Poisoned SERPs: Why Your Rank Tracker Shows Fake Drops

By Linkzy · 2026-10-29

Poisoned SERPs are Google results pages that look normal but are not what a real searcher sees, and since September 2026 Google has been serving them to rank trackers and other automated tools.

So if your rank tracker showed a sudden drop out of the top 10 this autumn, there is a real chance your page never moved at all.

This guide explains what poisoned SERPs are, who found them, which tools have confirmed it, and what it means for your SEO and link building reports.

What are poisoned SERPs?

A poisoned SERP is a search results page served to a suspected scraper where the organic listings have been swapped out.

The page still looks complete. It has ads, it has ten blue links, it loads like any other result.

But the sites in it are not the ones a normal user would get for that query.

Semrush, in its statement of 27 October 2026, describes two types:

• Clearly off: the whole page is wrong, with results that do not match the query even on page one.
• Drifting: the top looks fine, but lower results include pages that only loosely match the search.

Its example is a search in Brazil for a baby stroller model that returned a dental equipment page, because both share the Portuguese word "carrinho".

Ready to stop worrying about tracker noise and focus on links that move real rankings? Linkzy gets you placements on trusted, relevant sites, and you can check them in Search Console, not just a tracker.

Who found poisoned SERPs?

The first detailed data came from Nozzle, a rank tracking company.

According to PPC Land's write-up of Nozzle's study, Nozzle looked at 27.5 million SERPs collected between 13 September and 9 October 2026, across 49 countries and desktop, iPhone and Android.

Here is what it found:

So the poisoning was short lived on most pages, but while it lasted it knocked tracked sites out of the top 10 more than half the time.

Barry Schwartz then reported it on Search Engine Roundtable on 12 October and kept the article updated as vendors responded.

What do the poisoned results look like?

Nozzle found that social and forum sites filled the gaps.

On poisoned pages, Reddit's share of the top 10 rose from 20.8% to 28.8%, Facebook's from 28.3% to 33.8% and LinkedIn's from 6.2% to 9.3%.

Paid listings, oddly, were kept more often than usual.

That matters because a lot of people already believe "Reddit is taking over Google". Some of what they saw in their tools this autumn may have been the poisoning, not the real SERP.

Why is Google doing it?

Google has not commented publicly.

Nozzle's Derek Perkins described three things Google appears to be doing: wide IP blocking of residential proxy networks, many more CAPTCHAs, and injecting bad data into suspect pages.

Barry Schwartz's view, which he labels as his own assumption, is that the target is AI bots and scraping services such as SerpApi rather than SEO tools. Semrush points to the same likely cause: limiting large scale data collection for AI systems.

It fits a pattern. Google required JavaScript for search in January 2025, dropped the num=100 parameter in September 2025, and began testing "goto" redirect links in 2026, all of which make scraping harder.

Which SEO tools have confirmed it?

Almost every big data provider has now said something.

DataForSEO, Advanced Web Ranking and Morningscore also posted notices.

Notice how far apart the numbers are. On 29 September Nozzle measured 27% while AccuRanker measured 1.4%. They use different methods, and neither has been independently audited, so treat every figure as one vendor's view.

What poisoned SERPs mean for your SEO reports

This is the part that matters most, because a sudden dip on a chart is the fastest way to make a client panic.

If you report on rankings, three things are true right now:

1. September and October 2026 drops need a second look. The poisoning overlapped Google's spam update (24 September to 8 October), which caused real drops too. You cannot tell them apart from a tracker alone.
2. Positions beyond 10 are the least reliable. Semrush and BrightLocal both say most problems sit beyond page one.
3. Historical charts may change. Ahrefs says it is cleaning past data, so a chart you exported in early October may not match what you see today.

Not sure whether your brand is showing up in AI answers as well as Google? Get a free AI visibility audit and see where you stand.

How to protect yourself from poisoned SERP data

You do not need to throw out your rank tracker. You need to stop treating one data point as the truth.

Quick checklist:

• Compare any big drop with Google Search Console clicks and impressions for the same page.
• Check traffic and conversions in GA4 for the same dates.
• Spot check the keyword in a private browser window, from the right country.
• Look at a second tool. If only one shows the drop, be sceptical.
• Wait two or three days before reacting. Most poisoned episodes cleared within that time.
• Report odd results to your tool's support with the domain and keywords.

I go through this step by step in how to check if a ranking drop is real, and compare the two main data sources in Search Console vs rank tracker.

If you use Semrush, I also answer is Semrush accurate after all this.

What does not change: links still drive rankings

Poisoned SERPs break the thermometer, not the weather.

The things that actually move rankings are the same as they were in August: useful pages, good internal linking and links from trusted, relevant sites.

What does change is how you prove results. When tracker data is noisy, a link building report that leans on Search Console clicks and real traffic is far more convincing than a position chart.

If you want to avoid the classic reporting traps, my list of link building mistakes to avoid is a good next read.

Want placements you can measure in Search Console, not just in a tracker? Linkzy's guest post service puts your brand on real, relevant sites.

FAQ

What does poisoned SERP mean?

It means a Google results page served to an automated tool where the organic results have been swapped for different, often less relevant, sites. Real users do not see that version.

Is Google poisoning SERPs on purpose?

Google has not said. Nozzle's data shows a clear pattern of blocking followed by swapped results, and Schwartz and Semrush both suggest it is aimed at large scale scraping for AI, but that is not confirmed by Google.

Does poisoning affect my real rankings?

No. It affects what tools see, not what searchers see. Your real position only changes if Google actually ranks you differently, for example after an update.

Which rank trackers are affected?

Semrush, Ahrefs, SE Ranking, AccuRanker, BrightLocal, SISTRIX, DataForSEO, Advanced Web Ranking and Morningscore have all acknowledged the issue or posted notices. The size of the impact they report varies a lot.

Is Search Console affected by poisoned SERPs?

There is no report of it being affected, because Search Console uses Google's own data rather than scraping. It only covers sites you have verified, though, so you cannot use it to track competitors.

Final thoughts

Poisoned SERPs are a reminder that every SEO tool is an estimate built on scraped pages.

Use your tracker for trends, use Search Console for truth, and keep building the links and content that real searchers reward.

When you are ready to build authority that shows up in your own data, start with Linkzy.