The disavow file is the most over-prescribed fix in SEO. Rankings dip, a tool paints part of the backlink profile red, and someone uploads a list of domains to Google hoping it heals the drop.
In most of those cases the file does nothing, and in the worst cases it cuts links that were quietly helping.
Here is when a disavow genuinely belongs in your plan, and when it does not.
Key Takeaways
- Google’s own documentation limits the disavow tool to sites with a manual action (or one likely coming) from links they built; most sites never qualify
- Since Penguin 4.0 in 2016, Google devalues spammy links instead of demoting the site, and the December 2022 SpamBrain update neutralizes them at scale
- Disavowing from a tool’s “toxic score” regularly cuts real, earning links; that mistake costs rankings while the spam was costing nothing
- A recurring disavow habit is really a self-report of your own link building; if the profile keeps needing cleaning, the fix is to stop building manipulative links, not to file a longer list
- A ranking drop is not evidence of a link problem; check the Manual actions report and the actual cause of the drop first
- The legitimate disavow cases are narrow: a manual action for unnatural links, or a paid-link and PBN history you know exists
Where the disavow reflex came from
The reflex is a leftover from 2012. Penguin used to punish sites for bad link profiles, recoveries could take until the next refresh, and an entire industry of link audits and toxic-link scores grew around that fear.
Back then aggressive cleanup made real sense, because a bad profile could hold the whole site underwater for months.
The tools kept the fear alive long after the algorithm changed. A dashboard that flags a chunk of your backlinks as “toxic” creates urgency out of nothing, and a vendor selling monthly link audits has little reason to correct the impression.
So the disavow file became the default response to any ranking drop, which is exactly the wrong lesson to have carried forward.
What Google’s disavow documentation actually says
Google’s disavow documentation is unusually blunt. The tool is meant for sites that have a considerable number of spammy or paid links and a manual action because of them, or a real reason to expect one.
It says most sites will not need the tool at all, and it warns that used incorrectly, the tool can harm your site’s performance.
The tool itself says the same thing. It does not even live in the main Search Console navigation anymore; it sits behind a separate page with a caution banner across it.
I generally treat Google guidance as an interested claim rather than settled fact, and I will get to that caveat in a moment.
But notice what this doc is doing: Google gains nothing by talking you out of a disavow, so a warning this direct is worth taking seriously.
What happens to spammy links now, without a disavow
Two documented changes did most of the work:
- Penguin 4.0, September 2016: Google moved from demoting sites to devaluing the spam itself, adjusting ranking on spam signals rather than punishing the whole site.
- The December 2022 link spam update: SpamBrain started neutralizing unnatural links at scale, so the links simply stop counting.
Neutralized means the links stop counting in either direction. Any link equity they were faking evaporates, but no penalty follows.
John Mueller
John MuellerSearch Advocate, GoogleGoogle's Search Advocate and the main on-the-record voice of Google Search Relations.LinkedInXSearch Central has been blunter than the docs. He has said Google’s systems are really good at dealing with random spammy links, and he calls routine disavow work without a manual action a waste of time.
He has also accused disavow service sellers of cashing in on clients who don’t know better.
Now the caveat I promised. “We handle it” is a claim from the company that benefits when you stop investigating, so I don’t take it at face value either.
Some practitioners still report recoveries after large cleanups, almost always on sites with a genuine manipulation history, and those reports are hard to verify causally.
The honest reading: devaluation handles random junk well, and the disputed territory is deliberate, concentrated link schemes, which is exactly where Google’s own doc already points the tool.
| The belief | What the evidence supports |
|---|---|
| Spammy links are dragging my rankings down | Google devalues or neutralizes most junk links automatically; they usually pass nothing at all |
| A high toxic score means I must disavow | “Toxic” is a tool metric, not Google’s; tools flag harmless links and real earning links alike |
| Disavowing is harmless insurance | Google’s own doc warns incorrect use can harm performance; over-broad files cut real links |
| My drop must be a negative SEO attack | Drops mostly track algorithm updates, indexing shifts, or measurement; link attacks rarely move rankings on their own |
The real cost of disavowing from a tool’s toxic score
The disavow file gives you no feedback. Google treats it as a strong suggestion to ignore those links, takes weeks to reprocess them as it recrawls, and never tells you what changed.
You are steering blind, which is a bad place to make aggressive cuts from.
That is what makes over-broad files expensive. Toxic scores are computed from proxies: spammy-looking anchors, low authority metrics, odd TLDs, and real, earning links fail those proxies all the time.
The pattern that keeps showing up in audits and community post-mortems is a site disavowing hundreds of domains off a tool export, a handful of which were quietly passing value, and rankings sag with nothing “toxic” left to blame.
The file also keeps costing after upload, because almost nobody remembers it exists. A forgotten disavow file from a previous SEO can sit there suppressing legitimate links for years.
If you inherit a site, checking for an existing disavow file belongs on your first-week checklist.
What a recurring disavow habit says about your link building
The tool’s own scope is the tell. Google built it for links you, or an SEO you hired, actually made: the paid placements, the PBNs, the anchor-stuffed guest posts, paired with a manual action or one you can see coming.
So a fat disavow file is really a paper trail of your own link building. Reaching for it every month is not a sign Google is misjudging you; it is a sign the junk is still being built.
If your backlink profile keeps needing a disavow, the honest question is who keeps building the links you keep disavowing. The fix is not a cleaner list, it is not building manipulative links in the first place.
To be fair, the file is not quietly “outing” you to a reviewer. Google devalued spammy links from 2016, and since the 2022 spam update it effectively ignores them, so random junk usually passes nothing in either direction.
The real risk from a heavy, self-built spam profile is a different one. As Search Engine Land reported, a strong enough pattern of spammy links can push Penguin to distrust the whole site, rather than ignore the links one by one.
That is the line Google’s doc and Mueller’s “waste of time” point both circle. Routine disavowing fixes nothing, but a real, concentrated manipulation history is a genuine liability, and the cure is to stop building it, not to keep filing it away.
When a disavow file is the right call
A manual action for unnatural links is the clear case. A human reviewer flagged your profile, it shows in the Manual actions report in Search Console, and cleanup plus a disavow of what you cannot get removed is part of the reconsideration path.
This is the job the tool was built for.
The second case is a link history you know is dirty because you, or an SEO you hired, built it: paid placements, a PBN, mass guest posts with commercial anchors.
If that past is concentrated enough to invite a manual action, disavowing ahead of one is the doc’s own “about to get one” scenario, and doing it narrowly is cheap insurance.
Negative SEO panic is mostly not a case. SpamBrain exists precisely to neutralize hostile junk, and Mueller’s criticism was aimed at the services selling that panic.
If a genuinely concentrated attack worries you, a targeted disavow of that one pattern costs little; just don’t let it grow into a monthly toxic-link ritual.
What to check before you upload a disavow file
Most drops that get blamed on links turn out to be something else: an algorithm update window, an indexing recount, or a measurement artifact.
I have covered how Google can engineer drops that look like penalties and why a falling indexed-pages count is usually bookkeeping, and I wrote on LinkedIn about why I stopped reacting to ranking drops at all. The disavow question deserves the same discipline, in this order:
The pre-disavow checklist, in order
- Open the Manual actions report in Search Console; a real link penalty is named there
- Diagnose the drop itself: update timing, indexing changes, measurement noise
- Ask what you (or a past SEO) actually built: paid links, PBNs, anchor-stuffed guest posts
- Only a manual action or a real manipulation history earns a disavow; otherwise leave it alone
So, should you use a disavow file in 2026?
For most sites, no. The tool solves a problem most sites do not have, the evidence says Google already devalues the junk your toxic-score report is worried about, and the downside of an over-broad file is real while the upside is usually zero.
If this were my site and I had a manual action, or knew there was a paid-link past behind it, I would disavow without hesitation, narrowly and carefully; that is the tool doing its actual job.
Everything else is a placebo tax, and the hours are better spent on the pages, the internal links, and the content that actually earn rankings.
Not sure whether your links are really the problem?
Don’t hesitate to contact us or email me; I’ll tell you honestly whether a disavow would help your situation or just waste your time. Remember, diagnosing the drop correctly matters more than reacting to it fast.
Update Logs
11 Jul 2026
- Added a section on what a recurring disavow habit signals about your own link building, including the Search Engine Land finding that a strong pattern of spammy links can lead Google to distrust the whole site rather than ignore them one by one.
Want our posts to show up more often on Google?
One step & Google will surface this site in your Top Stories.
