The quiet collapse of the click
Look at those headlines again and a pattern jumps out: AI overviews, answer engines, “Read more” links, AEO vs GEO, utility news content, why ChatGPT cites one page over another.
Everyone’s dancing around the same problem: the web you built your growth machine on is disappearing in slow motion.
Google, OpenAI, Anthropic, Meta – they’re all turning the open web into a training set and a preview layer. Users get answers, summaries, and recommendations without ever needing to click. The “visit site” is now the exception, not the default.
For CMOs, performance leaders, and media buyers, this isn’t an SEO story. It’s a P&L story. You’re still forecasting on a world where impressions lead to clicks lead to sessions. But the platforms are quietly moving to impressions → answers → actions that may never touch your site.
If you keep optimizing for clicks that will never come, you’ll mis-price your media, mis-read your attribution, and misallocate your team’s time.
The work now is simple to describe and hard to do: stop thinking in terms of “search traffic” and start thinking in terms of “answer presence” and “action capture.”
The new funnel: from search engine to answer engine
Three shifts matter more than the rest:
- AI overviews and answer boxes are eating the top and middle of the funnel. Users get synthesized guidance, not a list of blue links.
- Social is becoming query-driven. TikTok, Instagram, YouTube, and even WhatsApp are where people ask “what to buy” and “how to do X,” and creators or AI surfaces answer it inline.
- Chat interfaces (ChatGPT, Gemini, Claude, Copilot) are becoming the default interface for research and planning, especially in B2B and high-consideration consumer categories.
The old funnel assumed:
- Discovery happens on social or display
- Research happens on Google
- Decision happens on your site
The new funnel often looks like:
- Discovery happens on social or in chat (“what’s the best…?”)
- Research happens inside an answer layer (AI overview, creator explainer, chat thread)
- Decision happens on-platform (native checkout, lead form, app store, marketplace) – or with one click to a very specific page
In this world, “rankings,” “sessions,” and “time on site” are the wrong north stars. You care about:
- Presence in answers, not just positions in SERPs
- Attribution to actions, not just visits
- Surfaces where intent shows up (search, social, chat, marketplaces), not just Google.com
AEO, GEO, and the mirage of “owning” the query
The industry is already naming this: Answer Engine Optimization (AEO) vs Generative Engine Optimization (GEO). Fast Company calls GEO a mirage. They’re right in one sense: you will never “own” how a model answers a question the way you could own a SERP with a #1 ranking and a featured snippet.
But writing off AEO as hype is equally wrong. You can’t control the answer, but you can influence:
- Which sources the engines trust
- How your brand is described
- What actions are easiest to take next
The operators who win in this environment will treat answer engines the way the best teams treated early Facebook and early Google Ads: as systems with rules you can study, test, and exploit – not as mysterious black boxes.
What actually changes in your operating model
This isn’t about adding “AEO” to a slide. It’s about changing how you plan, brief, buy, and measure.
1. From “keyword research” to “question mapping”
The SEO headlines are still obsessed with keywords, title tags, cannibalization. Useful, but incomplete. The unit of competition is shifting from keyword to question.
Practical move:
- Export your top 200-500 revenue-driving queries from search, social, and internal site search.
- Normalize them into canonical questions: “how much does X cost,” “best X for Y,” “X vs Y,” “how to implement X,” etc.
- For each question, document:
- Where it shows up (Google, YouTube, TikTok, Reddit, Quora, chatbots)
- What the current top answers look like (publisher content, creator videos, AI summaries, your own assets)
- What action is easiest from that surface (buy, subscribe, contact sales, get directions, download)
This “question map” becomes more important than your keyword list. It’s the backbone for content, media, and product surfaces.
2. From “SEO content” to “answerable authority”
Ahrefs is asking why ChatGPT cites one page over another. Moz is rewriting 8,000 title tags. Copyhackers is warning about AI’s trust problem. The subtext: generic AI-written content doesn’t get cited, doesn’t get linked, and increasingly doesn’t get surfaced.
Answer engines need:
- Clear, structured, factual information
- Signals of expertise and trust (authorship, references, real-world data)
- Specificity: numbers, comparisons, examples, caveats
That means your content strategy needs to move from “publish more” to “be the best possible answer on a small set of commercially important questions.”
Operationally:
- Pick the 50-100 questions with the highest revenue impact from your question map.
- Assign each question to an internal subject-matter owner (not just a copywriter).
- Build a canonical “answer hub” per question with:
- Short, direct answer in plain language
- Deeper sections with data, examples, and comparisons
- Clear, structured markup (schema, FAQs, pricing tables)
- Explicit author, sources, and last updated date
Then use AI as a research and editing tool, not a writing engine. Let it:
- Summarize dense internal docs
- Suggest counter-arguments and missing questions
- Generate variants for different surfaces (short, long, video scripts)
But the core message – the thing you want answer engines to trust – has to be human-owned. In a SaaS recession, as Copyhackers points out, outsourcing your message to a model is a fast way to sound like everyone else and get priced like everyone else.
3. From “traffic targets” to “surface-level conversions”
You can’t judge success by “organic sessions” when answer engines are doing the reading for the user. You need to move one layer up in the funnel and one layer closer to the action.
Start tracking:
- On-platform actions: lead forms in Google Ads, native lead gen on LinkedIn and Meta, TikTok instant forms, WhatsApp replies, marketplace “add to cart.”
- Assisted actions: calls from search, map directions, app installs, store visits, DM conversations started.
- Answer presence proxies: share of voice in AI overviews, citations in major answer engines, inclusion in “Read more” clusters.
Then rework your media and SEO reporting to:
- Report cost per surface-level action (e.g., cost per qualified lead form on-platform), not just cost per click.
- Attribute revenue to source + surface (e.g., “Google AI overview → site → purchase”) where possible using blended modeling, not just last-click.
- Trend answer presence for your top 50-100 questions monthly, the way you used to trend rankings.
4. From “search-only” to “intent wherever it lives”
Adweek is flagging that consumers are taking a new purchase journey on social. Search Engine Land is talking about utility news content in AI search. Social Media Examiner is pushing AI deep research and AI creative. The message: intent is fragmenting.
Your search budget is probably over-concentrated in Google and under-represented in:
- YouTube search and suggested videos
- TikTok search and Spark Ads on creator content
- Reddit search and targeted placements in high-intent threads
- Marketplaces (Amazon, vertical platforms) where “search” is one step from purchase
- Chat-based discovery (sponsored recommendations, plugins, actions – where available)
The practical move is to build an intent portfolio:
- For each canonical question, list every surface where that question appears in the wild.
- Decide: are you going to own it (your content), borrow it (creators, partners, affiliates), or buy it (paid placements)?
- Set a target blended CAC or ROAS across surfaces, not just per channel.
Media buying in a world where the platform answers first
So what does this mean for your paid search and paid social teams, practically?
1. Paid search: from “keyword coverage” to “answer adjacency”
As AI overviews push organic results down, your ads are either:
- Above the answer (still premium)
- Between answer and organic (less visible, but still useful)
- Or increasingly, integrated into the answer experience itself (formats still evolving)
Your search strategy should:
- Segment campaigns by question type (informational, comparison, transactional) rather than just match type.
- Bid more aggressively on queries where:
- You’re weak or absent in organic/answer layers
- The AI overview is generic and doesn’t feature strong competitors
- On-platform conversion paths are well-instrumented (lead forms, calls, store visits)
- Pull search term reports and tag queries that consistently trigger AI overviews, then test ad copy that:
- Complements the likely overview (“Get pricing in 30 seconds,” “Talk to a specialist,” “Local options near you”)
- Offers a clear next step beyond the generic answer.
2. Paid social: buying signals, not just audiences
Adweek’s line – stop chasing data, start harnessing signals – is the right one. With Meta, TikTok, and YouTube, your job is less about micro-targeting and more about feeding the system:
- High-intent creative (clear who it’s for and what it solves)
- Clean conversion events (purchase, lead, qualified lead, high-value engagement)
- Enough volume for the algorithm to learn
In the answer engine era, social is where you:
- Seed the narratives and proof points you want answer engines to pick up
- Test which angles and objections actually move people
- Drive direct actions when search is increasingly “read-only”
Shift some of your “upper funnel video” budget into:
- Creator-led explainers that literally answer your top questions
- Shorts/Reels/TikToks that mirror how people ask and answer in search
- Retargeting that assumes the user has already seen an AI or creator answer and just needs a reason to act now
How to brief your team for the next 12 months
If you want this to turn into action instead of another trend deck, your brief for 2026-2027 should sound something like this:
- Objective: Increase revenue from high-intent queries by X%, regardless of whether the journey starts on Google, social, or chat.
- Strategy:
- Identify and dominate the 50-100 highest-value questions across surfaces.
- Shift content from volume to authority on those questions.
- Measure and optimize for surface-level actions, not just site traffic.
- Key initiatives:
- Build and maintain a cross-channel question map.
- Stand up an “answer authority” squad (SEO, content, product marketing, CX).
- Re-architect paid search around question types and answer adjacency.
- Reallocate X% of “search” budget to high-intent social and marketplace placements.
- Implement new reporting: answer presence, surface-level conversions, blended CAC by question cluster.
The operators who treat answer engines as just another feature of Google will keep chasing vanishing clicks. The ones who treat them as a new distribution layer – with its own rules, metrics, and economics – will quietly compound an advantage while everyone else argues about whether SEO is “dead.”