Google is serving rank trackers deliberately falsified search results, and 8.0% of tracked SERPs now come back "poisoned," according to a deep dive published by Derek Perkins of Nozzle. "Since September 2026, Google has been poisoning the SERPs it serves rank trackers: complete, normal-looking pages whose results have been swapped out," Perkins wrote in a post titled Poisoned Google SERPs. "Nothing on the page gives it away. Each keyword's own history does."

The finding confirms what many SEOs have suspected for weeks: third-party rank tracking tools are struggling more than they have in a long time, and in some cases the data they return is not merely incomplete — it is wrong on purpose.

What exactly is Google doing?

Perkins, who has been the most transparent tool operator about the problems, described a three-pronged approach by Google:

• Super wide IP blocking, targeting major residential proxy networks.
• A significant increase in the use of captchas.
• Injection of bad data into suspect SERPs, so tools that successfully scrape still get wrong results in some cases.

The sequence matters. According to Nozzle's findings, Google blocks first, then poisons. The page looks completely normal — but the results do not. Social sites and forums fill the gaps where real results should be, tracked sites fall out of the top 10, and the poisoning passes within days.

The July tests of goto redirects in search results were an earlier step in the same direction, alongside more aggressive human-verification checks and the removal of the num=100 parameter in 2025. As Daniel Skov Jacobsen put it on X: "Trackers are paying for Google making SERP scraping expensive: num=100 gone, JS required. Most now sample instead of counting, so some daily swings are measurement noise."

Which tools are affected?

Several vendors have acknowledged problems publicly. Sistrix has commented on the issue, Data for SEO posted a notice on its status page, and SE Ranking stated this morning that it was "missing positions for some keywords" but that most of the issue has been resolved. Advanced Web Rankings, Morningscore and others have also posted about disruptions.

However, many tools either did not respond to inquiries or conflated the goto redirects with the newer problems — even though, as noted in the original reporting, goto redirects are the easy part for the tools to handle. The bad-data injection is the harder problem, and most tools may not detect it quickly, instead assuming everything is fine while Google feeds them false results.

How are practitioners coping?

Independent tests by SEOs show the uneven impact. Glenn Gabe documented a site hit on September 29 during the spam update, confirmed by the site owner: ahrefs picked up the drop accurately, Semrush picked it up late, and Sistrix did not show the drop at all. In an earlier September 18 test of a site that dropped nearly 100% and recovered after the September 4 unconfirmed update, ahrefs showed the recovery, Sistrix did not, and Semrush lagged.

Lily Ray posted on October 8: "Trying to share some findings on this but rank trackers are too unreliable at the moment 😩😩😩"

Others echoed the frustration:

• Alex Wolk: "Spot-checking a few queries in an incognito window has been the only way I trust a movement right now."
• Antony Deroshan: "They are able to detect bot traffic and instead of blocking them, they are poisoning the results with junk webpages."
• Marty Eigner: "Nothing matches the SERPs."
• Dr. Pete Meyers, on handing off MozCast: "Google certainly doesn't make it easy to maintain."

Confirmed vs. speculation

Confirmed: Google's goto redirect tests from July, increased captcha and verification rates, the num=100 change, and the public tool outages acknowledged by vendors. The poisoning mechanism itself comes from Nozzle's measurement — 8.0% of SERPs — which is tool-side analysis, not a Google statement. Google has not commented on its scraper-blocking tactics.

Speculation, clearly labeled as such in the original reporting: the crackdown likely targets AI bots scraping search results and Google's dispute with SerpApi, not the rank tracking tools SEOs use. The tools are collateral damage.

What to watch next

For site owners who need reliable data now, Google Search Console remains the one source of first-party ranking information — though it lacks click and query data for AI Overviews and AI Mode. Whether vendors can detect poisoned SERPs at scale, and whether Sistrix and other lagging tools close the gap with ahrefs and Semrush, will determine how trustworthy third-party rank tracking remains through the rest of 2026.