Wanamaker said half his advertising was wasted — he just never knew which half. He spent it anyway. Built one of the great department store brands in American history on that bet. A hundred years later, CEOs are staring at a new channel, watching it send almost no clicks, and concluding it doesn’t work. They’re about to make Wanamaker’s mistake in reverse — walk away from the half that’s actually working, because they can’t trace it. Hi, I’m Jeff Payne. You’re listening to The Jeff Payne Show, Episode #33: The Click Isn’t Coming Back. AI search stopped being optional a while ago. ChatGPT crossed the 1 billion monthly user mark this year. Google’s AI Mode did too. Seventy-one percent of software buyers now research with a chatbot before speaking with a salesperson. So every CEO wants the same slide: what did AI search do for revenue? And almost nobody can build that slide credibly. In a recent survey of marketers, the single most common complaint wasn’t budget or content — it was that nobody trusts the attribution. Here’s why. Users don’t click AI citations. Barely one percent do, across every platform anyone’s measured. So if you’re grading AI search the way you graded a search ad — by the click that lands on your site — the scoreboard reads zero, every time, forever. That’s not a measurement gap. That’s a measurement built for the wrong kind of channel, pointed at a new one. AI search doesn’t behave like a search ad. It behaves like a billboard. Nobody scanned a QR code off a billboard either, and nobody sane concluded the billboard wasn’t working. They measured it differently — awareness, recall, the slow creep of someone typing your name into a search bar three weeks later. AI search is asking marketers to relearn a skill performance marketing let them forget: measuring the half you can’t trace to a click, without pretending you can’t measure it at all. Take Daniel Goodwin of Provident 1031, a 1031 exchange specialist and Qualified Opportunity Zone advisor. He publishes with Kiplinger twice a month — not a guest post, a standing byline. He wrote a book that hit #1 Bestseller on Amazon across multiple categories the week it launched. His video Masterclass series became his best lead source, outperforming everything else by 5 to 1. And starting this August, our team is producing 10 new short-form videos a month for Instagram, YouTube, and LinkedIn — real footage, no shortcuts, one topic per clip, so nothing gets miscategorized by the models parsing it. None of that is built to win a click-through report. It’s built so that when someone asks an AI which advisor actually understands Oil and Gas tax strategy, there’s a mountain of original, citable material with his name already on it — and he’s willing to run that play for a full quarter before he even looks at whether to keep going. So stop asking one question. Ask three in order. First — Can the bots even find you? Pull your server logs and check for GPTBot, PerplexityBot, and ClaudeBot. If they’ve never touched your best page, nothing downstream matters. Second — When they do find you, how are you framed? Are you on the shortlist, or just named in a paragraph nobody remembers? Third — and this is the one every CEO can start tomorrow, for free — add AI platforms to your “How did you hear about us?” field. ChatGPT. Claude. Perplexity. It’s the cheapest instrument in your entire stack, and it’s the only one that catches the buyer who read about you and never clicked anything at all. The click isn’t coming back. That’s not a crisis — it’s just the truth about the kind of channel this is. The companies that win inside the answer won’t be the ones who found a way to trace it perfectly. They’ll be the ones who kept building the case with the proxies they had, the way Wanamaker did, while their competitors waited for proof that was never going to load. Thanks for listening. See you next time.
Why grading AI search by the click is the wrong scoreboard — and what to measure instead.
John Wanamaker famously said half his advertising was wasted — he just never knew which half. He spent it anyway. Built one of the great department store brands in American history on that bet.
A hundred years later, CEOs are staring at a new channel, watching it send almost no clicks, and concluding it doesn’t work. They’re about to make Wanamaker’s mistake in reverse: walk away from the half that’s actually working, simply because they can’t trace it.
You’re not measuring a channel. You’re measuring the wrong century’s idea of proof.
The channel nobody can prove is working
AI search stopped being optional a while ago. ChatGPT crossed the 1,000,000,000 (BILLION) monthly user mark this year. Google’s AI Mode did too. Roughly 7 in 10 software buyers now research with a chatbot before they ever talk to a salesperson.
So every CEO wants the same slide: What did AI search do for revenue? And almost nobody can build that slide credibly. Attribution — not budget, not content — is the single most common complaint marketers raise about AI search measurement.
The scoreboard reads zero, every time, forever — and that’s not a measurement gap. It’s a measurement built for the wrong kind of channel
Why the Click Was Never Coming
Users don’t click AI citations. Barely 1% do, across every platform anyone’s measured. So if you grade AI search the way you graded a search ad — by the click that lands on your site — the scoreboard reads zero, every time, forever.
AI doesn’t behave like a search ad. It behaves like a billboard. Nobody scanned a QR code off a billboard either, and nobody sane concluded the billboard wasn’t working. They measured it differently — awareness, recall, the slow creep of someone typing your name into a search bar three weeks later.
AI search is asking marketers to relearn a skill performance marketing let them forget: measuring the half you can’t trace to a click, without pretending you can’t measure it at all.
What BUILDING THE CASE ACTUALLY LOOKS LIKE
Take Daniel Goodwin, a 1031 exchange and Qualified Opportunity Zone advisor. He publishes with Kiplinger twice a month — not a guest post, a standing byline. He wrote a book that became an Amazon #1 Bestseller across multiple categories the week it launched. His video Masterclass series became his best lead source, outperforming everything else by 5 to 1.
And starting this August, our team is producing 10 new Reels and short-form videos a month for Instagram, YouTube, and LinkedIn — real footage, no shortcuts, one topic per clip, so nothing gets miscategorized by the models parsing it.
None of that is built to win a click-through report. It’s built so that when someone asks an AI which advisor actually understands Oil and Gas tax strategy, there’s a mountain of original, citable material with his name already on it — and he’s willing to run that play for a full quarter before he even looks at whether to keep going.
Daniel Goodwin is not chasing a click-through report. He’s building a mountain of proof and letting the answer engines find it.
THE QUESTIONS TO ASK INSTEAD OF ONE
Stop asking whether AI search sent you traffic. Ask three better questions, in order:
First — Can the bots even find you? Pull your server logs and check for GPTBot, PerplexityBot, and ClaudeBot. If they’ve never touched your best page, nothing downstream matters.
Second — When they do find you, how are you framed? Are you on the shortlist, or just named in a paragraph nobody remembers?
Third — The one every CEO can start tomorrow, for free. Add AI platforms to your “How did you hear about us?” field: ChatGPT, Claude, Perplexity. It’s the cheapest instrument in your entire stack, and it’s the only one that catches the buyer who read about you and never clicked anything at all.
The buyer who never clicks is still the buyer. Your dashboard just can’t see them.
The Bet Has NOT Changed
The click isn’t coming back. That’s not a crisis — it’s just the truth about the kind of channel this is. The companies that win inside the answer won’t be the ones who found a way to trace it perfectly. They’ll be the ones who kept building the case with the proxies they had, the way Wanamaker did, while their competitors waited for proof that was never going to load.
Framework and research referenced: Kevin Indig and Kyle Poyar, “How to measure the impact of AI search the right way,” Growth Unhinged, July 2026. See also Kevin Indig’s Growth Memo.
COMPLETE THE FORM TO
BOOK A STRATEGY CALL
"*" indicates required fields
CHAPTERS:
COMPLETE THE FORM TO
BOOK A STRATEGY CALL
"*" indicates required fields
Subscribe and Share – WE APPRECIATE YOUR SUPPORT
