Hi everyone, this is Neo.

The underlying logic of traffic is undergoing a profound shift. This week alone: Google’s Personal Intelligence deeply integrating personal data, the attack-and-defense game over OpenAI’s crawler strategy, and Google officially warning about domain trust. This string of signals shows that SEO has evolved from a battle of keywords to a fight over “data access” and “brand assets.”

For independent site sellers, this isn’t just about whether your traffic faucet survives — it directly affects your brand’s visibility in the AI era.

Today, let’s break down these three big changes so you can see which way the wind is blowing and build a sturdier traffic moat as the algorithms iterate.

01 Google AI Search Knows You Better: Personal Intelligence

Google is making a big move: connecting Gmail and Google Photos data to its AI Mode (i.e., Gemini), under a feature called Personal Intelligence.

What does this mean? Simply put, when users search on Google in the future, the AI won’t just know what’s on the internet — it’ll know users’ private information too. For example, if a user searches “what coat should I wear in Tokyo next week?”, Google used to give only a generic Tokyo weather suggestion. Now, if the user has this feature enabled, Google AI will directly read the flight booking in his Gmail (so it knows the exact dates), check his Google Photos (so it knows his style), and give an extremely precise recommendation.

Neo’s take

For independent site owners (especially in B2C fashion and travel), this is both an opportunity and a challenge.

• Opportunity: if your brand can appear in a user’s “personal context” (like confirmation emails or receipts), the odds of being recommended again go up.
• Challenge: users’ search queries will get shorter and vaguer. Before, they might search “Tokyo March men’s trench coat recommendation”; now they might just ask “what should I pack?”. This kind of “invisible” context makes traditional long-tail keyword SEO much harder to pin down.
• Privacy concerns: of course, Western users are sensitive about privacy. Google says the data won’t be used to train models, but handing your email and photos to AI still makes a lot of people raise an eyebrow.

02 The AI Crawler Offense-Defense Game: Block GPTBot or Let It In?

Hostinger recently published some hard data, analyzing 66 billion bot requests. It found a fascinating pattern: training AI crawlers are being blocked on a massive scale, while search AI crawlers are being let through.

• GPTBot (OpenAI’s training crawler): getting blocked at higher and higher rates, because nobody wants to hand their content to OpenAI for free to train models.
• OAI-SearchBot (OpenAI’s search crawler): coverage staying stable, because everyone wants to show up in ChatGPT Search and get traffic.

Neo’s take

A lot of independent site operators see an AI crawler and panic, going for a blanket “block everything” in robots.txt. That’s actually wrong!

We need to learn to divide and conquer:

1. Block GPTBot: don’t let it freeload off your content to train models — protect your original copyright.
2. Allow OAI-SearchBot: this is ChatGPT Search’s crawler. If users search for related products in ChatGPT, you want your site to be cited and bring in traffic.

As Aleyda Solís recommends:

“Disallow the ‘GPTbot’ user-agent but allow ‘OAI-SearchBot’.”

Action tip: go check your robots.txt right now — don’t slam the door on traffic (ChatGPT Search) that’s walking right up to you.

03 Free Subdomains: The Invisible Poison of SEO

Google’s big gun John Mueller recently came out with some “life advice” again. Replying to a site owner on Reddit, he said bluntly: free subdomain hosting makes SEO extremely difficult.

The backstory: a site owner had decent content and no technical errors, but could never rank in normal search results. The reason: he was using a free subdomain (like yourbrand.freewebsite.com).

Neo’s take

Mueller’s analogy is vivid: using a free subdomain is like moving into a low-rent apartment complex full of “problem tenants.”

Even if you’re a good person (a quality site), all your neighbors are spamming ads and running black-hat operations. When Google’s algorithm scans this “neighborhood,” it assumes nothing good lives there — and your site gets dragged down with them.

Advice for independent site newbies: Don’t save that few dozen dollars in domain fees by using Shopify’s default subdomain (.myshopify.com has decent authority, but for real operations you must bind a custom domain) or some other free platform’s subdomain. Your domain is your plot of land on the internet. Buy a proper .com — it’s the first step of building a brand and the most cost-effective SEO investment you’ll ever make.

Summary

This week’s SEO signals are crystal clear: access is advantage.

1. Data access: Google is starting to use users’ private data (email, photos) to deliver answers — SEOs need to figure out how to win in “invisible search.”
2. Crawler access: manage your robots.txt with precision — block training, allow search.
3. Domain access: don’t live in a dump. Buy a good domain and give Google a reason to trust you.

I’m Neo — follow me to grow independent site traffic with a technical mindset.