You asked for this guide because something in that carousel hit a nerve.
Maybe you’ve watched your organic traffic slowly bleed out while your rankings stayed the same. Maybe you’ve sensed that what you were taught about SEO doesn’t quite fit what’s happening on your screen. Maybe you’re a marketer who’s good at what you do and you’re wondering if “good” still means the same thing.
It doesn’t. Not entirely.
Google AI Mode crossed 1 billion users in 12 months. AI Overviews now dominate the top of search results. Organic click-through rates on informational queries have dropped 61% as a result. You can rank number one and get almost no traffic. That’s not a bug. It’s the new default.
Here’s what this guide is not: a panic piece about AI replacing marketers. Most of that content is useless. It’s either catastrophising or reassuring in equal measure, and neither does anything for you.
What this is: a clear breakdown of which marketing skills have actually gained value, which have lost it, and what you should be building right now if you want to be harder to compete with in 2026 than you were in 2024.
Google AI Mode reached 1 billion users within 12 months of launch. Queries are doubling every quarter. The search box now accepts text, images, videos, files, and Chrome tabs. Google described it as the biggest Search upgrade in 25 years.
Table of Contents
First, Understand What Actually Changed (And What Didn’t)
Before we get to skills, get the diagnosis right.
The search channel didn’t disappear. It changed its job. Google used to be a referral engine, sending people to websites to get their questions answered. Now, for a large and growing category of queries, Google is the answer. AI Mode synthesises a response from multiple sources and delivers it directly. The source gets a citation. The user stays on Google.
What’s gone: clicks from short informational queries. “What is content marketing?” doesn’t send traffic anymore. “Best tools for social media scheduling?” doesn’t either. These queries now resolve inside AI Mode before anyone reaches your site.
What’s still alive: commercial intent queries, comparisons, high-stakes decisions, and content where the reader needs to trust a source rather than just extract a fact. Queries like “best performance marketing agency for D2C brands in India” or “is YUP’s Advanced Digital Marketing course worth it” still generate clicks because an AI summary isn’t enough. The reader needs depth, credibility, and a real opinion before they act.
So the question isn’t “how do I save my old SEO strategy?” It’s “which marketing skills transfer to the new environment, and which ones do I need to build from scratch?”
Organic click-through rates on informational queries dropped 61% as Google AI Overviews dominate search results. 93% of searches in AI Mode end without a click to any external website. Marketers who built traffic strategies around short informational keywords are seeing the steepest declines.
The Marketing Skills That Have Lost Value (Be Honest With Yourself)
Some of this will sting. That’s fine.
Keyword-stuffed content production is essentially dead as a standalone skill. If you or your team can produce 800-word articles targeting a single keyword and not much else, that output has almost no SEO value left. AI Mode eats exactly that content. It synthesises it, cites maybe one source, and the rest of the articles in that category disappear from the click funnel entirely.
Traffic volume as a success metric is misleading more teams than it helps. If your monthly reporting still leads with “we got X organic sessions,” and you’re not also tracking branded search volume, citation frequency, and direct traffic trends, you’re measuring a channel that’s been structurally altered as if nothing happened.
Shallow generalist execution – managing every channel at a surface level without deep ownership of any one – is getting squeezed. Not because generalists have no place, but because the ceiling on generalist execution has dropped sharply. You can automate a lot of the generalist work now. What you can’t automate is strategic judgment on a specific channel you know deeply.
The Marketing Skills That Have Gained Value
1. Generative Engine Optimisation (GEO)
GEO is the practice of structuring content so AI systems cite it in their responses. Not just Google AI Mode. Also ChatGPT with browsing, Perplexity, and Bing Copilot.
Only 14% of marketers currently track AI visibility at all. That’s the gap. Most of your competitors are not doing this yet.
What GEO actually requires:
Answer-first writing. Every section of your content needs to open with a direct, standalone answer. AI systems scan for passage-level answers, not page-level ranking. If your section on “what is brand equity” opens with two sentences of context before defining anything, it’s less citable than a section that states the definition cleanly in the first sentence and explains it after.
Named entity signals. AI systems trust content that references specific, verifiable things. Organisation names, product names, named studies, specific tools, real data points. “Many marketers use analytics tools” is less citable than “marketers using GA4’s Explorations report to segment by channel source.” One is a general claim. The other is a specific piece of knowledge.
Structured data markup. FAQ schema, HowTo schema, Article schema. If you’re on WordPress, RankMath or Yoast handles this. But you need to configure it deliberately, not just install it and forget it.
Content modelled for extraction. Think about whether your content can answer a question as a standalone passage, even if the reader never sees the rest of the article. That’s the test. If the passage only makes sense with the context around it, it’s harder for an AI to cite cleanly.
GEO is a writing skill as much as a technical one. Marketers who understand this and apply it consistently are building a moat that most of their competitors haven’t started digging yet.
Generative Engine Optimisation (GEO) is the practice of structuring content so AI systems cite it inside generated answers. It requires entity signals, structured data, modular answer blocks, and content that reads as editorial rather than promotional. Only 14% of marketers currently track AI visibility at all.
2. Brand Authority Building (The Signal Layer)
Google’s Information Agents now continuously monitor the web and synthesise brand signals for buyers. Your buyers, increasingly, won’t visit five websites before making a decision. An agent will do that research for them and surface a summary.
What that agent looks for: what are people saying about this brand outside of the brand’s own website? Reviews, third-party mentions, press coverage, forum discussions, community reputation.
This means brand authority building, which used to feel like a soft, long-term play, is now a direct input to whether AI systems recommend your brand or your competitor’s. Your Google Business Profile, your reviews on G2 or Trustpilot, mentions in industry publications, coverage in newsletters your audience reads, your LinkedIn presence as a founder or practitioner – all of this feeds the signal layer that AI systems use to evaluate credibility.
The skill here is actively shaping what the internet says about you beyond your own properties. PR thinking applied to organic AI discovery.
Google’s Information Agents continuously monitor the web and synthesise brand signals for buyers. Third-party reviews, press mentions, and community reputation now carry more weight in AI-driven discovery than a brand’s own webpage copy. Brand signals, not just on-page SEO, determine how AI systems evaluate and recommend brands.
3. Commercial Intent Content Strategy
The content that still drives clicks and conversions is content that helps people make a decision that an AI summary can’t complete for them.
Comparisons where opinion and context matter. Specific product or service choices with non-obvious trade-offs. High-intent queries where the reader needs to trust someone, not just extract information.
If you’re a marketing team of five, and you’re still producing twenty blog posts a month targeting informational keywords, you should probably be producing eight posts that go deep on commercial intent topics instead. Shorter list, higher value per piece, better aligned to where clicks actually still happen.
The skill is being able to identify commercial intent topics in your category and write content that treats the reader’s decision with the depth it deserves. Not “here are 15 tools for X” with a two-sentence description of each. The actual comparative reasoning a smart peer would give you if you asked them in person.
4. AI-Native Marketing Operations
This is different from “using AI tools.” Most marketers are using AI tools. That’s table stakes.
AI-native marketing means rebuilding your workflow around AI, not adding AI to the edges of an existing workflow. The distinction matters.
A team that drafts content manually and uses AI to proofread at the end is using AI tools. A team that uses AI to research the competitive landscape, identify citation gaps, build structured content briefs, draft initial passes, run A/B tests on hooks, and monitor AI visibility across platforms – that team has rebuilt around AI. The output per person is genuinely different.
From what we’ve seen with YUP learners in the AI Marketing cohort, the biggest unlock isn’t knowing which tools to use. It’s knowing how to structure a workflow so AI handles the high-volume, repeatable parts and humans make the strategic calls. That takes real effort to build. It’s also the most durable skill because it compounds with every new tool that comes out.
5. Data Literacy Specific to Marketing Attribution
AI has made it easier to produce marketing outputs. It’s made it harder to attribute which outputs drove business results. The two are connected.
As the funnel gets noisier – more touchpoints, more AI-mediated discovery, more zero-click brand impressions – the marketers who can build clear attribution models and read the actual data confidently are the ones making better resource decisions.
This isn’t data science. You don’t need to write SQL to be data literate as a marketer. But you do need to be able to navigate GA4 without a tutorial, build a basic performance dashboard without relying on your analytics team, and understand the difference between last-click attribution and data-driven attribution well enough to have an opinion about which one is misleading your reporting.
Marketers who can do this earn more and get more autonomy. That’s still true regardless of what AI does to the channel landscape.
The One Framing That Ties It All Together
The carousel’s closing line was: “Write for the human. Structure it for the machine. That’s it.”
That’s not a simplification. It’s genuinely the mental model.
Writing for the human means your content has a real point of view, a clear reader benefit, and enough depth to be worth reading. It doesn’t read like it was produced by someone who was paid per word.
Structuring it for the machine means you’ve thought about how AI systems will read it. Answer-first sections. Clear entity signals. Schema markup. Modular passages that can stand alone as citations.
Most marketers do one or the other. The ones who do both are getting cited, trusted, and clicked on in a search environment where most other marketers are watching their traffic decline and blaming the algorithm.
The algorithm isn’t the problem. The strategy is.
What to Do With This
If you read this far and want a practical starting point:
This week: Pick one important piece of content on your site. Open the sections. Check whether each one leads with a direct, standalone answer. Rewrite the ones that don’t. That’s your first GEO experiment.
This month: Search for your brand and your key product topics on Perplexity and in ChatGPT with browsing. Note whether you appear, who does, and what they’re getting cited for. That tells you your current AI visibility baseline.
This quarter: Audit your content calendar. Identify which pieces are targeting pure informational keywords (already losing click value) and which are targeting commercial intent or decision-stage queries. Shift the balance.
None of this is complicated. It just requires deciding that the old playbook isn’t enough.
Conclusion
The marketers getting hurt right now built their strategy around traffic volume from generic informational content. That model worked when search was a referral engine. It doesn’t work the same way when search is the answer.
The marketers who are harder to compete with understand brand authority, commercial intent, and how AI systems decide what to cite. They’re not waiting for the landscape to stabilise. They’re building while most people are still debating whether the shift is real.
It’s real. And the gap between those who’ve adjusted and those who haven’t is going to keep widening.
Want to go deeper on AI marketing? Young Urban Project’s AI Marketing course is a 4-week hybrid programme covering GEO, AI-native marketing workflows, and the practical skills marketers need to stay ahead of what’s changing. If you’re serious about building these skills properly, this is where to start.