📑

​AI Overviews have been found linking to Google SERPs rather than directly to website. This is, understandably, very controversial in the eyes of publishers and feels similar to the recently-discontinued Page Annotations feature.

🙅

​If users don't use your pages in search, they may fall out of the index. This answer to a user's questions seems to reaffirm Google's Navboost & user signals as ranking/indexing factors.

📊

​Large domains such as Quora, HMRC and Amazon have been impacted by the March core update. Some sites are also seeing some recovery from their December 2024 update losses.

📝

​Google appears to be testing new elements of its upcoming AI mode. Over 40 new search options with names such as 'Info Sleuth' and 'Smart Kitchen' were found and but didn't change the SERP when clicked.

📅

​Bing Webmaster Tools has added date range comparisons. This allows users to compare year-on-year for 7 days, 30 days & 3 months as well as comparison to the previous period for 6 months.

📊

​Copilot is now integrated directly into Bing Webmaster Tools. This new feature allows users to ask questions about their data to gain insights and learn how to use BWT directly on the platform.

🧑‍⚖️

​The European Commission finds Google in violation of two DMA regulations. These violations of the Digital Markets Act are for Google Search and Google Play and their preferential treatment of other Google services.

✅

​A recent Google Business Profile reverification bug has been fixed. If your GBP had been affected by this bug, you should have already been reverified automatically.

🗙

​Google changed the HTML code for desktop search results, creating errors in some SEO tools. Errors or significant ranking changes around 19 March may be caused by this change from Google's end.

Join the green web revolution

The web won’t get greener on its own! 🌱

Join us at our next GreenSEO event in Brighton on Wednesday, April 9th and hear from Nat Arney, Chris Green, Fiona MacNeill, and Kathryn Lawrence as they share insights on making the web more green - with sessions on content, rank tracking, and behavioural change.

Be part of the solution and book your free ticket now! 💚

Check if your site allows Googlebot's new IP range

If your website setup is authenticating Googlebot, they are now refreshing the JSON objects with those IP ranges daily, rather than weekly, so you should be too. ⏳
​
"Have you seen a recent drop in new users, and/or found that Google's crawl rate has dropped on your site while server response times seem to be higher than usual?" this is what Gianna Brachetti-Truskawa wrote a couple of weeks ago (here's Gianna's original post) when they pointed out Google had updated their IP ranges on 4/2/2025 and some larger CDNs were running into issues.

How much does your content contribute to traffic?

When planning for migrations, one thing I like to do with informational content is a quick cumulative traffic graph to demonstrate how much of the existing content is currently contributing to traffic. ⤵️
​
📈 In this case we can see that around, 1-2 pieces of traffic contribute around 50% of the traffic and the first 1/4 of the content provides ~90% of the site visitors.
​
💸 While you also need to look for outliers in GA in terms of revenue/leads, if these fit closely the mean, it's a great way to get scope on risk and reward.

"Just make good content"...right?

Sorry, "Helpful Content™" is not enough.

It does not mean you will rank, even if your site is also technically perfect.

I will demonstrate this (and how to bypass a bunch of Google filters) next month at brightonSEO.

LLMs don't use schema in the ways you may think

I think there has been some confusion over "Bing's Fabrice Canel confirms Copilot and other LLMs at Microsoft use schema for its LLMs" ⤵️
​
🗨️ Here is the quote: "Gen AIs value fresh content in particular, partly as a reference check of their LLM training data. Use the API at indexnow to push that information as it’s published or updated,"
​
🤖 This means they are using schema as part of their grounding process to work out if information in the model is correct.
​
😵 This does not mean "LLMs ingest schema during training", it does not mean that LLMs use structured data in the way we expect machines to - to explicity label things and their relationships.
​
🔄 Any RAG process using a search engine sorted index is likely to have some second-order impact of structured data (if it's helped the search engine, which is in turn providing the URLs).
​
🔢 LLMs are not building models of "entities" in their base models, as Britney Muller so eloquently phrased it: "LLMs are essentially giant statistical knowledge graphs of language."

Easily digest SEO patents as a podcast

Being aware of patents involved in SEO is a great way to "ground" your knowledge and theories. While not all patents may be used as-written in search, it gives fascinating insight into how challenges are being solved.
​
🎙️ I run the SEO Patent Podcast, which provides NotebookLM overviews of SEO patents that are easily digestible. This week it's covering the "Using concepts as contexts for query term substitutions" patent which details a system for enhancing search queries by using concepts, which are groups of words with a unified meaning, as contextual clues for suggesting better search terms.

Top Core Updates referrer leaderboard

A big thank you to our top referrers, who have signed up over 10 people to the Core Updates newsletter, go follow them!

🥇 MJ Cachón Yáñez (LinkedIn / Bluesky)

🥈 Nikki Pilkington (LinkedIn / Bluesky)

🥉 Lidia Infante (LinkedIn / Bluesky)

🏅 Saijo George (LinkedIn / Bluesky)