- Answer Engine News
- Posts
- It's Official: Google Just Killed Traditional SEO
It's Official: Google Just Killed Traditional SEO
Here's what's next, and here's what you should do:

Thanks for reading. Feedback?
Reply to this email or message [email protected]
Subscribe to My YouTube Channel
Let’s get into it
At Google’s I/O Conference, They Killed Traditional SEO. Here’s What You Need to Do:
Google just dropped some major updates at I/O 2025 that change everything we thought we knew about SEO.
This isn’t just a feature tweak…it’s a full-on rewrite of how search works.
From AI Overviews replacing blue links to AI agents doing the searching for us, the old rules of SEO are done.
Here’s your breakdown of what’s new. And exactly what to do next.
1. Search Isn’t Static Anymore…It’s a Thread
Google has fully committed to making search a conversational experience, not a list of links.
Instead of typing in one query and getting 10 blue links, users are starting a thread. A real conversation. Each follow-up question builds on the last. And Gemini’s AI Overviews respond to each turn in real time.
This is called multi-turn search, and it means your website content has to anticipate not just the initial question, but the entire journey of intent.
What this means for marketers:
Build content clusters around primary topics with supporting detail pages
Optimize for follow-up queries, not just the head term
Use natural language that mirrors how people think and talk in conversations
Add FAQ sections that mimic conversational dialogue
Create next-step links in your content: “If you’re wondering X, here’s what to read next”
Think like a product designer building a user flow, not a blogger writing a standalone post. You’re not just answering one question. You’re joining a search dialogue.
2. Multimodal Search: Touch, Talk, Type, and See
Search is no longer a keyword box. It’s voice, camera, screen, and context—all in one.
At I/O, Google showcased how Gemini can take a voice prompt and a visual scan simultaneously to deliver layered responses.
What this means for you:
Add alt text and schema to images
Transcribe your videos
Use structured data to describe visual features (color, size, shape, etc.)
Optimize content for how people speak, not just how they type
We’re moving from search engine optimization to sensory optimization.
3. AI Overviews Are Becoming Default
AI Overviews aren’t experimental anymore. They’re going mainstream and showing up at the top of more and more SERPs.
Google now uses Overviews as the first layer of answers for many queries. And these Overviews aren’t just basic summaries. They synthesize answers across sources, cite real pages, and even recommend next questions.
Sometimes users will get what they need and never scroll. Other times, they’ll explore based on follow-ups the AI suggests. Either way, you want your content in that first layer.
How to show up in Overviews:
Write content that clearly and concisely answers questions
Use clear H2 and H3 headers that align with search intent
Structure your writing like a Wikipedia article meets a Stack Overflow answer
Be trustworthy—link to sources, show the author’s name, and update content regularly
Include well-structured summaries at the top of each article
AI Overviews are now the first screen of SEO visibility. Everything else comes after.
4. The Fragmented SERP Is the New Normal
Today’s search page is no longer a linear list of 10 links. It’s an explosion of modules, formats, and surfaces. This fragmentation is by design—Google is giving users multiple entry points into content depending on their intent.
You might see:
AI answers at the top
Shopping cards with images and reviews
People Also Ask questions
Short video previews from YouTube or TikTok
Reddit or Stack Overflow embeds
Long-form articles under the “Web” filter
Translated content modules
Lists, comparisons, or infographics pulled directly into the SERP
This fragmented structure rewards format diversity. If your only play is 1,000-word blog posts, you’re losing ground.
What to do about it:
Optimize for video (YouTube, shorts, reels)
Use question-and-answer formatting on your blog
Submit your products to Google Merchant Center
Participate in expert forums and discussions
Add long-form content with rich metadata
Repurpose single pieces of content into multiple formats that can win in different modules
To win the modern SERP, you have to think like a multimedia publisher, not just a writer.
5. Personal Context and Privacy-Aware Discovery
This is one of the most important long-term changes to search: personalization now goes beyond browser history. Google is pulling from your entire Google ecosystem.
With user permission, Gemini can access data from Gmail, Calendar, Docs, YouTube history, Maps, and previous queries to tailor your search experience.
If you’re searching for a hotel in LA, Google might:
Reference your upcoming flight in Gmail
Filter based on check-in times from your calendar
Recommend places you’ve starred in Maps
Exclude locations you’ve negatively reviewed before
Even your previous searches, restaurant preferences, delivery history, and YouTube viewing patterns play a role. Gemini turns all this into real-time context enrichment.
Key implications:
Keyword tracking tools will become unreliable because each user sees something different
Brand interaction across Google surfaces (emails, Maps, video watch history) now affects visibility
If users don’t trust you enough to allow access, you’re cut off from high-context personalization
Loyalty programs, subscriptions, and logged-in app experiences that sync with Google services are now SEO factors
What you can do:
Ensure brand interactions span across Google (Gmail, Maps, YouTube, Shopping)
Prioritize email marketing that integrates with Google services
Build trust with transparency and consent-first strategies
Use schema to help Google connect the dots across services
Encourage opt-ins by clearly explaining the value of personalization
The takeaway: In the age of Gemini, AI uses your brand’s relationship data to influence rankings. Build that relationship, or lose relevance.
6. Query Fan-Out and Deep Search
Search is no longer a one-and-done. It’s a burst of intent resolution.
With query fan-out, Google breaks a user’s single search into multiple subqueries to better understand and meet their needs.
You search: “best AI tools for business.”
Google fans this out into:
Best AI tools for small business
AI tools for marketing automation
Comparison: Jasper vs. ChatGPT
Pricing for enterprise AI tools
Reviews of AI tools by G2 and Capterra
This happens in milliseconds, and the final result is a synthesized response based on the best source for each micro-intent.
What that means for you:
You need content that covers broad topics and niche subtopics
Be the best answer to one facet of a big idea
Create layered content: overview pages + deep-dive pages
Interlink related content pieces to support content discovery by models
Think like a research assistant, not just a content writer. Give Gemini what it needs to build a smart, multi-angle answer.
7. Model Context Protocol (MCP): Feed the Model or Be Forgotten
MCP is a set of structured metadata guidelines that help large language models (like Gemini) understand, store, and retrieve contextual information about your content.
Think of it like Schema.org but specifically built for feeding LLM memory. It tells the model:
Who wrote the content
When it was last updated
What organization it represents
What product or service is being discussed
What the authoritative source is
Why it matters:
LLMs can hallucinate or forget. MCP helps them remember the truth from your site and resurface it later across Overviews, agents, and chat.
What to do:
Implement JSON-LD with organization, author, and FAQ schema
Include named entities (companies, tools, dates) with consistency
Embed structured context in footers, author bios, and product pages
Use canonical URLs and make metadata crawlable
MCP is how you create structured memory inside Gemini.
8. Shopping, Fashion, and Try-On: AI Meets Ecommerce
Gemini isn’t just searching. It’s shopping.
In I/O 2025, Google introduced powerful ecommerce experiences:
AR-based virtual try-on for glasses, makeup, and outfits
Context-aware shopping suggestions (based on weather, events, and your calendar)
Size and style recommendations based on past purchases and user similarity
Review summarization, sentiment tagging, and Q&A from real users
Here’s how to win:
Submit high-res images and AR assets
Use full product schema (brand, reviews, specs, size, price)
Add video try-ons or demos to product listings
Feed dynamic content to Google Merchant Center
Answer product FAQs in structured format
Collect review content that answers real buyer questions
Your product pages must now be visually rich, data-rich, and context-rich.
AI doesn’t just want titles and SKUs. It wants immersive, informed content it can package as advice.
You’re not selling to people. You’re selling to Gemini, who is advising people.
9. AI-Literate Content Wins
Gemini isn’t a browser. It’s a summarizer.
You’re writing for a system that needs to:
Parse your intent
Trust your expertise
Quote your facts
Make your content machine-eligible:
Clear subheads and bullets
Source citations
Short, fact-based paragraphs
Linked internal resources
10. First-Hand Experience Is the New E-E-A-T King
Google’s original E-A-T guideline emphasized Expertise, Authoritativeness, and Trust.
Now? The extra “E”—Experience—is front and center.
Gemini wants to know: Has this person actually done the thing they’re talking about?
This changes how you write content:
Add first-person voice: “Here’s what happened when we tested X”
Use original screenshots, photos, and quotes
Share opinions and lessons—not just facts
When Gemini builds summaries, it favors real insights from real people over regurgitated info.
If you want to rank, you need to prove lived experience.
11. AI Agents Are Redefining Discovery
We’re entering a world where users won’t just use search. Their AI agents will use search for them.
Here’s what this looks like:
Gemini gets a task like “Plan a birthday dinner”
It searches your calendar, Maps, YouTube reviews, and restaurant listings
It books the dinner, sends an invite, and even adds a reminder to your phone
Or…
You say, “Find me a CRM that works with Stripe and has AI lead scoring.”
Gemini does the research and returns a shortlist without showing you the search results
Your content is now competing in the background.
The user might never see your page. Gemini just extracts the key data.
To show up in this future:
Build machine-parsable content that answers agent tasks
Use structured data to label bookable, buyable, schedulable actions
Optimize for zero-click, zero-interface experiences
Include decision-enabling content: comparisons, recommendations, filtered lists
This is more than SEO. It’s Agent Engine Optimization.
You’re not just being discovered. You’re being delegated to.
Summary FAQ
Q: What’s the biggest shift from I/O 2025?
Search is now AI-driven, thread-based, multimodal, personalized, and assistant-powered.
Q: Is traditional keyword SEO dead?
Not dead..but no longer enough. You need AI-literate, structured, and experience-rich content.
Q: What is Model Context Protocol?
It’s a structured way to feed consistent, machine-readable metadata to LLMs like Gemini so they can accurately represent your content in AI summaries.
Q: How is ecommerce search changing?
It’s becoming interactive. Visual search, try-ons, and Q&A are the new norm.
Q: What should I do right now?
Audit your content. Ask: Can Gemini use this in an answer, an agent task, or a follow-up? If not, rewrite it.
If you need any help with this, just email me at [email protected] or reply to this email.
Answer Engine Optimization Strategies and Tips
These 4 Moments Will Transform How AI Sees Your Business
ChatGPT Has Changed Shopping Forever. Here’s How:
Why Ahrefs Content Director Thinks Top-Funnel Content is Dying
Google’s Vision for AI Agents
If you liked this newsletter, feel free to forward to your friends in marketing and AI.
If you got this from a friend, you can sign up to get the newsletter: https://aen.beehiiv.com/subscribe