Link footprints: the patterns that expose a link scheme

A link footprint is a repeated pattern that shows a group of links or sites share one origin. One link from one site says little. Twenty sites on the same hosting, with the same theme, all linking to the same three commercial pages, say a lot. Schemes are exposed by repetition, not by any single link.

This lesson describes the patterns so you can recognise them, in a vendor’s inventory or in your own profile. It is a reference for assessing risk, not a guide to hiding a network.

Why footprints matter

Google is reported to have said, with its December 2022 link spam update, that SpamBrain detects sites that buy links and sites used to pass them on. A detection system of that kind works on patterns across many sites. When a group is identified, the typical result is devaluation of the whole group at once. The links still exist and pass nothing.

For a buyer, this turns a footprint into a pricing question. A link from a network with obvious footprints has a short expected life, whatever the seller charges.

The five groups

Group	Examples	Who can see it
Infrastructure	Shared IP addresses, nameservers, hosting, registrars, registration dates	Anyone with DNS and WHOIS tools, and Google
Site build	Same themes and plugins, same analytics, ad or tag manager IDs, same Search Console account	Anyone viewing source. Account links are visible to Google
Content and outbound links	No about page, no author, 5 to 30 posts, every post linking to a commercial site, unrelated niches side by side	Anyone who reads the site
Link graph	Many sites linking to the same set of money sites	SEO tools unless blocked. Google always
Anchors and timing	Exact-match anchors, bursts of similar links, hockey-stick growth in referring domains	SEO tools and Google

Infrastructure footprints

The oldest footprints are technical. Sites that share an IP range, nameservers or a hosting account are trivially connected. So are domains registered on the same day at the same registrar.

Specialist “SEO hosting” that offers many IP addresses was built to break this pattern, and it often creates a new one, because the addresses sit in sequential ranges that few real sites use. Operators of better-built networks use ordinary hosting companies and content delivery networks for that reason.

Site build footprints

Sites made in a batch look alike. The same theme, the same plugins and the same page structure repeat across the set. The strongest signals in this group are shared account identifiers: one analytics ID, one advertising ID or one tag manager container used on several sites ties them to a single owner with certainty.

Anyone can check these by viewing the page source. Tools that search the web for a given ID make the check fast.

Content and outbound link footprints

This group is visible without any tool. Respona’s July 2026 guide to PBN links lists signs such as no author bios, thin content, identical themes and unrelated outbound links. Practitioners add others: no about page, a small number of posts, and every post linking out to a commercial site.

Two patterns deserve attention.

Unrelated niches. A marketing blog that links to crypto, casino and CBD sites has no editorial logic. The only thing those links share is that someone paid.

History mismatch. A domain that was a local sports club site until last year and is now a finance blog has been repurposed. Archive services show the switch, and the 2024 leaked Google documentation is reported to indicate that Google keeps a history of page versions. Expired domains explains why this matters.

The same signs identify link farms that are not private networks at all. BuzzStream’s June 2026 study rated 96.2% of roughly 257,000 guest post sellers as low quality.

The link graph is the biggest footprint

Every other footprint can be hidden with money and care. The link graph cannot, because it is the purpose of the scheme.

If thirty sites exist to link to the same five money sites, then those thirty sites share a set of outbound targets that unrelated sites would almost never share. Analysts describe this as the main way a network is found: the system models relationships between sites, not just the sites.

The same logic applies to bought placements. If you and forty other buyers all purchase from one seller list, your profiles overlap. A competitor can find the list by comparing who links to several of you, which is ordinary link gap analysis. So can a reviewer.

Blocking crawlers

Many networks block the crawlers of Ahrefs, Semrush and similar tools in robots.txt or at server level. This keeps the links out of competitors’ reports.

It has two limits. It hides nothing from Google, which does not rely on those tools. And the block is a footprint in its own right, because a normal publisher has no reason to refuse SEO crawlers. When a site you are offered a link on blocks them, ask why.

This is also why third-party tools under-report link profiles in markets where networks are common.

Anchor and timing footprints

Patterns in the links themselves:

• A high share of exact-match commercial anchors.
• Many links with near-identical anchors appearing in a short period.
• Referring domains that jump from a few hundred to a few thousand in months and then go flat.
• Anchors in one language on sites written in another.

The 2024 leaked documentation describes fields that record spikes of similar anchors, according to analyses by iPullRank and Hobo. The fields are documented. How they are used is not confirmed. See anchor text and link velocity for what normal looks like.

Outreach and vendor footprints

Footprints are not limited to networks.

• Pitch templates. Editors recognise the same opening line sent by hundreds of senders, and fake personas are a known pattern.
• Author bios. The same guest author name and bio across dozens of unrelated sites.
• Seller lists. A domain that appears in circulated price spreadsheets and on several marketplaces is sold to everyone.
• Placement style. A paragraph inserted into an old article, with a commercial anchor, in the same position on every site.
• Reciprocal patterns. Two sites that link to each other repeatedly, or rings of three. See link exchanges.

How to check your own exposure

1. Export your referring domains and sort by first seen date. Look for batches.
2. For each batch, open ten sites. Do they share a theme, an author or a seller?
3. Check what else those sites link to. If the same competitors appear every time, you share a source.
4. Review the anchor mix for the batch against your whole profile.
5. Ask any vendor whether they own the sites they place on. A vendor-owned network is a footprint you are renting.

A backlink audit covers the full review, and the backlink tools comparison lists software that can run these exports.

When the pattern is found

Expect the links to stop counting, usually without notice. Rankings that rested on them slide. A manual action is possible where the pattern dominates a profile.

One rule holds across every case: never assume a firewall between assets that share footprints. Sites connected by hosting, accounts or link patterns can be treated as one group. Private blog networks covers how that plays out for networks specifically.

Where to go next

How Google detects paid links covers the detection side in more depth. Terms such as PBN, money site and referring domain are defined in the glossary.