Search is changing so fast it feels like somebody swapped the internet while we were asleep.
A few facts from McKinsey really punched me in the face this week:
Preference Shift: AI-powered search is now the #1 most preferred source of information among consumers—beating Google, social platforms, brand sites, and review sites.
Category Variation: When Google AI Overviews generate an answer, the underlying sources vary dramatically by industry. Some categories lean heavily on affiliate blogs; others lean on brand sites; others rely on news or academic research.
Full Funnel: AI search is used across the full customer journey; not just discovery but also comparison and late-stage funnel and decision-making.
Traffic Impact: Meanwhile, traditional search traffic is already declining and is expected to fall off a cliff over the next 12–24 months (up to 50%).
None of this means Google is dead.
But it does mean the way people find information is already shifting, and the pace of the shift is just accelerating at an insane pace.
The Old SEO Playbook Had a Long Run
SEO used to be a pretty predictable playbook:
Keyword research
Content creation
On-page optimization
Technical SEO
Backlinks and off-page work
That system still matters, but buyer behavior no longer starts with typing a keyword into the Google search bar (or Bing, if you were a lunatic).
Increasingly, it starts with asking a question to an AI assistant, and that requires a different strategy and a different set of tactics.
A Meaningful Chunk of Your Organic Traffic Is at Risk
McKinsey is saying 20-50% of your traffic is at-risk. Honestly, that feels right. But who the heck really knows? Nobody.
But let’s just all agree it’s gonna be YUGE.
Here’s what we know:
AI search tools (ChatGPT, Claude, Perplexity, Google AI Overviews) are becoming the primary entry point (even boomers are on board)
User preference is already shifting away from traditional search
AI search often gives an answer without requiring a click
Put all that together and it’s reasonable to expect a crazy amount of organic traffic volatility over the next 12–24 months, especially for informational queries.
This isn’t doomsday, just the reality that distribution is shifting.
And distribution shifts always hit the slowest companies the hardest.
AI Search Is Already Part of the Entire Customer Journey
The McKinsey data makes it clear that AI search isn’t just “top of funnel.”
Consumers are using for all 3 phases of the decision-making journey:
Awareness
Discovery
Decision-making
That means AI isn’t just augmenting or replacing Google, it’s also messing with the core functions of:
Comparison sites
Review sites
Forums
The “ask a friend” moment
And consumer preference has already moved in this direction.
AI-powered search is now the most preferred digital information source, except for older demographics who still lean traditional.

Again, Google doesn’t disappear here.
It just becomes one of several entry points — and often not the first one.
Different Industries Show Up Very Differently in AI Search
This is where the McKinsey research gets interesting.
Similar(ish) to old SEO, AI engines aren’t treating every category the same.
Some examples from their July 2025 analysis of Google AI Overview answers:
CPG leaned heavily on affiliate blogs and user-generated content
E-commerce leaned overwhelmingly on brand and retailer sites
Financial services pulled from affiliate blogs, news and media, and market research
Telecom pulled heavily from review sites and news coverage

In other words: your industry’s “information ecosystem” determines what AI pulls from.
If your category is dominated by affiliate content, AI will reflect that. If your industry has stronger research or journalism, you’ll probably see more of that, too.
My Experiments: CEO Networks vs Life Insurance
I wanted to see specific results for very different industries, and how the source material varied for those industries. Basically, I wanted to fact-check McKinsey’s assumptions.
So, I ran two tests in Claude.
Test #1: “Which CEO network is best for me?”
First, I asked this question using only training data:
Sources included:
Traditional business media
Company marketing materials
Business books
Tech community discourse
Some general business chatter
Screenshot below:

Then I asked Claude to search the web, screenshot below:

Comparing the pre-search answers to the 2025 web results, the pattern was obvious: AI leaned heavily toward first-party content.
Nearly 40% of all cited sources came directly from founder content, company marketing, or member testimonials. In a niche category like CEO communities, that’s not completely surprising, since most groups publish a lot of their own material, and there isn’t much external reporting to counterbalance it.
What was mostly missing:
Independent reviews (G2/Trustpilot-style)
Comparative journalism
Analyst-style “pros vs cons” breakdowns
Reddit threads or candid forum posts
So the picture you get online is almost entirely produced by the communities themselves or their customers.
Test #2: “What life insurance should I buy?”
The results were completely different.
Sources included:
Most sources were “Best of 2025” rankings from major personal finance publications
Ranking methodologies were very similar across sites (AM Best ratings, J.D. Power scores, NAIC complaint index)
There was limited critical analysis
Much of the content appeared affiliate-driven
Some older information was still referenced
Screenshot below:

Almost everything came from “Best Life Insurance Companies of 2025” lists published by major personal finance sites, all using nearly identical criteria — financial strength ratings, customer satisfaction scores, and complaint indexes.
The same handful of carriers showed up in every source, and most of the content was built on the same underlying rating agencies (AM Best, J.D. Power, NAIC), rather than original analysis.
There was also very little critical perspective. Most of the pages were affiliate-driven comparison sites or lightly updated listicles, not deep reviews or investigative reporting. A few results even referenced companies that no longer exist, which tells you how often these lists get recycled.
The overall takeaway: this category has a highly uniform information ecosystem, so AI ends up surfacing the same structured data over and over.
Comparing the Two Side-by-Side
The life insurance answers couldn’t have been more different from the CEO networks.
See below:

The comparison matched the McKinsey research almost exactly:
CEO networks → more diversity of sources, more first-party materials, less investigative content
Life insurance → lower diversity, dominated by the same ranking sites, heavy reliance on third-party evaluators
The broader point: AI search mirrors the quality and variety of information available in your category.
If your industry has shallow content, AI will give shallow outputs.If your industry has rich, structured content, AI will use it.
So What Should Founders and CEOs Actually Do?
Here’s a practical checklist:
1. Strengthen Your First-Party Content
AI search frequently cites:
Product and feature pages
Category and comparison pages
Clear FAQs
Case studies
Testimonials
Well-structured blog posts
If you don’t publish clear, updated, authoritative content, you won’t be represented in AI answers.
2. Create “Answer-Ready” Pages
Build simple, structured, up-to-date pages for the main questions buyers ask.
Make them scannable, make them current, and make them easy for AI to parse.
3. Show Up Where AI Pulls From
Depending on your industry, that may mean:
Affiliate sites
Review platforms
News outlets
Research reports
Comparison sites
UGC and forums
Study your category’s ecosystem and then try your best to participate in it.
4. Monitor How You Appear in AI Search
Have someone on your team regularly check:
Which questions you show up for
Which sources are cited
How competitors appear
What information is missing
This needs to become a new part of your marketing rhythm.
5. Train Your Team on GEO
This includes:
Understanding how AI engines select sources
Using structured data
Keeping content fresh
Answering questions directly
Maintaining clarity and authority
It’s not SEO 2.0 — it’s a related but slightly different discipline.
6. Build a Cross-Functional Plan
Include marketing, product, data, sales, and customer support.
AI search affects the entire customer journey, not just “top of funnel.”
Summing It Up
AI search is becoming the new entry point for how people learn, compare, and decide.
It won’t replace every channel, and it won’t kill SEO, but it’s already changing the shape of demand.
The companies that win will be the ones who:
Understand how their category shows up
Build content AI likes
Participate in the right external ecosystems
And refresh that work regularly
Try to do that consistently, and you’ll stay visible in a landscape where the buyer's decision journey is changing even faster than we realize.

Or sign up for my weekly-ish newsletter, Signal // Noise, where I share my honest notes on work, startups, life, and money. No bullshit or spam!

Looking For More?
—> Here are 3 ways I can help you:




