Google is poisoning SERPs to stop scrapers, and rank trackers can't tell the difference

Google has been serving rank-tracking tools deliberately inaccurate search results since mid-September, according to research published this week by rank-tracking vendor Nozzle and corroborated by a second vendor, DemandSphere, plus independent SEO trade press — meaning weeks of ranking dashboards marketers relied on to judge organic performance may reflect Google's anti-scraping defenses rather than real position changes.

What changed

Nozzle, a SaaS rank-tracking platform, analyzed 27.5 million search results pages across 49 countries between August 20 and October 9. It found that Google started blocking scraper requests harder on September 13, then began returning results pages that look complete and normal but contain the wrong sites — what Nozzle calls poisoning rather than blocking.

"Since September 2026, Google has been poisoning the SERPs it serves rank trackers" — Nozzle

The share of poisoned pages rose from a 0.17% baseline to 8.0% overall, peaked near 25% between September 27 and 29, and has held at roughly 5% to 11% a day since October 1. On a poisoned page, a tracked site fell out of the top 10 results 58.6% of the time, compared with 2.7% normally — and AI Overviews, "people also search for" boxes and image packs all appeared less often, replaced by more results from Reddit, Facebook and other forums.

The disruption isn't evenly spread. Nozzle's country breakdown put Austria at 18.8% of SERPs poisoned, South Africa at 18.4% and Australia at 11.5%, against 8.5% for the US and just 3.0% for the UK. Legal-industry searches were hit hardest of any category, at 15.9% overall and 51.3% on the single worst day, and "near me" queries poisoned at 11.4% versus 7.8% for everything else. Desktop, iPhone and Android all showed similar rates, so this isn't a mobile- or desktop-specific bug in one tracker's crawler.

A second vendor, and SEO's own press, back it up

This isn't one vendor's theory. A separate rank-tracking company, DemandSphere, reported its own disruption starting around September 11 and told ppc.land that roughly 11% of its daily checks were coming back spurious, using a different methodology and keyword set than Nozzle's — a different number, but the same sequence: Google blocks first, then swaps in misleading results on pages it suspects are automated.

Search Engine Roundtable rounded up tool vendors and SEO practitioners dealing with the fallout in real time. Sistrix, DataForSEO, SE Ranking, Advanced Web Ranking and Morningscore all posted notices about tracking problems, and SEO consultant Lily Ray described rank trackers, in a post the outlet cited, as "too unreliable at the moment." Other practitioners compared tools directly: Ahrefs picked up the disruption, Semrush lagged behind it, and Sistrix missed it entirely. Search Engine Journal, which is independently owned, reported the same Nozzle data separately and noted that SERPs also keep only 20.7% of the same sites across repeated checks on a poisoned page, versus 88.5% normally.

Who this affects

Anyone who reports organic search performance using a third-party rank tracker is exposed right now, which covers most software marketing teams. Semrush, Ahrefs, Sistrix, SE Ranking, Advanced Web Ranking, Morningscore and the trackers built into agency reporting stacks all pull from the same scraped SERPs Google is now actively disrupting. A dashboard showing a page-one keyword suddenly vanishing, or a competitor jumping into a spot it never held, may be measuring Google's bot defenses instead of your content, your backlinks or your last algorithm scare.

The practical risk is downstream: a marketer who sees a ranking crash and escalates it to the content or SEO team — or worse, to a board update — is now at real risk of chasing a problem that doesn't exist, while a genuine drop gets lost in the noise around it.

Teams running international SEO have the most to untangle, since the poisoning rate in Nozzle's data varies by more than six times between the UK and Austria. A global software company comparing regional performance off the same tracker right now could be comparing real differences in one market against pure measurement noise in another, with no way to tell which is which from the dashboard alone.

• Hold reporting for 2-3 days before treating a sudden drop as confirmed
• Cross-check any alarming move against Search Console impressions and clicks
• Ask your rank-tracking vendor directly whether they've detected and filtered poisoned SERPs
• Avoid citing single-day rank-tracker swings in client or executive reporting until the vendor confirms its data is clean

None of the outlets covering this have an on-record comment from Google explaining or confirming the behavior. Nozzle and others read it as a side effect of Google hardening its defenses against AI scrapers and bot traffic generally, not a move aimed at SEO tools specifically — but the effect on marketers reading those tools is the same either way.