4. Sort by the number of competitors each domain links to. Domains that link to two or more go to the top.
5. For the page gap, record the referring domain count for each ranking page and for your own.
6. Vet every domain on the list before anyone contacts it.
7. Group what survives by how the link was won: editorial coverage, a guest article, a directory or association, a paid placement.

Count unique referring domains, not raw links. Ten links from one site are not ten votes. The leaked Google documentation includes a field for repeated links from the same source to the same target, which analysts read as diminishing returns from one domain. That reading is an inference, but it matches the correlation data, where referring domains outperform backlink counts.

What to compare beyond the count

A raw count hides most of what matters. Compare these as well.

Dimension	What to look at	Why
Page type	Share of links going to the homepage, informational pages and commercial pages	Shows whether competitors win with assets or with direct links to money pages
Anchors	Mix of branded, partial-match and exact-match anchors	Google publishes no safe ratio, so the ranking pages are your only benchmark
Pace	New referring domains per month for each competitor	Shows whether the gap is closing or widening
Relevance	Topic, language and country of the linking sites	A gap made of off-topic or foreign sites is not worth closing
Method	Earned, exchanged or bought	Tells you what the market accepts and what it costs

See anchor text for how to read the anchor comparison, and link velocity for the pace comparison.

Read the numbers with care

Correlation is modest. In Ahrefs’ study of one million US keywords, published in January 2025, referring domains correlated 0.255 with rankings. The link was weaker for commercial and transactional searches than for informational ones. Pages that rank also attract links, so some of that correlation runs the other way.

Authority scores are not Google’s. Domain Rating, Domain Authority and Authority Score are third-party estimates built on each vendor’s own crawl. In the same Ahrefs study, Domain Rating alone correlated only 0.131 with rankings, lower than the plain count of referring domains. Do not rank a gap list by DR.

Tools disagree. Each tool crawls a different part of the web, so the same site shows different link counts in each. Use one tool for the whole comparison.

Some of the gap is not worth having. A competitor’s profile includes scraper links, dead forums and links Google most likely ignores. You do not need to copy them.

Filter the list before you use it

Competitors buy poor links too, and a gap report repeats their mistakes back to you as opportunities. BuzzStream’s 2026 analysis of roughly 257,000 sites that sell guest posts found that only 1.37% met a basic quality bar and 39.6% had no organic traffic at all. If your competitors bought from that pool, those sites are now in your gap list.

Run each domain through the checks in how to vet a publisher. A site that links to all five competitors from articles covering casinos, loans and CBD in the same week is a seller list, not an editorial endorsement.

Turn the gap into a budget

Once the list is clean, the arithmetic is simple. Multiply the page gap by a realistic cost per link for your market. A March 2026 survey of 518 SEOs by Editorial.link put the acceptable price of one high-quality link at about $509 on average. A page that is 20 quality referring domains behind is therefore a five-figure decision.

Check that figure against what the ranking is worth. BuzzStream suggests one cap: the monthly traffic value of the keywords multiplied by 24 months, as a rough lifetime value of holding the ranking. If the gap costs more than that, pick a different keyword or a different page.

You do not always have to close the gap on the commercial page itself. Most natural links go to homepages and informational assets, and internal links can carry the benefit onward. Which pages need links covers that choice.

Where to go next

Take the clean list into how to vet a publisher, then price it with link building cost and budget. The tactics chart shows which method fits each kind of prospect.