SEO Weekly Roundup: The 40% Reality, August Volatility, and Why GEO Is No Longer Optional

New data shows only 40% of U.S. searches send a click to the open web. Plus: August ranking volatility hits again, Google recommends 304 codes for crawl budget, the Anthropic robots.txt mess, and why GEO is no longer optional.

SEO NewsGoogle UpdatesWeekly RoundupAI SearchGEOCrawl Budget
By Emil Mequita

Happy Monday. If you’re like me, you spent part of last week refreshing your ranking tracker wondering what the hell was going on. Spoiler: something is definitely going on. Let’s get into it.


1. The Number That Should Be Keeping You Up at Night: 40%

The Datos x SparkToro State of Search report dropped last week and if you haven’t read it yet, put everything else on hold. Here’s the headline: only 40% of Google searches in the U.S. result in a click to the open web. The EU/UK sits at 40.7%.

Sit with that for a second. Six out of ten searches on Google never leave Google.

It gets more interesting. Google-owned properties now capture 17.1% of all U.S. searches — YouTube, Maps, Images, Shopping, and everything else in the ecosystem. In the EU/UK that number climbed to 20.4%, likely driven by the French AI Overviews rollout in July.

I’ve been in this industry for over 16 years and I’ve watched Google slowly tighten its grip on traffic. But this data makes it viscerally clear: we are no longer purely in the business of winning clicks from Google searches. We’re in the business of being valuable enough that Google’s AI decides to cite us — or finding ways to build an audience before people even get to the search box.

The encouraging bit buried in the report: informational searches are actually up. In the U.S., informational queries jumped from 59.69% to 65.12% year over year. People are asking more questions. They’re just getting more answers directly in the SERP. Your content still matters — it just needs to feed AI answers, not just rank for clicks.

One more note worth flagging: AI Mode usage has plateaued. Maybe even dropped slightly in the U.S., sitting at just 0.13% of visits. The EU/UK is at 0.29%. Search isn’t dead. It’s being hollowed out from the middle. Big difference.

Source: Search Engine Journal


2. We’ve Got Volatility — Again. Still No Word from Google.

If you woke up Saturday to rankings that looked like a different website, you’re not alone. Barry Schwartz flagged significant Google Search ranking volatility starting August 1st, with movement continuing into today, August 3rd. The third-party tracking tools are spiking and the SEO community chatter is loud.

There’s been no official announcement from Google. Which is becoming the norm. The days of Google dutifully posting a blog update saying “hey, we rolled out an update, here’s what we know” feel increasingly distant. Now we’re watching Semrush volatility charts and WebmasterWorld threads to piece together what’s happening in real time.

The last confirmed update was the June 2026 spam update. Nothing official for July or August yet. So whatever’s happening right now could be a quiet algorithmic adjustment, a data refresh, or something else we won’t find out about for weeks.

My advice: don’t panic-change anything. Document what you’re seeing with dates and traffic data, wait for it to stabilize, then assess. Reactive SEO decisions made during an active volatility window almost never help — they just add noise to the signal you’re trying to read.

Source: Search Engine Roundtable


3. Google Quietly Penalizing AI Content — What Last Week’s Case Tells Us

You may have seen headlines about Google treating some AI-generated content as “thin content.” If you caught our roundup from last week, you know we went deep on the WindowsForum.com manual action — a forum where an AI chatbot had been auto-posting responses since 2023, racking up hundreds of thousands of entries, and Google apparently flagged those specifically.

The important thing I want to emphasize this week now that the story has had time to settle: Google didn’t just penalize AI content. They penalized AI content that added no genuine value to the experience. The forum had sections with far more obvious thin content problems — syndicated feeds, archived clutter — and those weren’t touched. The AI chatbot section was.

That’s not random. That’s specific.

Here’s how I’ve been framing this for clients: if your content can’t pass the “so what?” test from a real user’s perspective, it doesn’t matter whether a human or an AI wrote it. Google has always penalized hollow content. AI just makes it dramatically easier to produce hollow content at scale — which means the blast radius of a bad content strategy is now much larger and faster.

If you’re using AI in your content workflow (and at this point, who isn’t), the bar is whether the piece says something useful that a real person cares about. “It covers the topic” is no longer enough.

Sources: Search Engine Journal | Search Engine Roundtable


4. Google’s 304 Recommendation: Small Detail, Big Impact at Scale

Google updated its crawl budget documentation this week with something worth bookmarking if you’re managing a large site: a direct recommendation to implement 304 Not Modified HTTP response codes.

The logic is simple. If a page hasn’t changed since Googlebot last crawled it, return a 304 — a signal that tells Google to use its cached version and move on. No full page download, no wasted crawl capacity, no server load. Google explicitly added this to their documentation as a way to preserve crawl budget for pages that actually need attention.

Google also clarified something I find genuinely useful: all their crawlers — Googlebot, Googlebot-Image, and others — share a single crawl capacity limit per site. High image crawling activity can starve your main web crawler. Worth auditing if you run a site heavy on product images, video thumbnails, or CDN-hosted media.

For sites under 10,000 pages, this is probably a low priority. For enterprise sites, news publishers, or large e-commerce catalogs, this is good technical hygiene that belongs in your implementation checklist.

Source: Search Engine Journal


5. Google Search Console Platform Properties: Now Live Globally

Quick practical note: Google Search Console Platform Properties went globally live last week. You can now connect your brand’s social platform presence and track its visibility alongside your traditional web property data — all within the same GSC interface.

When this feature was first announced, I was skeptical of how useful it would be day-to-day. That skepticism has softened. The more we see AI Overviews pulling from YouTube, Reddit, and social content, the more relevant it becomes to have consolidated visibility data across those surfaces. This isn’t just a nice-to-have; it’s a data source that’s going to matter more as AI Mode’s content diet expands beyond traditional web pages.

If you haven’t connected your social properties yet, spend ten minutes setting this up. The data is free and there’s no downside to having it.

Source: Search Engine Land


6. The Anthropic/Claude Robots.txt Situation Is a Textbook Cautionary Tale

This one flew under the radar for some people, so let me surface it. Google indexed private Claude.ai conversations because Anthropic had not properly configured their robots.txt to block Googlebot from those pages.

The irony of a leading AI company accidentally exposing its AI conversations to a search engine’s crawler is almost too perfect. But the lesson isn’t about irony — it’s about how many sites are operating with robots.txt configurations that were set up once and never revisited.

Whether you’re a solo site operator or running infrastructure at scale, if you have pages that shouldn’t be indexed, you need to explicitly block them. Assuming Googlebot understands your intent doesn’t work. Never has.

Anthropic fixed it, but for a window, private conversations were appearing in search results. Data privacy and SEO hygiene failure rolled into one. Don’t let that happen to you.

Source: Search Engine Land


7. GEO Is Now a Real Strategy, Not a Buzzword

If you’re still debating whether “Generative Engine Optimization” is a real thing or just consultant jargon, this week’s coverage should settle it. Multiple pieces dropped on measuring brand visibility in Gemini, optimizing for AI-driven revenue targets, and the “ghost citation” problem — where AI search uses your content to answer queries but doesn’t attribute it back to your brand by name.

That last one deserves its own conversation. Your information shows up. Your expertise is being used. But the AI answers as if it simply knows the thing — your brand stays invisible even as your knowledge powers the response.

The implication is uncomfortable: content that used to create brand awareness by ranking is now increasingly creating value for AI systems that may not surface your name in return. The click was the credit. Without the click, the credit disappears too.

The SEOs who thrive in this next chapter are the ones building brand recognition through channels AI can’t easily absorb — direct audience relationships, email lists, YouTube presence, podcast listeners, owned communities. The click is becoming optional. The brand needs to be the moat.

Sources: Search Engine Land | Search Engine Journal


Quick Hits


The Week in Summary

The throughline across all of this is the same one that’s been building all year: search is becoming a passive consumer of your content, not necessarily the primary reward for producing it. The clicks are declining. The citations are rising. Brands that adapt their measurement and content strategy to this reality — building for AI visibility alongside human traffic — are going to be fine.

The ones waiting for the old model to come back are going to have a rough second half of the year.

Stay close to your data. Don’t chase the volatility. See you next Monday.


Your Friendly SEO is a weekly roundup written from the perspective of someone who has been in this industry for 16+ years. If something here was useful, share it with someone who needs a level-headed take on a loud week.