Ahrefs, SEOmonitor, AccuRanker Explain Handling of 'Poisoned' Google Results

Ahrefs said Thursday it is "cleaning up any past results that may have been affected" by what it calls the "poisoning" of search results by Google, in which suspected SERP data providers are shown different results. The announcement did not specify which dates, reports or metrics the cleanup covers.

The statement is one of several from rank-tracking vendors describing how they detect, re-check and correct Google results that appeared incomplete or wrong during a period beginning in mid-September. The pattern is significant enough that Ahrefs estimates 5% to 20% of search results have been affected over the past few weeks, depending on the data provider. The post does not say whether that share is measured before or after its filters run.

What did Nozzle measure?

An Oct. 10 article by rank-tracking company Nozzle examined how Google's results pages appeared without many, or sometimes all, of the sites that normally rank for a given search term. Roger Montti covered the article on Oct. 14, and Nozzle updated it on Oct. 21.

Nozzle counts a first page as poisoned when less than 50 percent of the normal first-page sites for a keyword appear anywhere in the collected results. The test does not require those same sites to return in a follow-up review.

The figures:

• 8.7 percent of Nozzle-determined first pages were poisoned between Sept. 13 and Oct. 21.
• 0.17 percent were poisoned from Aug. 20 to Sept. 9.

Nozzle states the trend "is Google's response to AI scraping, and rank trackers are caught in it."

How does SEOmonitor handle suspect results?

SEOmonitor's help center states that between Oct. 9 and 13, "some of the incorrect results were recorded in the platform." The company reprocessed affected keywords on Oct. 13, the day it put a new check in place.

The company estimates that on average, around 8 percent of the results Google returns each day are incomplete or incorrect and get pulled again before being recorded. It does not say what period that average covers. On Oct. 13, SEOmonitor switched to checking each result's content against the keyword's own recent history as results arrive.

The company also disclosed corrections to stored data. It reprocessed all keywords affected by incorrect September results for the period of Sept. 25 through 28, and it blocked erroneous data from entering its product until Oct. 8.

According to SEOmonitor, approximately 10% of monitored keywords experienced issues with incorrect results from Oct. 9 through 13. The company reports most movements in its Visibility metric were minimal. It did not indicate which prior day's values were revised.

What does AccuRanker's process look like?

AccuRanker's pipeline works in stages. The company first eliminates shortened Google results pages, then uses another browser session to fetch the page again. It compares the new page-1 results to the sites listed on page 1 for at least two of the previous three days. When less than half of those sites appear, AccuRanker fetches the page again; if the second pull shows the same new sites, it records the change as real.

The company discloses the limits of its history check: approximately one in every 20 desktop pages and about one in every 12 mobile pages lack enough history data to run the check at all.

For stored data, AccuRanker applies a "sandwich check." This flags any day where less than 50% of page-1 websites match both the previous day and the next day. A fixed sample of one in every 200 keywords produced a flag rate of 30.2% for desktop keyword-days on Sept. 25, falling to 0.58% on Oct. 17. The blog post does not state whether flagged days will be revised within customers' stored data.

AccuRanker also says that when a page loses most of its normal sites, it performs a secondary fetch before the result reaches a customer's dashboard. It does not state what appears at the dashboard during that re-check. Users can request a keyword refresh within the application, which triggers a new fetch with a new session.

What are Ahrefs and BrightLocal doing?

Ahrefs says it operates filters that detect and reject abnormal results and request replacements. Alongside its cleanup announcement, that is the extent of its public detail so far.

BrightLocal, which sells local SEO software, is engaging with its current data suppliers and adding additional suppliers, so that if one provider fails, another is available. CEO Myles Anderson wrote in an Oct. 16 blog post that there is "no single fix that solves it completely" for BrightLocal or anyone else in the industry. The post does not mention verifying every individual result.

How do SERP API providers handle it?

On Oct. 14, SerpApi, which sells Google results through an API, wrote in a public issue that it had deployed an update that "should reduce the number of instances in which poisoned results are returned." It said it was still working on the problem.

An Oct. 13 comment in the same issue passed along a customer's request for a field that would flag searches without relevant results. The comment noted downstream services currently have to review each search response manually.

AccuRanker's API, AccuSERP, returns the page Google sends for a single search, so it cannot run the history check. The company says AccuSERP shares a pool of browser sessions with its rank tracker and draws first on sessions that passed the tracker's check and have stayed clean since.

Why does this matter for reporting?

Teams reviewing potential late-September or October drops should first examine how their vendor reported during that window. SEOmonitor's volatility safeguard, in place since around May, may keep showing a previously recorded rank for up to two daily updates after unusually large drops — such as Rank #4 to Rank #14, or Top 50 to Rank >99 — until it confirms the lower rank persists. SEOmonitor also links a recrawling option and a Chrome extension for comparing rank values.

Google Search Console can serve as additional evidence in a dispute. Its performance reporting help page states that average position is based on the highest-ranking result on a site, and that results vary by device, location, time and user search history. According to SEOmonitor, a high average position does not inherently mean an incorrect rank exists in its platform.

If an agency resends September or October data, it should ask the vendor whether stored data for those dates was updated. Older exports may not reflect the current tool state, so compare the two before sending anything to a client. Teams pulling Google results via API should determine which safeguards exist on their endpoint — screened sessions, history checks — and whether their internal systems validate API responses.

What remains unresolved?

As of Oct. 22, none of the vendor statements includes a statement from Google. AccuRanker writes plainly: "Google has made no statement."

Each vendor described only its own systems. None cites independent testing of how well its checks perform. The published percentages also measure different things — SEOmonitor's 8% covers results Google returned to it, while AccuRanker's sandwich figures came from a sample of stored keyword-days — so direct comparison between vendors is not reliable.

SEOmonitor says it will update its incident page as Google's changes develop and its own fixes go live. Ahrefs says it will post updates in its dashboard. Check those updates before using figures from an affected period in your next client or management report.