Skip to content
LoudCrowd
Reviewing a backlink report on screen

Spammy links and bad backlinks: how to check, and what to actually do

Most advice on this subject tells you to disavow everything. That advice has cost people more rankings than the links ever did.

Last updated 2026-09-04

What are spammy links?

A spammy link is one that exists to manipulate rankings rather than to point a reader at something useful. In practice that means links from sites built purely to host links: private blog networks, scraped directories, hacked pages, comment and forum spam, and paid link placements dressed up as editorial. The defining test is intent: would this link exist if search engines did not?

"Bad links", "spammy backlinks" and "backlink spam" all describe the same thing. What matters is the distinction between spammy and merely low quality: a small, obscure, genuinely relevant site is not spam, and treating it as such is how people damage their own link profile while trying to clean it.

How to check for spammy backlinks

You do not need a paid tool to start. Google Search Console lists your links free, and it is the only source that reflects what Google actually sees.

  • Open Search Console → Links → Top linking sites and export the list.
  • Sort by domain, not by individual link. One site linking four thousand times is one relationship, not four thousand problems.
  • Look for the obvious patterns: domains you have never heard of in languages you do not sell in, hundreds of links appearing on one date, anchor text that is a commercial phrase you never asked anyone to use, and site names containing "directory", "articles" or "submit".
  • Open a sample. Twenty domains is enough. If the page is a wall of unrelated outbound links with no readership, that is spam. If it is a small real site that mentioned you, it is not.
  • Check the dates against your history. A spike that coincides with an agency you no longer use, or with nothing at all, is the thing worth investigating.

Free tools that widen the picture: Bing Webmaster Tools has its own link report, and Ahrefs, Semrush and Moz all offer limited free backlink checks. Treat any tool's "toxicity score" as a prompt to look, never as a verdict. Those scores are vendor heuristics, not Google's opinion.

First, the thing nobody tells you

Most bad links do not matter. Google has spent the best part of a decade learning to ignore low-quality links rather than punish the sites they point at, because the alternative would make it trivial to attack a competitor.

That matters because the standard advice (audit everything, disavow aggressively) dates from 2012. Following it in 2026 is how people accidentally cut off the referring domains that were doing them good.

What actually counts as a spammy link

Judge patterns, not individual links. The signals worth acting on:

  • Paid link networks. Sites that exist only to host outbound links, usually with no real traffic and a mix of unrelated industries.
  • Exact-match anchor at volume. Hundreds of links all reading "seo agency sheffield" is not something that happens naturally.
  • Sitewide footer links. One link from every page of a site is one link, and it looks bought.
  • Irrelevant language or geography. A Doncaster plumber with four hundred links from Russian-language forums did not earn them.
  • Hacked-site links. Injected into pages whose owners do not know they exist.

And what is not spam, despite frequently being flagged as such: directory listings in real directories, links from low-traffic but genuine local sites, and old links from sites that have since declined. Low quality is not the same as harmful.

How to audit without breaking anything

1. Get the data

Search Console's link report is free and shows what Google actually sees. A third-party tool adds detail and a quality score, but the score is an opinion, not a verdict.

2. Group by domain, not by link

Three thousand links from one domain is one decision, not three thousand. Working at domain level turns an impossible job into an afternoon.

3. Ask one question per domain

Would a real person plausibly click this link? Not "is this an authoritative site". Plenty of small, unglamorous, genuinely useful sites fail an authority score and pass this test.

When to disavow, and when not to

Disavow when: you have a manual action for unnatural links, or you know a previous agency bought links and you can see the campaign in the data.

Do not disavow when: rankings dropped and you are looking for a cause. Check the core update calendar first. A drop that coincides with a known update is not a link problem. A disavow file will not fix it, and it can make things worse.

The disavow file is a chainsaw. It is the right tool for a specific job and a poor choice for general tidying. Once a domain is in there, the links from it stop counting for you as well as against you.

Where crawling fits into this

Links are also how your site gets discovered. A crawler follows what it can find, so a page nothing links to is effectively invisible. That is why orphan pages fail so reliably.

We see this constantly on sites that generated hundreds of location or service pages: the pages exist, the sitemap lists them, and nothing on the site links to them. They get crawled once, judged thin, and quietly dropped. Internal linking is not a finishing touch; it is the mechanism.

How to avoid acquiring bad links in the first place

Cheaper than any cleanup, and mostly a matter of declining things:

  • Do not buy links, including from anyone offering "guest posts on DA50+ sites" by email. That is the single largest source of the problem.
  • Ask any agency what their link building actually consists of. "We have relationships with publishers" is not an answer. Ask to see three placements from the last quarter.
  • Be wary of bulk directory submission, which is the 2011 version of the same mistake and still sold.
  • Earn the boring ones. Trade bodies, accreditations, suppliers, local press, sponsorships. Hard to fake, easy for a real business to get, and they double as the corroboration that makes you identifiable to AI assistants.

What to do this week

  • Open Search Console, check Manual Actions. If it is clean, you almost certainly do not have a link penalty, and you can stop worrying about links and look elsewhere for the drop.
  • Export your linking domains and skim the top hundred for the patterns above. An hour, at most.
  • Check that every page you care about is linked from somewhere else on your site.
  • Only then consider a disavow, and only for domains you can positively identify as paid or manipulative.

A link audit is part of the initial work in every one of ourpackages, and if you would like a second opinion before you touch a disavow file, ask us to look first.

A developer and a marketer reviewing a site
FAQ

Common questions

How do I know if I have spammy backlinks?

Pull your link profile from Search Console and look for patterns rather than individual links: sudden volumes from one domain, anchor text that repeats your money keyword exactly, links from sites in unrelated languages or industries. One odd link means nothing. A thousand of them from the same footer means something.

Do I need to disavow bad links?

Usually not. Google ignores most low-quality links automatically, and a careless disavow file does more damage than the links did. You can cut off perfectly good referring domains. Disavow when you have a manual action, or when you can see a paid-link campaign someone ran on your behalf.

What is a manual action?

A human reviewer at Google has looked at your site and applied a penalty. It appears in Search Console under Manual Actions, and unlike an algorithmic drop it will not recover on its own. You have to fix the cause and submit a reconsideration request.

Can a competitor hurt me with bad links?

It is far harder than the people selling "negative SEO protection" want you to believe. Google has been discounting rather than penalising for junk links for years. It is not impossible, but if your rankings dropped, the cause is much more likely to be a core update or something you changed.

How do web crawlers actually see my site?

A crawler requests your pages like a browser would, follows the links it finds, and queues what it discovers. Anything it cannot reach (content behind a script it does not run, pages nothing links to, URLs blocked in robots.txt) effectively does not exist to it. Most crawl problems are link problems.

Get in touch

Ready when you are

Tell us what you are trying to rank for and we will look at the site and the search results before we reply. You get a straight answer, including if SEO is not the right spend for you yet.

Call0114 697 4482Email[email protected]
Pop inCourtwood House, Silver Street, Sheffield S1 2DD

No sales sequence, no chasing. One reply from the consultant who would run the work.