🔔 Today I successfully claimed a Knowledge Panel on Google UK for my #SEO blog and I did a quick write-up covering: ✅ The Advantages of a Knowledge Panel ✅ The Math Behind Knowledge Panels ✅ Insights on Securing One seodepths.com/seo-research/g…

I never realized how much the overall SEO industry misunderstood schema until now. It was never a ranking factor. It just helped get the search features that Google added and took away. It was never used by LLMs - they get summaries of sites that they tokenize - stripping it all away in the process. I thought everybody knew this.

Go into Google Search Console right now. Filter your queries to INCLUDE your brand name. Note the clicks. Now filter to EXCLUDE your brand name. Note the clicks. Add them together. They don't equal your total clicks. Not even close. For most sites, you'll be missing 30-50% of your actual click data. This isn't a bug. Google designed it this way. Here's what's actually happening: ➡️ The chart at the top of GSC and the table below it are two completely different datasets ➡️ The chart includes clicks from "anonymised queries" the table doesn't ➡️ The moment you apply ANY query filter, those anonymised clicks vanish entirely ➡️ Google uses something called Bloom filters (a probabilistic data structure) that intentionally trades accuracy for speed ➡️ This is why filtering a specific page can sometimes show MORE clicks than the unfiltered view Ahrefs studied 22 billion clicks across 887,534 GSC properties. 46.77% of all clicks had no query data attached. Nearly half. And with AI Overviews pushing people toward longer, more conversational queries that number is almost certainly higher now. I wrote a full breakdown covering: • Why your data is missing (5 distinct reasons) • How Bloom filters create arithmetic that doesn't add up • 5 exercises you can run in your own GSC right now • What you can actually do about it Link in the comments. Try the brand filter test and drop your percentage gap below I bet most of you will be shocked. seo-stack.io/blog/why-is-dat…

Hour 2 will feature @andreasvoniatis, @Optimisey, @mintcstudios, Joe Hale, @MontseCano, @RanaAbuQuba, @RyanJonesSEO, @SukhSingh84, Annabelle Sacher, @damienrobertweb, Folashade Uba, Myriam Jessier, @RamonaJoita, @SimoneDePalma2, Tom Winter & @ValentinaStra1

To all OASIS fans around the world young middle aged old as fuck THANKYOU from the bottom of my heart you absolutely LICKED IT UP TO RAS forever grateful for your ENERGY and BIBLICAL vibes without you were just a good band with you were the BEST BAND ON THE FUCKING PLANET LG x

🆕 Founder of AI visibility tracking tool Lorelight is now shutting down the tool and said "there's no such thing as "GEO strategy" separate from brand building. For the large majority of brands at least." in LinkedIn post linkedin.com/posts/benjaminh… and also wrote a long blog post about it growwithless.com/shutting-do…

My favorite thing about the new AI search data studies coming out is how the takeaways always support obvious SEO best practices like “use the keywords in the URL” and “use bulleted lists and tables to clearly present information”

Hot take: Y'all are just rediscovering SEO by way of "AI SEO" or "GEO"... Everyone's acting like optimizing for LLMs is an entirely new discipline that we need to do ASAP. It's not. Let's clarify a few things: Two different mechanisms: 1. Getting mentioned by an LLM (no URL) = Being in training data 2. Getting a URL cited by an LLM = Ranking in search results For # 1 it's a popularity contest. The more you appear online in contextually relevant discussions, the higher your odds of being baked into the model's outputs. "Brand mentions are the new backlinks" thinking applies here. For # 2 👏 Every 👏 Single 👏 URL 👏 you see in an LLM output comes from a search engine API (Google or Bing) LLMs ARE NOT INFORMATION RETRIEVAL SYSTEMS (been trying to tell y'all this for years..) So, when you hear or say things like: "LLMs aren't citing Reddit URLs as much" → Reddit's search rankings dropped (or G's &num=100 removal impacted this) "LLMs fall for spam listicles" → These rank well in search (unfortunately) "LLMs prefer fresh content" → Search engines prioritize recency You're not discovering AI behavior patterns. You're rediscovering SEO! I'm tired & just trying to help 😅 ....it's also the last day to join my last AI Course of 2025 if you wanna level up & evolve your AI SEO thinking, hit me up. 🚀

Liz Reid from Google says that AI in Search is driving more queries and higher quality clicks 🤔 Here's the reality👇 * She says: the "dramatic" traffic decline reported is due to "third party reports" based on "flawed methodologies, isolated examples, or traffic changes that occurred prior to the roll out of AI features in Search." This is simply wrong. There are plenty of first-party GSC and Analytics data showing the decrease in organic search traffic directly correlated with the inclusion of AI Overviews. Heck! The Crocodile like graph has been even addressed directly by the Google Search Relations team showcasing GSC data 😂 * But here's the key, she also says: "While overall traffic to sites is relatively stable, the web is vast, and user trends are shifting traffic to different sites, resulting in decreased traffic to some sites and increased traffic to others. People are increasingly seeking out and clicking on sites with forums, videos, podcasts, and posts where they can hear authentic voices and first-hand perspectives." So if we read between lines, this might explain the high level claim that: "AI search hasn't hurt the traffic sent to sites" simply because the traffic sent by Google hasn't decreased at an *overall* basis, but to *which* sites this traffic is what has actually changed, *and* of course, the sites benefited are the big UGC platforms she talks about aka Reddit and YouTube (which is also Google), which have continued to grow despite AIO's and showcased more via other search features. * Additionally she mentions that: "As a search company, we care passionately — perhaps more than any other company — about the health of the web ecosystem. We continue to send billions of clicks to websites every day and believe that Search’s value exchange with the web remains strong." Sorry, but if most of the sites that have ultimately increased in clicks/traffic despite AI Overviews are mainly the biggest platforms on the Web -owned by big corp's- (that Google has partnerships with as well, like Reddit) and Google's own ones (like YouTube), rather than independent sites offering information, products and services, owned by small or medium sized businesses it *does* hurt the health of the Web ecosystem. It's sad how they try to manipulate our perception of what's happening using half-baked half-truths. Liz: Please, try to do better; we know better than that. PS: Why don't you give us the filters/segments for the data directly in GSC and GA? It's time to take the necessary steps for transparency here. Thanks! Link to the article: blog.google/products/search/…

Today we launched a new product called ChatGPT Agent. Agent represents a new level of capability for AI systems and can accomplish some remarkable, complex tasks for you using its own computer. It combines the spirit of Deep Research and Operator, but is more powerful than that may sound—it can think for a long time, use some tools, think some more, take some actions, think some more, etc. For example, we showed a demo in our launch of preparing for a friend’s wedding: buying an outfit, booking travel, choosing a gift, etc. We also showed an example of analyzing data and creating a presentation for work. Although the utility is significant, so are the potential risks. We have built a lot of safeguards and warnings into it, and broader mitigations than we’ve ever developed before from robust training to system safeguards to user controls, but we can’t anticipate everything. In the spirit of iterative deployment, we are going to warn users heavily and give users freedom to take actions carefully if they want to. I would explain this to my own family as cutting edge and experimental; a chance to try the future, but not something I’d yet use for high-stakes uses or with a lot of personal information until we have a chance to study and improve it in the wild. We don’t know exactly what the impacts are going to be, but bad actors may try to “trick” users’ AI agents into giving private information they shouldn’t and take actions they shouldn’t, in ways we can’t predict. We recommend giving agents the minimum access required to complete a task to reduce privacy and security risks. For example, I can give Agent access to my calendar to find a time that works for a group dinner. But I don’t need to give it any access if I’m just asking it to buy me some clothes. There is more risk in tasks like “Look at my emails that came in overnight and do whatever you need to do to address them, don’t ask any follow up questions”. This could lead to untrusted content from a malicious email tricking the model into leaking your data. We think it’s important to begin learning from contact with reality, and that people adopt these tools carefully and slowly as we better quantify and mitigate the potential risks involved. As with other new levels of capability, society, the technology, and the risk mitigation strategy will need to co-evolve.

#SEO #HotTake If you listen really carefully, you can hear the desperation of many of the big "SEO Tools" trying to justify their existence and subscription fees with AI correlation studies based on ever-changing nonsense. IYKYK

Oasis sounds amazing!!! it's fucking Noel Gallagher playing the Cigs and Alcohol riff after 16 years and Liam singing his heart out!!!! but please don't leak the whole setlist 🙏🙏 It's nice to hear a little taste, but let's wait a month, it won't be long now!!! We are so fucking back

🔔A faulty placeholder segment generated within the #React Router component can seriously impact SEO 👿It’s all down to client-side routing on SPAS and their major impact on indexation and rendering Thanks @sengineland for using your space !

Il Leeds United ha annunciato che il comune di Leeds ha dato il primo via libera alla ristrutturazione di Elland Road, che passerà da una capienza di 37.645 persone a oltre 53.000. E il risultato sembra spettacolare. La città di Leeds e i suoi tifosi meritano un tale palcoscenico (solo la waiting list degli abbonamenti ha toccato recentemente quota 26 mila).

NEW Research: sparktoro.com/blog/new-resea… If we include ChatGPT, what is Google's *REAL* market share of search? After years of speculation, we finally have enough reliable data to build high quality numbers for both.

Happy birthday, mr. Crouchée And regards to your family 🤣 @petercrouch @PeterCrouchPod

🔔 About the role of cache in your eCommerce In my latest post, I cover how caching strategies boost eCommerce performance and SEO. Key tips on hard disk, server, and JS caching inside! seodepths.com/seo-research/t…

🔥 Just released in Chrome 129 today, a new Performance panel landing page! 🔹 monitor your local Core Web Vitals in real time 🔹 compare them to real-user data from CrUX 🔹 and more Learn all about it in today's blog post: developer.chrome.com/blog/de…

🔔 The strange case of #WebP images stuck as Crawled, Not Indexed WebP images bring speed, but what if they start causing more harm than good? Dive into the full story here: seodepths.com/seo-research/w… #SEO #WebP

“This is it, this is happening” Tickets on sale this Saturday 31st August (🇮🇪8AM IST / 🇬🇧9AM BST) Dates: Cardiff Principality Stadium - 4th/5th July Manchester Heaton Park - 11th/12th/19th/20th July London Wembley Stadium - 25th/26th July & 2nd/3rd August Edinburgh Scottish Gas Murrayfield Stadium - 8th/9th August Dublin Croke Park - 16th/17th August