top of page

AI didn't kill SEO, it just made experience the only thing that matters

  • Writer: Caius Tenche
    Caius Tenche
  • Sep 19, 2025
  • 5 min read

Updated: 3 hours ago

Is SEO dead in the age of AI?

Not quite, but it is changing fast.


Diagram showing how AI search connects directly to brand experience and conversions
AI search connects directly to brand experience and conversions.

According to Profound’s Josh Blyskal, ChatGPT referrals to websites fell by 52% in just one month. Meanwhile, sites like Reddit and Wikipedia are surging, and were cited nearly 90% and 60% more often, because they deliver quick, authoritative answers.


The same pattern is showing up in broader search. Pew Research found that when an AI summary appears in Google results, people click a traditional link only 8% of the time, compared with 15% when there’s no AI summary. In other words, the traffic that marketers have spent years engineering funnels around is being cut in half.


So what should brands actually do?


The Opportunity Hiding in Plain Sight

AI didn't kill SEO, it just raised the bar.


Insights from SEO and AI conversion experts highlight two critical shifts:


  1. Blog traffic is falling, but service pages are steady and they convert at up to 5x higher rates. Translation: the content that explains what you actually do is more valuable than ever.

  2. Conversions from AI-driven clicks are 2x higher than traditional search, even though the raw numbers are smaller. That means the people who do arrive have higher intent.


So yes, the SEO tactic is changing. But in many ways, it’s evolving in your favour if you know where to focus.


What to Focus On

Here are five ways to adapt and win in the AI-first search era:


1. Prioritize clarity and authority

AI pulls from content that answers questions directly. That means your site needs to be clear, complete, and trustworthy. But “clarity" doesn’t just mean simple copy, it means content that looks like it was written for an AI model to cite.


Next-level example: Instead of a generic “About Our Services” page, an HVAC company could publish a Q&A section that mirrors how people phrase questions in AI prompts (“What size air handling unit do I need for a 2,000 sq. ft. space?”). This goes right back to really understanding your customer and is a structure is far more likely to get surfaced in AI answers than a marketing-heavy paragraph.


AncestryDNA's FAQ page is an example of a service page with Q&A format designed for AI search.
AncestryDNA's FAQ page is a great example of a well-structured page with clear, direct questions and answers. Questions are separated into categories and the ability to view all the questions on a single page makes searching easy.

2. Double down on service and product pages

If blog traffic is softening, shift energy to the pages that convert. These pages should be treated as both conversion hubs and AI training data.


Next-level example: A boutique fitness studio could add short-form explainer videos and customer voice clips directly on the service page. Why? Because AI scrapes transcripts, FAQs, and captions, not just body copy. Rich, multi-format detail signals authority in ways competitors miss.


Capture of Airtable's website showing their solutions page for marketing teams. It leads with a prominent video showcasing how the product works. As you scroll, you see bold, pull-out quotes from well-known customers (e.g., "The team at Netflix uses Airtable..."). These quotes are strategically placed between feature descriptions, and a CTA button is consistently available, tying the visual proof and testimonials directly to the product's value.
Airtable's solutions page for marketing teams, leads with a prominent video showcasing how the product works. As you scroll, you see bold, pull-out quotes from well-known customers (e.g., "The team at Netflix uses Airtable..."). These quotes are strategically placed between feature descriptions, and a CTA button is consistently available, tying the visual proof and testimonials directly to the product's value.

3. Track the right metrics

Vanity traffic isn’t the goal anymore. AI-driven clicks are rarer but higher intent. The real win is optimizing what those visitors do next.


Next-level example: A SaaS startup could run GA4 event tracking specifically on AI referral domains (like chat.openai.com or perplexity.ai). By comparing that behaviour against Google Search visitors, they can see whether AI-driven leads convert faster, and adjust CTAs accordingly.


A custom-build AI Traffic report in Google Analytics GA4.
AI Traffic report in GA4. Source: grandcrudigital.com.au

4. Go after brand mentions, not just backlinks

AI doesn’t rank pages the way Google does. It pulls from trusted mentions in authoritative sources — see the reference to Wikipedia and Reddit above. This shifts PR, community, and partnership work from “nice to have” to “essential.”  [Update: As of July, 2025, Reddit citations are up +400% in ChatGPT according to Josh Blyskal.]


Next-level example: A real estate brokerage could join local Reddit or Facebook groups where first-time buyers vent frustrations. Posting useful, evidence-based responses gets their brand name into the discussions that AI already favours. It's an indirect but powerful tactic for visibility.


Example of a Reddit community forum with a brand name appearing in a conversation thread.
In a Reddit discussion titled, "What's the best standing desk for a new WFH setup that will get daily use?" users provide detailed and helpful recommendations for different brands based on their personal experience.

5. Think audience-first, not keyword-first

Traditional SEO often starts with a keyword list. But AI search mirrors human language and decision-making. That means your content playbook needs to be rooted in unmet questions and counter-narratives, not just keyword density. It's less about optimizing for bots and more about creating compelling, authoritative, and helpful content that directly addresses the complex questions and pain points of real users.


Next-level example: Instead of chasing to rank for “best mortgage broker Toronto,” a broker could write: “The 3 worst bidding-war strategies everyone tells you to use (and what works instead).” It’s not just search bait, it's "solution bait." It promises to solve a specific, frustrating problem for the reader and provides immediate value, which is exactly what AI is being trained to reward.


SOLUTION-BAIT HEADLINE

  • Why the Lowest Mortgage Rate Is Not Always the Best Deal

  • The Unwritten Rule of Car Buying: Ignore Your Test Drive

  • What Your Insurance Agent Doesn't Want You to Know About Your Policy

  • The 3 Most Overrated Tourist Traps in Toronto (and What to Do Instead)

  • The One Item You Should Never Buy at a Department Store

TRADITIONAL HEADLINE

  • Get the Best Mortgage Rates in Toronto Today

  • How to Choose the Right Car for Your Family

  • Save Money on Your Home Insurance: A Complete Guide

  • Top 10 Must-See Attractions in Toronto

  • Your Guide to Finding the Perfect Wedding Guest Dress

Comparison of counter-narrative blog headlines versus generic keyword-optimized headlines.


Why This Matters

AI may be disrupting search, but it isn’t disrupting what people want. They still want brands they can trust. They still want answers. They still want confidence that they’re choosing the right partner.


What’s shifting is how they get there. AI tools are cutting through the noise, skipping funnels, and pointing people straight to the brands that look most credible.


Ready to Future-Proof Your Marketing?

The businesses that win won’t be the ones clinging to yesterday’s SEO hacks. They’ll be the ones designing memorable, trustworthy experiences at every touchpoint, from the first AI-generated mention to the moment a customer clicks “buy.”


That’s where Gearbox comes in. We've been focusing on crafting brand experiences that are clear, memorable and consistent across every touchpoint since day one. Heck, it's in our name.


👉 If you're ready to make your marketing AI-search ready, let's talk.





bottom of page