The monthly marketing report lands, and organic sessions are down for the third quarter in a row. Someone on the marketing team wants to publish twice as many articles to win the traffic back, and your CFO asks why you're still paying for SEO.
Both reactions assume traffic and demand are the same number. They used to move together. Now many B2B buyers get their answer from a Google AI Overview, ChatGPT, or Perplexity before they click, and the AI-referred visits you do get often land in Direct or Referral instead of search. The chart can fall while the number of buyers looking for you holds steady.
In Revenue Rewired, the framework I use at StringCan to find where B2B revenue leaks, answering that chart with more content is a Volume Trap. Cutting the channel because sessions fell is usually a Metrics Misfire. Before you do either, check whether the pipeline from search actually dropped.
A drop in organic traffic doesn't mean SEO stopped working. Buyers now get many answers from Google AI Overviews, AI Mode, and AI assistants like ChatGPT before they click, and AI-referred visits often land in Direct, Referral, or Organic Search in your analytics instead of a clean AI channel. Before you cut SEO, compare Search Console impressions with clicks, then check branded search, AI referral sessions, Direct traffic to deep pages, and leads and pipeline from search in your CRM. Don't assume traffic equals demand, and don't try to fix the drop by publishing more. Getting this wrong is a Volume Trap paired with a Metrics Misfire, and it can cost you a channel that's still influencing deals.
Google now answers many questions on the results page through AI Overviews and AI Mode, and more B2B buyers ask ChatGPT, Gemini, Claude or Perplexity what they used to type into Google. When Pew Research Center tracked the browsing of 900 U.S. adults, people clicked a traditional search result on 8% of visits when an AI summary appeared, compared with 15% when it didn't.
If your page is the source an AI answer draws from, a buyer can read your thinking, skip the visit and come back weeks later by typing your company name. Search Console won't show you that pattern either, because Google counts clicks from AI Overviews and AI Mode inside the regular Web totals.
This is the part most traffic reports miss. AI-driven visits don't arrive in one tidy bucket, so you won't find them by looking for one. In Google Analytics 4, they usually split four ways:
Put those together, and the picture changes. Organic sessions can fall while AI tools influence more of your pipeline than they did a year ago, with the evidence spread across three other channels and one that never gets recorded.
Compare the last 12 months with the 12 before that. Most of this takes an afternoon.
When impressions, branded search, and pipeline from search held while clicks fell, the drop is mostly a reporting problem. Cutting SEO here removes a channel that's still feeding deals, so fix the measurement and keep going.
When pipeline from search fell too, and you don't appear in AI answers for your core questions, you have a visibility problem. The fix is changing how your content is built so AI tools can find and cite it. Publishing twice as much in a format those tools skip is still a Volume Trap, and an expensive one. If you can't connect any pipeline to search at all, you're not ready to make the call in either direction.
At StringCan, this sits in the Adapt stage of our Revenue Flow methodology, where you improve the system with real data before adding activity.
Most often, buyers are getting answers before they click. Google AI Overviews and AI Mode answer many questions on the results page, and more buyers research in ChatGPT, Gemini, Claude or Perplexity. Some of the AI traffic you do get lands in Direct or Referral, so it doesn't look like search. Check impressions and pipeline from search before assuming demand fell.
Some of it does. GA4's AI Assistant channel catches visits it recognizes, and others show as Referral from chatgpt.com. Visits from the ChatGPT app or links opened without a referrer often land in Direct, and buyers who come back later show up as branded search. A custom AI channel group and a self-reported source field close most of that gap.
Not until you've checked the pipeline from search. If impressions, branded search, and pipeline hold, cutting SEO removes a channel that's still feeding deals. If the pipeline fails and you're missing from AI answers, change the content approach instead. If you can't tell which it is, a Growth System Session can map where revenue is leaking and what to fix first.
SEO helps your pages rank in search results. AEO, or answer engine optimization, structures content so search engines can pull a direct answer from it, like the answer in an AI Overview. GEO, or generative engine optimization, helps AI tools such as ChatGPT and Perplexity understand and cite your company. B2B companies need all three, because buyers now move between them.
Take the Revenue Leak Finder. If your result feels familiar, send it to me and I'll tell you where I'd inspect first.
Jay Feitlinger is the CEO of StringCan and the creator of Revenue Rewired, a framework that helps B2B leaders find where revenue leaks between marketing, sales, and operations.