Insights
What Personal Intelligence Means for SEOs and Marketers
Personal Intelligence in AI Mode has been called an "apocalypse" and "the beginning of the end for many SEO strategies."
Now that everyone in the US can use Personal Intelligence through AI Mode, Gemini Chrome and the Gemini app, will website traffic take a bigger hit?
I won't try to answer that question fully because it has layers of complexity.
But if Google can assemble an answer using personal information and the open web, it's likely more traffic will shift away from sites that previously appeared in Search.
This would include e-commerce, travel, and even sites offering tech support.
I didn't need to guess there, because Google uses them as examples of how Personal Intelligence can be helpful.
How helpful is Personal Intelligence?
If you read Google's blog post, Personal Intelligence "helps you find exactly what you need without having to give all the context."
If you dig further (like I did), you'll find a PDF hidden in a corner of Google AI that lists some of the known technical limitations and issues.
2 samples from the PDF [edited]:
1. Overpersonalization based on your interests.
A known challenge is the tendency for the model to rely too heavily on a personalized inference where it’s not appropriate, AKA “tunnel vision.”
Example: You might be a big fan of coffee shops. When you ask it to “plan a trip to Australia”, it may inadvertently provide an itinerary focused on coffee shops.
2. Mistaking another person’s preferences for your own.
In testing we’ve also seen a challenge around the model conflating subjects — for instance, attributing a family member’s interests to you.
Example: Based on a receipt in your email, the model might think you enjoy listening to heavy metal and offer suggestions on concerts nearby, when you actually purchased the tickets as a birthday gift for your brother.
Personal Intelligence isn't for everyone
Sharing my email messages, personal photos, travel plans, and shopping receipts with Google isn't on my 'to-do' list. And that's not entirely related to privacy. It's because I prefer to eliminate as much bias as possible from the results I get when I search.
I'm also trying to avoid the "Answer Bubble."
That term was highlighted in March 2026 research comparing 11,000 user queries across platforms like vanilla ChatGPT and Google AI Overviews.
The results showed that information presented by AI is rarely objective and highly dependent on which AI search engine people happen to use.
In other words, the outputs vary based on user history, purchases, and behavior across different models.
Where to focus your SEO and content efforts
I've taken relevant sections from the "Answer Bubbles" research and used snippets from Mack Grenfell's comprehensive summary of the same to create the list below.
Specific content types and strategies to try:
1. Invest in "authoritative brevity" (but keep it long)
AI models have a significant length bias. Sources under 200 words are under-represented in synthesis by up to 18%, while sources over 800 words are systematically favored.
2. Adopt assertive, explanatory prose
AI gravitates toward content that sounds authoritative. Use causal language (words like "because," "therefore") and certainty markers ("always," "definitely") to increase the odds of your content being synthesized into the final answer.
3. Prioritize your Wikipedia presence
Wikipedia is one of the most-cited and over-represented sources across all AI systems.
It is the "load-bearing" signal for your brand; ensuring it is accurate, complete, and well-sourced is now a top-tier SEO priority.
4. Double down on "unfakeable" experience
To survive personalized answer substitution, create content that a personal assistant cannot easily replicate. This includes:
- First-hand testing and original reporting
- Proprietary data, tools, or calculators
- Narrow, niche expertise that goes deeper than generic guides
5. Shift from keywords to entities
Instead of trying to rank for a specific phrase, aim to "be the source worth visiting" when the AI layer cannot fully satisfy the user's need.
Focus on solving complex audience problems and providing recurring use-case depth.
6. Platform-specific optimization
a). For Search GPT: Focus on authoritative, data-rich, long-form editorial content.
b). For Google AIO: Maintain a presence on forums and social media (like Reddit and YouTube), as Google draws from a "messier," broader web—though you must still provide the authoritative depth to bridge the citation-synthesis gap.
It's clear Google (et al.) is taking increasing control of what users see and where to 'corral' them in their search for answers.
To remain relevant, SEOs and marketers need to consider that simply publishing something relevant to a user's query may not be enough. Your goal is be the strongest reference, not the easiest summary (I still believe E-E-A-T applies).
We must execute a strategy grounded in trust, authority, and original value that earns a place after an AI model has done its best to prevent a site click.
Recommended: Check the section headed "What this means for brand visibility" in Mack Grenfell's article.
Related: As Cyrus Shepard wrote last week:
"Google has moved beyond simply ranking “good content” to proactively rewarding what AI can’t replicate."
How do you optimize your brand stories for GEO?
A "universally effective" writing strategy appears to be the key.
I went down a rabbit hole and looked at what generative engines need in order to surface and recommend your brand content.
My latest article shows you the patterns in prompt optimization for GEO, and gives you the 5 elements of a successful GEO strategy.
Check out my article and please drop a comment if you find it helpful.
Quoteworthy
1. "Clueless imposters"
Preeti Gupta, a Growth Consultant, wrote an article on her blog titled, "I don’t like how the word ‘Guru’ is misused in the SEO industry."
Gupta is from India where the word Guru has deep meaning.
Her post on Bluesky attracted the attention of John Mueller, who commented:
2. Is it too late?
Thinking of getting into seo (digital marketing/copywriting background), but am wondering if seo people are about to be mass replaced by ai. Love to get some opinions on this and whether it’s worth getting into, or is it too late?
I saw this on Reddit (where else?). Unsurprisingly, the question attracted numerous comments.
What would you have said?
Reply with your best line and I'll feature some in my next newsletter.
Do you follow me on LinkedIn? I share regular tips and stories I don't have room for here. Come and join me.
