Early November Search Console click spikes look like bot traffic

Google Search Console reported click spikes of 2x to 6.5x in published data from affected sites between 31 October and 2 November, and up to 10x in one SEO's reports, while impressions on the measured sites barely moved. The click patterns point to automated activity rather than people, though Google has not confirmed a cause. By 4 November clicks were falling back to normal on the sites SEOs were tracking.

Marie Haynes tracked click spikes of up to 6.5x

The first reports came from SEOs comparing properties on 2 and 3 November. Dan Nutter saw spikes on 31 October across a range of sites and industries in a mainly UK client base, sometimes 10 times a normal day's clicks. Joy Hawkins posted a chart showing a vertical jump in clicks with impressions flat, and said the CTR patterns made her fairly confident it was bot traffic. Glenn Gabe saw it on a few properties and said most did not show it.

SEO consultant Marie Haynes then published row-level data from three affected properties she monitors, pulled from the Search Console API with fresh, unfinalised data through 2 November. Site A, a multi-location service business, went from about 50 clicks a day to 327 on 2 November, a 6.5x jump (14.1x on desktop). Of those 327 clicks, 297 were desktop, spread across 120 pages, and the traffic was from the United States.

Site B, a regional service business, went from about 16 clicks a day to 66, a 4.1x jump (7.9x on desktop), with 55 of the 66 clicks on desktop across 36 pages, again from the United States. Site C, a technical Q&A site, went from about 68 clicks a day to 144. Its Canadian clicks jumped from about 11 a day to 75 while US clicks stayed flat at 67, and 40 of those clicks hit a single page across 23 reworded versions of one query.

Why are page 2 results getting clicks?

Organic clicks normally concentrate in the top three to five results. On 2 November the click-through rate on these sites exploded on results ranking between position 11 and position 55. On Site A, desktop clicks went from a six-month daily median of 21 to 297 and sitewide desktop CTR rose from 0.20% to 2.89%. Mobile clicks held at 30, in line with the 25 to 40 of prior days.

Site C shows the other pattern. One long-tail troubleshooting page that usually gets zero or one click a day, and had never taken more than seven in a day across the previous six months, logged 13 clicks on 1 November and 40 on 2 November, all 40 of them from Canada. The queries were 23 slightly reworded versions of the same troubleshooting question, including number permutations such as 3 times, 4 times, 5 times, 7 times and 10 times, almost all at 100% CTR between position 10 and 33. Haynes reads that as an AI agent or research tool generating variants of one question and opening the candidate pages in a headless browser.

Haynes offers four explanations for bot clicks

According to Haynes, loading a results page, as a basic rank tracker or scraper does, only adds impressions. A click is logged only when something actually follows the tracking link Google attaches to each result. Haynes lists four candidates: scrapers or headless browsers that click through page 2 to 6 results to fetch page content or resolve Google's redirect to the final URL; CTR manipulation networks that click decoy results on deep pages to make fake desktop profiles look human, hitting innocent sites in the same niche; AI agents running query fan-out retrieval; or a gap in Google's bot-click filtering, which normally strips known automated clicks before they reach the report.

Gabe, after checking more properties that spiked, said it looked like AI agents or bots researching products or services. Google was asked for comment and had not responded by the time the spike ended. Early reports also described three sites where the Search Console API showed the spike but the web interface showed it on only one of them, so an API-fed dashboard may carry the jump even where the Search Console report looks clean.

Rank trackers may show goto redirect URLs

Since late August, Google's search results have carried www.google.com/goto redirect links in place of direct URLs, according to DataForSEO, which sells search result API access. The links cannot be decoded, so a tool has to follow each one to learn where it goes. Google has been limiting how fast the redirects on a single results page can be followed, according to DataForSEO and Serper. Serper's status page first logged the issue at 1:30 am UTC on 30 October, and SerpApi recorded goto links in organic results on the same date.

The dates line up with the click spike, and one site owner asked whether rank trackers resolving the goto parameters could be behind it. Haynes lists resolving Google's redirect as one way a scraper would register a click. No one has shown that the two are connected, and Google has not announced the goto limit or explained the spike.

There is a separate practical consequence. Some rank tracking reports and in-house URL-level reports from about 30 October onward may show google.com/goto redirects instead of destination URLs. Before you report a landing page as having lost a ranking, check whether the stored URL is a Google redirect and ask your provider how it collected the data. It is the second data-quality problem for rank trackers in a fortnight, after Semrush said Google was serving poisoned results to rank trackers.

Isolate bot clicks with Performance report filters

Search Console has no button that strips only the bot clicks from the top chart. These checks work.

• Open the Performance report and compare 31 October to 3 November with the week before. A jump in clicks with flat impressions is the signature.
• Tick Average CTR and Average position above the chart, open the Pages tab (Search Console hides most one-off single-impression queries, so the Queries tab will not show the pattern), click the filter funnel in the table header, choose Position, Greater than, 10, then sort by clicks descending. Page 2 to 6 URLs logging 30% to 100% CTR are the affected ones.
• Switch the Device filter to Mobile. It hides your real desktop traffic too, but it shows a client the untouched trendline. Site A's mobile clicks were 30 on 2 November against 27 to 40 on prior days, and Site B's were 11 against 11 the day before.
• Open the Countries tab. All of Site C's extra clicks were Canadian: about 11 a day became 75, while US clicks held at 67. For an Australian business, a sudden jump from the US or Canada is an obvious tell, and you can set the Country filter in the top bar to just the countries you serve, such as Australia (Search Console now lets you select several countries at once).
• If only one or a few pages spiked, use Add filter, Page, URLs not containing, and enter the slug or a subfolder such as /locations/. That removes the spiked pages from the chart and keeps the rest of your traffic intact.

Keep the spike out of CTR and conversion comparisons

Haynes' advice is that there is nothing to fix on the site: no page changes, no disavow, and no alarm when the clicks drop back to baseline. Annotate early November 2026 in your reporting so anyone reading it later knows why the line jumped.

The spike also needs to come out of any comparison that uses clicks as a denominator. Any conversion rate calculated against Search Console clicks will look lower for the window on affected sites, because the extra clicks were not real visitors. One site owner noted the clicks did not seem to have translated to actual traffic on the site, and one person on X suggested checking GA data for 31 October as well. It is the same discipline that applied to October's Search Console data delay, when the performance report ran more than 64 hours behind and then caught up, and to the GA4 audit of Christmas orders that found script-running bots inflating one store's GA4 sessions while its Search Console clicks did not move: a metric that moves with no matching change in enquiries or sales is a tracking question first.

Pull the Performance report for 26 October to 4 November before the monthly report goes out, run the Position greater than 10 check on the Pages tab, and footnote the window if you find the pattern. If you want a second set of eyes on a client's Search Console data before it goes out, Mojo Dojo can run that check as part of an SEO audit.