A CEO reviews next year’s marketing budget. Two lines catch her eye. One reads Search. The one below it reads AI Visibility.
Different vendor. Different dashboard. A different person presents it at the meeting.
She asks one question. What work appears on the second line that isn’t already on the first?
And nobody at the table can answer it.
Hi, I’m Jeff Payne. You’re listening to The Jeff Payne Show, Episode 109: If You Care About Search Traffic.
If you follow marketing at all, you’ve seen the argument this month. A well-known marketer posted a short video asking whether AI visibility is a distraction. It went viral. Hundreds of comments. By lunch, everyone in the industry had picked a side.
Here’s what started it. One researcher pulled a full year’s worth of analytics from a hundred companies of every size. Picture a thousand people walking through your front door. Nine hundred ninety-three came from Google. Seven came from all AI chatbots combined. Seven.
Yes, those seven converted at roughly four times the rate of the other visitors. They showed up ready. But the other 993 still accounted for more than 96% of the business. And the chatbot traffic stopped growing over a year ago. Since then, it’s been flat or drifting downward.
A separate test asked whether AI mentions shifted branded search or direct traffic. Once the numbers were shuffled, the relationship didn’t hold.
So the conclusion that traveled was: stop chasing AI. Get back to the work that sends clicks.
And if you’re a CEO who’s been paying for something with “AI” in the name, that conclusion landed on your desk with a question.
Here’s what didn’t travel. If you read the whole argument instead of the headline, the people fighting agree on more than they admit.
The researcher with the thousand dots said, in her own words, that the AI search that actually sends people to businesses isn’t the chatbots. It’s Google. AI Overviews. AI Mode. A ranking system that’s now mostly machine learning. Millions of people use AI search every day and never leave Google. Those 993 blue dots are already AI traffic.
The marketer in the viral video set Google’s AI aside as a separate question and ended by saying AI should be measured like a brand channel. A billboard, not a storefront. Nobody shared that part.
And the sharpest rebuttal conceded the point, too. The tactical layer is a distraction. The special files for robots, the markup tricks, the write-for-the-machine advice. He drew the line at recognition. Being the company the machine names.
So step back and look at what the argument is actually about. Nobody is saying the answer doesn’t matter. They’re arguing about the ruler. Clicks or recognition. Traffic or trust.
That’s a measurement fight, and it was sold to you as a strategy fight.
Here’s the strategic version. The work that earns you a place in Google’s answer and the work that gets you named by a chatbot are the same. Being the specific, best answer to the questions your buyers ask. Being described by others, in words, in places you don’t control. Being the name someone types after they stop searching.
That’s one body of work. It was never two bodies.
The industry took that single body of work and billed it twice. A search line and an AI visibility line. A rankings dashboard and a mentions dashboard. Two vendors, two monthly reports, and two people in the budget meeting, each explaining why their number went up or down.
Same answer. Second invoice.
And the distraction isn’t that someone’s watching the chatbots. The distraction is the second department. Because the second department needs its own metric to justify itself, and the only metric it can show you is seven dots out of a thousand, you end up with meetings about the seven.
Here’s the move. Go back to the CEO in the cold open and ask her the question for real. What work on the AI line isn’t already on the search line?
If the answer is “we track mentions,” that’s a report, not work. Fold it into the report you already get.
If the answer is a list of tactics aimed at the machine, that’s the layer everyone in this debate agrees is a distraction. Cut it out.
If the answer is “we’re getting the company described more often, in more places, by people who aren’t us,” that’s not an AI line. That’s the search line. That’s the brand line. It’s the same line it’s always been. Pay for it once, measure it once, and stop grading it on 7 dots.
One more thing from that same study, and it’s the part nobody argued about. 45 of those hundred companies weren’t tracking a single conversion. Nearly half. The whole industry spent a month arguing over what AI traffic is worth, yet half the companies in the data couldn’t tell you what any traffic was worth.
That’s the real headline. Not whether the machine is a distraction, but whether anyone in your building knows what happens after the door opens.
Here’s your self-audit.
How many dashboards does your marketing team report on? If the AI column disappeared tomorrow, would any of the actual work change?
Second, of the people who walked through your door this year, do you know what the 993 did once they got inside? Or has every meeting been about the seven?
I’m Jeff Payne. Thanks for listening. See you tomorrow.
A CEO is reading next year’s marketing budget. Two lines catch his eye. One says Search. The one under it says AI Visibility.
Different vendor, different dashboard, different person presenting it in the meeting.
She asks one question: What work is on the second line that isn’t already on the first?
Nobody at the table can answer it.
That question is the whole of this episode, and this month the marketing industry spent a week arguing about it without quite asking it.
THE FIGHT THAT WENT VIRAL
On October 21, 2026, Marie Haynes published a post titled GEO is a distraction. She pulled 12 months of Google Analytics data (GA4) from 100 websites of varying sizes, for more than 145 million sessions in total.
Sessions from ChatGPT came to 0.62% of the volume Google Organic delivered. Add every other LLM, and it reached 0.73%. Put differently: for every 1,000 visits, 993 came from Google and 7 came from all the AI chatbots combined.
Of the 55 sites tracking conversions, ChatGPT visitors converted at 13.8% versus 3.02% for Google organic — more than 4X the rate. Google still drove 96.5% of total conversions — 27 times what all LLMs produced.
And the collective ChatGPT referral count across those 100 properties stopped growing in August 2025 and has drifted down since.
The next day, Rand Fishkin bundled Marie’s post with two other studies in a 5-minute LinkedIn video asking whether AI visibility is a boondoggle, a sham, or a distraction. He said he left less confident than ever that visibility in ChatGPT, Claude, Perplexity or Copilot would have much business impact for most companies. The video drew hundreds of comments and a week of rebuttals.
For every thousand visitors, 993 came from Google. 7 came from every AI chatbot combined.
What DIDN’T TRAVEL
Read the whole argument instead of the headline, and the combatants agree on more than the headline suggests.
Marie’s own conclusion was that the AI search actually sending people to businesses isn’t ChatGPT.
It’s Google:
AI Overviews,
AI Mode,
and a core ranking system now largely driven by machine learning.
Millions of people use AI search every day without leaving Google. Those 993 blue dots are already AI traffic.
Rand exitly set Google’s AI faces as a separate question, and ended the video by suggesting AI be measured as a brand channel—a digital billboard rather than a storefront. The headline traveled. The ending stayed home.
Mike King’s rebuttal in Search Engine Land, “AI search is a branding channel, not a distraction,” conceded the tactical layer: the schema hacks, the llms.txt files, the write-for-the-robots advice are a distraction, and he said the debate was about brand recognition—being the company the machine names.
Nobody in this debate is arguing that the answer doesn’t matter. They are arguing about the ruler.
Clicks or recognition.
Traffic or trust.
It is a measurement fight, and it was sold to business owners as a strategy fight.
It’s a measurement fight. And it got sold to you as a strategy fight.
ONE BODY OF WORK, BILLED TWICE
Here is the strategic version.
The work that earns a place in Google’s answer and the work that gets a company named by a chatbot is the same work.
Being the specific, best answer to the questions buyers ask. Being described by other people, in words, in places the company doesn’t control. Being the name someone types after they stop searching.
That was always one body of work. The industry took it and billed it twice: a search line and an AI visibility line; a rankings dashboard and a mentions dashboard; two vendors, two monthly reports, two people in the budget meeting, each explaining why their number moved.
The distraction was never that someone is watching the chatbots. The distraction is the second department. A second department needs its own metric to justify its existence, and the only metric it can show is seven dots out of a thousand. So the company ends up holding meetings about the seven.
This is one of the best written, most informative email newsletters I’ve ever read!
Thank you for sharing your expertise, and hope you’re doing well.
Beth Zimmerman
Founder, Executive Director
The Move
Ask the CEO’s question for real: What work is on the AI line that isn’t already on the search line?
If the answer is “we track mentions,” that is a report, not work. Fold it into the report already being produced.
If the answer is a list of tactics aimed at the machine, that is the layer everyone in the debate agrees is a distraction.
Cut it.
If the answer is “we’re getting the company described more often, in more places, by people who aren’t us,” that is not an AI line. That is the search line and the brand line, the same line it always was. Pay for it once, measure it once, and grade it on seven dots.
One more number: Marie’s dataset, the one nobody argued about: 45 of the 100 sites aren’t tracking a single conversion in GA4. Nearly half. The industry spent a month fighting over what AI traffic is worth while half the companies in the data couldn’t say what any traffic is worth.
45 of 100 companies weren’t tracking a single conversion. That’s the real headline.
SELF-AUDIT
How many dashboards does your marketing team report on? If the AI column disappeared tomorrow, would any of the actual work change?
Of the people who walked through your door this year, do you know what the 993 did once they were inside? Or has every meeting been about the 7?
Sources & Credits: The traffic, conversion and plateau figures above come from a single 100-property GA4 dataset and should be read as one sample, not an industry benchmark. Kevin Indig’s twelve-site test of AI visibility against branded and direct traffic (Growth Memo, October 2026) is referenced in one line; it was the basis of Episode 98 and is not re-run here.
COMPLETE THE FORM TO
BOOK A STRATEGY CALL
"*" indicates required fields
COMPLETE THE FORM TO
BOOK A STRATEGY CALL
"*" indicates required fields
Subscribe and S—are – WE APPRECIATE YOUR SUPPORT
