SEObserver now filters spam backlinks from your analysis
Top backlinks in SEObserver now automatically detect and grey out spam links. More signal, less noise: Majestic's exhaustiveness, now filterable.
The problem: noise in your backlink data
When you analyze a site’s link profile, you want to see the real links. The ones that matter. The ones that influence rankings.
But the more exhaustive your data source, the more noise it surfaces. Automated link farms, mass-generated domains like domain-list-941 or seo anomaly, pure spam with zero analytical value.
These links aren’t dangerous. Google ignores them. But they pollute your analysis. When your top 20 backlinks are link farms, you spend cognitive effort mentally filtering them out before you can actually work. That’s unnecessary overhead when your attention is your scarcest resource.
Until now, you had to choose: use the most complete source and accept the noise, or settle for a partial source and risk missing important links.
What we tried (and why it wasn’t enough)
Before landing on the current solution, we explored several approaches.
Filtering by indexed pages. This was a fairly revolutionary approach — we’re actually still the only SEO tool that does it. It’s very expensive, but that’s how it is. The logic is simple: if the source page is in Google’s index, the link is legitimate; if the page isn’t indexed, you can ignore it. Two problems. First, spam does get indexed — temporarily, but enough to pass the filter. Second, many perfectly legitimate links aren’t indexed (or lose their indexation over time): deep pages on authority sites, articles with no internal linking, partially noindexed pages. Sure, they’re not taken into account by Google, but the problem is different from pure spam, and the strategy to adopt is different in both cases. So declaring that not-indexed = spam means throwing out signal with the noise.
Building our own link index. Since summer 2025, we explored building our own backlink database from URLs that rank in Google, starting from FR-only data. We spent months on it. Conclusion: it doesn’t replace a source that has been crawling the web in depth for 15 years. We were finding a fraction of the coverage. Historical exhaustiveness, and building a link scoring system, can’t be rebuilt in a few months. We haven’t abandoned this path though — but what we’ve learned from it is taking a different and even more powerful shape than what we originally envisioned. More on that soon.
Our approach: filter the noise, keep all the signal
We flipped the problem.
Rather than reducing coverage to get less noise, we keep the most exhaustive source available and build a detection layer on top.
Here’s how it works:
- Automatic detection via a blacklist of known spam domain patterns (link farms, automated networks, mass-generated domains)
- Metrics analysis at both URL and domain level, compared against a sample of legitimate sites to identify anomalies
- Manual whitelist and blacklist of sites clearly identified internally, continuously refined
- Spam links are greyed out, not deleted. They remain in your data. Nothing is hidden. Full transparency.
- Three views to choose from: “All” to see every link, “Zero spam” to hide the noise, “Spam” to isolate only the detected spam links.
- Your real backlinks finally stand out, free of visual pollution.
Before / After
The change is immediate. Where your top backlinks were flooded with domain-list-941, seo anomaly and other index spammers trying to sell you their expired soup through telegram groups, you now see the links that matter straight away.
Columns are enriched: link anchor text, quality metrics, Trust Flow visible at a glance. The analysis that used to require manual sorting becomes instant.
Why we don’t delete spam links
This is a deliberate choice. Deleting data — even spam — would be an arbitrary decision. Majestic’s CEO told us directly when we discussed how to handle this problem: these links genuinely exist on the web, and they need to be represented somewhere.
Our role isn’t to decide what exists or not. It’s to give you the tools to separate signal from noise. The three views are there for that: switch to “All” when you want to verify, stay on “Zero spam” when you want to work.
FAQ
What if a legitimate link is marked as spam? Report it. Every false positive improves detection for everyone. Our goal is zero false positives.
Are spam links factored into the metrics? The displayed metrics (Trust Flow, Citation Flow) come from Majestic and include all links. The filter is purely visual. We’re considering recalculating metrics on demand with spam excluded, but it involves latency and significant resource consumption at the scale of the entire database. It’s something we’re actively thinking about.
Is it enabled automatically? Yes. The filter is active for all accounts, effective immediately. No configuration needed.
Can I disable the filter? Yes. The “All” button displays every link, spam included.
The most complete link database.
Now also the cleanest.
Reducing coverage to reduce noise is easy. Anyone can do that. Keeping the exhaustiveness of 15 years of continuous crawling AND filtering spam — that’s the real challenge. That’s what SEObserver does today.

More signal, less noise.