A backlink profile is the complete set of external links pointing to a site, described by four properties rather than one number: how many distinct domains link to you, what anchor text they use, what types of links they are, and how that mix accumulated over time.
By Andrew Linksmith, Head of Link Acquisition · Update:
A backlink profile is the complete set of external links pointing to a site, described by four properties rather than one number: how many distinct domains link to you, what anchor text they use, what types of links they are, and how that mix accumulated over time. Referring domains matter more than total backlinks, because one arrangement can generate thousands of links from a single source.
That last distinction is where most conversations about link profiles go wrong. A site with 40,000 backlinks and 60 referring domains has a weaker profile than a site with 900 backlinks and 400 referring domains — but the first number is the one people quote.
Distinct websites linking to you. The closest single proxy for link authority, and the number to track month over month.
Benchmark: against page-one competitors, not in the abstract
How concentrated your links are. If a fifth of your referring domains share hosting, templates or outbound patterns, that is one source presenting as many.
Warning: any recognisable cluster of 10+
Share of referring domains using each anchor type. The most detectable manipulation signal, and the easiest to get wrong deliberately.
Benchmark: median of three ranking competitors
Editorial, contextual, footer, directory, and the split between followed and qualified links. A profile with no nofollow at all looks constructed, because real link accumulation is messy.
Normal: 15–35% qualified or nofollow
Share of referring domains that operate in or adjacent to your field. Relevance remains the factor practitioners rate above raw authority for most campaigns.
Aim: majority relevant or adjacent
Whether the domains linking to you have real readers. Ahrefs has reported that fewer than one site in ten in its index gets any organic traffic, so this is a genuine differentiator.
Track: share of referring domains with 500+ monthly visits
Dimension four is worth expanding because it produces a counter-intuitive rule. A profile consisting entirely of followed links is not a strong profile — it is an unlikely one. Real link accumulation produces a mixture: social profiles, forum mentions, syndicated copies, press listings, all of which are commonly qualified. A perfectly clean followed profile suggests every link was arranged.
Branded52%
Naked URL18%
Generic14%
Partial match11%
Exact match5%
Branded14%
Naked URL6%
Generic9%
Partial match28%
Exact match43%
Property | Healthy | Problematic |
Dominant anchor type | Branded and naked URL | Exact-match commercial |
Growth shape | Uneven but continuous | Flat lines punctuated by vertical steps |
Source variety | Media, blogs, directories, forums, partners | One format from many similar domains |
Followed / qualified mix | Mixed, with a meaningful qualified share | Almost entirely followed |
Publisher traffic | A visible core with genuine readership | High ratings, negligible audiences |
Topical spread | Concentrated in your field, with some adjacency | Scattered across unrelated industries |
One caveat that matters more than the table. "Natural" is niche-specific. A local services business will show far more directory and citation links than a software company; a news publisher will show far more incidental links than either. The correct reference is not a universal ideal but the profiles of the pages currently ranking for your terms — which is what the next section is about.
Export referring domains for the top three ranking pages — Page-level, not domain-level. The page that ranks is the unit you are competing against.
Find the intersect — Domains linking to two or more of them and not to you. These are publishers demonstrably willing to link within your topic — the highest-conversion prospect list you can build.
Measure their monthly referring domain growth — Twelve months of history. This tells you the pace you need to match, which is a more useful budget input than any absolute target.
Calculate their anchor distribution — Take the median across all three. That median is your benchmark, replacing every generic percentage recommendation you have read.
Identify their strongest links and where they point — Not just which domains, but which of their URLs receive them. This shows you whether they are building to money pages directly or routing through content.
Look for digital PR fingerprints — Clusters of media links appearing in the same fortnight indicate a campaign. Find the asset that earned them and you have learned what works in your market.
Step two produces the single most actionable output in link building. A publisher that has already linked to two of your competitors has demonstrated both topical interest and willingness — a fundamentally different prospect from a cold target. Campaign planning around this is covered in how to run a link building campaign.
The best illustration of why absolute benchmarks are useless comes from the search results for this industry's own head term. Looking at the US results for "link building services":
Position band | Referring domains to the ranking URL | Domain Rating | What it tells you |
Position 1 | ~1,300 | 74 | A decade of accumulated authority on one URL |
Positions 4–7 | 19–84 | 30–79 | The realistic competitive band for a new entrant |
Positions 8–14 | 1–94 | 41–91 | Some ranking URLs hold position with 1–4 referring domains |
Read that bottom row again. On a commercial head term with substantial search volume, pages sit on the first page of results with a single referring domain, while the top result holds roughly thirteen hundred. The question "how many links do I need" has no answer without specifying which position you are targeting — and the honest answer for most businesses is that positions four to seven are reachable at a fraction of what people assume.
Target | Typical referring domains to the URL | Realistic timeline |
Long-tail informational terms | 3–15 | 2–4 months |
Mid-competition commercial terms | 20–60 | 6–12 months |
Competitive commercial head terms | 60–150 | 12–24 months |
Position 1 on an established head term | Several hundred to 1,000+ | Multi-year |
Order matters here. Adding volume to a structurally skewed profile amplifies the skew, which is why the first two steps are corrective rather than additive.
If exact-match anchors sit above your competitor median, every new link should be branded or naked URL until the ratio corrects. Dilution is the only safe correction — you cannot retroactively change what has already been placed.
Fix broken inbound links pointing to 404s, collapse redirect chains, and convert unlinked brand mentions. This recovers authority you already paid for and costs almost nothing. Mechanics in how authority flows between pages.
If your referring domains are scattered across unrelated fields, prioritise publishers in your category even at lower domain ratings. Relevance outranks raw authority for most campaigns.
Work the publishers already linking to multiple competitors. Highest conversion, most direct competitive effect.
Only once the structure is sound. Cap monthly growth at roughly 20% of your current referring domain count so the curve stays continuous rather than stepped.
If the audit that precedes this turns up unexplained purchases in your history, run the seven-step audit before starting — you need to know what you are building on top of.
SERP benchmark figures are from Ahrefs SERP data for "link building services" (US), snapshot 30 July 2026, and can be reproduced with any backlink index. Organic traffic distribution figures: Ahrefs. Andrew Linksmith sells link building services. Google's guidance on link schemes is published in its spam policies.
Everything you need to know before starting a campaign. If something isn't covered here, email me — I reply within 24 hours.
A backlink profile is the full set of external links pointing to a site, described by four properties: the number of distinct referring domains, the distribution of anchor text, the mix of link types and rel attributes, and how that collection accumulated over time. Its shape matters more than its size.
Backlinks count individual links; referring domains count the distinct websites they come from. One sitewide footer link can produce forty thousand backlinks from a single domain. Referring domains is the more meaningful metric because it reflects how many independent sources have endorsed you.
In most profiles, branded and naked-URL anchors together form the clear majority, with exact-match commercial anchors a small minority. But the correct target is the median distribution of the three pages currently ranking for your terms, not a universal ratio — norms differ substantially between local services, ecommerce and software.
It depends entirely on the position you are targeting. On the US results for "link building services", pages hold first-page positions with between one and ninety-four referring domains to the ranking URL, while position one has around thirteen hundred. Check the referring domain counts of the pages you actually want to displace before setting a target.
It can be a signal worth examining. Natural link accumulation produces a mixture, because social profiles, forums, syndicated copies and press listings are commonly qualified with nofollow or sponsored. A profile that is almost entirely followed suggests every link was arranged rather than accumulated, and a qualified share somewhere between 15% and 35% is unremarkable.
I've spent 5+ years securing high DA backlinks for SaaS brands, e-commerce stores, and digital publishers across competitive niches. Every link I deliver comes from a real, independently-run website with genuine organic traffic and DA 30+ that actually moves the needle. No low-DA filler, no recycled inventory — just vetted, high-quality links with a 90%+ indexation rate that compound into lasting ranking authority.