If your content team is getting blamed for declining organic performance, ask one question before you accept the premise: what, exactly, is underperforming?
If the answer is “traffic from informational queries,” the content team didn’t break that. The funnel did — specifically, the stage of the funnel that was built on the assumption that a user with an informational question would travel from Google to your website to find the answer.
They don’t do that anymore. They ask AI Mode. They get an answer. They move on. Your content may have been cited in that answer. Your brand may have registered. But the click never fired, the session never opened, and the metric you built your reporting around recorded zero.
This is a funnel architecture problem, not a content quality problem. They require different diagnoses and different fixes.
What the funnel was built on
The content marketing funnel, in its classic form, worked like this. A user with a problem entered a question into a search engine. Your top-of-funnel content — the blog post, the guide, the how-to — appeared in the results. The user clicked through, consumed the content, encountered your brand, and entered a nurture sequence that eventually converted them. Simple, trackable, scalable.
That funnel had a first-order dependency that most practitioners never named explicitly: it required the user to click through to your site to get the information they were looking for. If the information was available in the search result itself — in a featured snippet, a People Also Ask box, an AI Overview — the click didn’t happen, the session didn’t open, and the funnel didn’t start.
Zero-click search has been eroding that first-order dependency for years. AI Mode accelerated it past the tipping point. The informational query — the lifeblood of top-of-funnel content — is now almost entirely captured before it reaches your website. The user has the answer. They don’t need the site.
Where the funnel still works
The funnel didn’t break everywhere. It broke at a specific stage for a specific query type.
High-intent queries — “hire a content strategist,” “content audit pricing,” “fractional managing editor services” — still drive clicks. A user who has identified a need and is evaluating solutions needs to see you, understand you, and trust you enough to take an action. That journey still runs through your website. The funnel there is largely intact.
Commercial and comparison queries behave similarly. A user choosing between options wants to see the options. AI answers for commercial queries tend to be less comprehensive and more navigation-focused — they point toward sites rather than substituting for them.
The mid-funnel content — case studies, specific methodologies, original research, point-of-view pieces that express a position only your organization can take — is also largely intact, because AI systems can’t substitute for it. They can summarize what’s generically true about content strategy. They can’t substitute for a specific case study about a real engagement with real results and a real person’s name attached to it.
What broke is the generic top-of-funnel. The “what is content strategy” post. The “how to do a content audit” post that covers the same ground as the forty other posts that cover the same ground. That content was never building brand affinity. It was capturing traffic on commodity queries and hoping some percentage converted. The AI took the commodity. What’s left is everything that was worth building in the first place.
The reframe that changes the strategy
If you stop thinking of the top-of-funnel as “content that drives traffic to start the conversion journey” and start thinking of it as “content that earns citations to build brand presence in the answer layer,” the strategic picture changes significantly.
The goal of a top-of-funnel blog post is no longer to rank and drive a click. It’s to be the source AI systems cite when your audience is forming opinions in your category. That’s not the same goal “ranking first” used to describe, even though the phrase hasn’t changed. That’s a different optimization target — less about volume, more about authority. Less about ranking position, more about being the specific source a machine trusts.
This is why brand voice and point of view matter more right now than they have at any point in the content marketing era. Generic content gets cited generically, if at all. Content with a specific, named, defensible position gets cited as the source that holds that position. The citation rewards distinctiveness. So does the conversion, eventually.
The fix isn’t more content
The instinct, when a content strategy underperforms, is to produce more. More posts, more frequency, more coverage of more topics. That instinct was wrong before AI search. It’s more wrong now.
The argument for stopping production before you fix the strategy applies directly here. If the funnel architecture is broken, feeding it more content produces more waste. The fix is upstream: redesign the funnel for the environment it’s operating in, clarify which query types your content is actually positioned to win, and build the content that serves those queries with the specificity and authority that earns citations rather than commodity traffic.
The content team didn’t break. Give them the right problem to solve.
This post is part of the Clifton Creative guide to GEO and AI search.
Jacob Clifton is the principal of Clifton Creative Agency. 25 years of professional writing, editing, and content strategy. He built Gawker’s Morning After and Tribune’s Screener to one million monthly readers in six months. Twice. He helps content teams solve the right problem — which is usually not the one they were assigned.
Before you rebuild the funnel, you need to know what you actually have. A content audit tells you which pieces are worth keeping, which clusters have real depth, and where the architecture needs to change. That’s the diagnosis that makes the strategy possible.
And once you know which clicks aren’t coming back, building the owned audience that doesn’t depend on them is the other half of the fix.
Because the content marketing funnel’s top-of-funnel stage was built on users clicking through from search results to find answers — and AI Mode now supplies those answers directly. High-quality informational content is still being used; it’s being cited in AI Overviews rather than clicked through to. The quality isn’t the problem. The funnel architecture is.
The top of the funnel — informational content designed to answer questions, explain concepts, and introduce users to your brand. High-intent bottom-of-funnel content (pricing, service comparisons, vendor evaluations) still drives clicks because users making decisions need to visit sites. Generic educational content no longer does.
High-intent queries where the user is evaluating options or ready to take action. Case studies, proprietary research, and original analysis that AI systems can’t synthesize from other sources. Point-of-view content that takes a specific, defensible position only your brand can hold. Anything with a named author, a real result, and a specific context that makes it irreplaceable.
The goal of top-of-funnel content shifts from “rank for this query and drive a click” to “be the source AI cites when users are forming opinions in this category.” That means prioritizing authority, specificity, and original perspective over volume and keyword coverage. It also means investing more heavily in mid-funnel content that can’t be substituted — case studies, original research, documented methodology.
Zero-click search happens when users get what they need from the search result itself — an AI Overview, a featured snippet, a knowledge panel — without visiting any website. It matters for content teams because it decouples content performance from traffic metrics. Content that earns citations in zero-click results is influencing buyers without generating sessions, which makes traditional ROI reporting unreliable and requires a different framework for measuring success.