Charleston clicks spiked in Search Console. It was bots.
Search Console clicks jumped 2x to 6.5x between Oct 31 and Nov 2 with flat impressions. Here are the three checks that tell you if your spike was real.
If you opened Google Search Console in the last week and saw the best click numbers your Charleston business has ever posted, do not call your accountant yet. Those clicks were almost certainly not people.
Here is what actually happened. Starting Friday October 31 and peaking on Monday November 2, sites all over the place saw their click counts in the Search Console performance report go vertical. Joy Hawkins flagged it first. Barry Schwartz wrote it up on Search Engine Roundtable on November 3, and by then Glenn Gabe, Dan Nutter, and a pile of other people were posting their own charts. Hawkins confirmed on November 4 that hers was back to normal, which means the whole thing was a four-day event that is already over and already sitting in your report.
Marie Haynes went and did the work the rest of us were too busy to do. She pulled three properties she monitors and published the numbers in Search Engine Journal on November 5. November 2 clicks came in at roughly 2.1x to 6.5x the previous seven-day average. One site hit 6.5x overall and 14.1x on desktop alone. Ninety-one percent of the extra clicks on that site were desktop.
And then the detail that settles the argument.
The extra clicks landed on pages ranking between 11th and 55th. Page two through page six of Google. On one site, desktop click-through rate at those positions went from 0.20% to 2.89%, and Haynes found CTR figures between 45% and 100% down in positions 12 through 54.
A hundred percent click-through rate at position 40 means every single person who saw that result, buried four pages deep, clicked it. Nobody does that. I have never once scrolled to page four of Google, and neither have you, and neither has the woman in West Ashley trying to find somebody to look at her water heater.
Barry covered it in his weekly video recap, along with the rest of a busy news week:
Nobody knows exactly what it was, and Google has not said
I want to be straight about the uncertainty here, because half the writeups floating around have already promoted a theory to a fact. Google has not commented. Barry asked and got nothing back.
The candidates on the table:
• Headless-browser bots clicking through results at scale • CTR manipulation networks, which are a real and ugly industry • AI agents doing query fan-out, opening every result to read it • A gap or temporary bug in Google's own bot-click filtering
Glenn Gabe's read is that it looks like automated tools researching products and services, which fits the AI-agent theory, and that one makes intuitive sense to me. When somebody asks ChatGPT or Gemini "who does crawl space encapsulation near Mount Pleasant," something on the other end opens a stack of pages, and it does not care whether a page sits at position 3 or position 43. That produces exactly the fingerprint Haynes found.
I am not going to tell you that is confirmed, because it is not. What is confirmed is the shape of the data, and the shape says non-human.
The three checks that tell you if your spike was real
This is the part that matters for your business, and you can do all of it in about ten minutes.
One. Put Search Console next to GA4. Open your performance report, note the click count for October 31 through November 3, then open GA4 and look at organic search sessions for those exact dates and those exact pages. If Search Console says clicks tripled and GA4 says sessions were flat, you have your answer. Bots that click a search result and never render a page do not fire a GA4 session.
Two. Look at the average position of the pages that got the clicks. In the performance report, open the Pages tab, click the filter icon in the table header, choose Position, set it to greater than 10. If your new clicks are concentrated on pages sitting at position 11 and below, those are not customers. Real Charleston search traffic lands on page one or it does not land.
Three. Check your phone log. The least technical check and the most important one. Did inquiries go up in that window? Did the phone ring more? Did anybody fill out a form? If clicks went up 400% and your call volume did not move at all, the clicks were not people.
Then do one more thing: write a note somewhere. A comment in your reporting spreadsheet, a line in a doc, anything that says "Oct 31 to Nov 3 click spike, bot traffic, ignore." Because in February when you are comparing quarter over quarter, that spike will still be sitting there making your real numbers look like a decline.
At this point a lot of people are thinking: so what, it is free data
Fair. A fake spike costs you nothing directly. Here is where it actually hurts.
Business owners make real budget decisions off a Search Console chart. A roofer sees clicks double, decides the SEO is finally working, and cancels the thing that was actually bringing in jobs. A dentist on Daniel Island sees the same chart from an agency and renews a retainer that has produced nothing in six months. The number went up, so the story wrote itself.
That is the cost. Not the fake clicks. The real decision made on top of them.
This keeps happening because Search Console has quietly stopped being a clean record of what people searched. I wrote a few weeks back about how AI Mode follow-up answers now show up in your query report as ordinary searches, which is how I ended up with the word "yes" ranking at position 6 on a wildlife removal site. Same underlying problem, different column. The report is increasingly measuring machines talking to machines, then handing you the total as if it were demand. Bot traffic cuts the other way too: last summer I wrote about Cloudflare's defaults quietly blocking Googlebot on sites that had opted out of AI training, and almost no owner has a clear picture of which bots are hitting them.
What to judge your search work on instead
Clicks were never the right number. They are the easiest number, which is a different thing.
For a Lowcountry service business the chain is short: impressions in your service area, position on the terms that carry buying intent, calls and form fills, booked jobs. Impressions held flat through this entire spike, which is your tell that search demand never changed. The market did not find you over a weekend in early November.
When we build the measurement side of our Charleston SEO work, calls and booked jobs lead the dashboard and clicks sit further down where they belong. Not because clicks are worthless, but because they are the metric most easily polluted by something that is not a customer. This month made the case better than I could have.
One more reason to care right now. The November 2026 core update started rolling out on November 10 and takes up to two weeks. Your report is about to be genuinely noisy, and it is noisy on top of a four-day bot artifact still sitting in your trailing averages. Read impressions and position in that window, not clicks.
Go pull October 31 through November 3 in Search Console tonight, hold it next to your GA4 sessions and your call log, and label it in your notes before you forget. Ten minutes now saves you a bad decision in February.