🤖

​AI Mode has started rolling out to all US users. It's been almost 4 weeks since Google's announcement at Google IO and many SEOs are reporting AI Mode appearing in the SERPs and Google Search Console data.

📈

​97% of URLs linked in AI Mode appear in the top 10 organic results. While there isn't a direct correlation, this data is very similar to AI Overviews, suggesting "ranking factors" are similar across AI Mode, AIOs and organic.

🎧

​Google launches Audio Overviews in Search Labs. As the name suggests, Audio Overviews are an audio summary of search results, similar to AI Overviews. US users can opt in to try this feature via Search Labs.

🔗

​Wikipedia, YouTube & Reddit are the most cited domains in AI search. The most-cited source varied by platform but Wikipedia & Reddit were consistently highly referenced across AIOs, ChatGPT and Perplexity.

🛑

​Google phases out support for 7 types of structured data. The discontinued types are: Book Actions, Course Info, Claim Review, Estimated Salary, Learning Video, Special Announcement and Vehicle Listing.

⛓️‍💥

​Google experienced two issues on 12 June, affecting indexing and Lens. One outage affected Google Lens, Discover, and Voice Search while the indexing was a broader issue. Both of the outages were fixed within 24 hours.

🔍

​Google CEO confirms plans to eventually integrate AI Mode into main search. As many of us have been theorising, it seems very likely that AI Mode will be the future of Google Search in the near future.

💳

​Google adds support for loyalty program structured data. Two types of loyalty program markup can be added: 'loyalty program' under Organization and 'loyalty benefits' in your Product structured data.

The Complete SEO Course

From Beginner To Professional

We are offering 1,000 free spaces on Mark's complete SEO course on Udemy!

Usually £199.99, if you use the link below you will gain free lifetime access to over 16 hours of on-demand video and over 80 downloadable resources.

The course covers key SEO concepts including indexing, internal linking, page experience, canonical tags and much more!

This is one of the highest rated SEO courses on Udemy with a 4.6/5 from over 1,700 reviews.

There are 1,000 redemptions for this code on a first come, first served basis. We'll send out more codes next week!

You should learn how LLMs actually work

At Zagreb SEO Summit last week, I got the question "What should SEOs be learning about to stay ahead?" It's an easy question for me, and the answer is "LLMs". ✅

LLMs are going to be at the core of the new search experience. Getting your head around how things like pre-training, tokenisation, system prompts, and similar concepts is how you ground your strategy and tactics going forward. It will insulate you from misinformation, help you sort fact from fiction, and allow you to focus on what works.

If you haven't seen it, my 100% recommendation is to watch this amazing video from Andrej Karpathy. It's 3.5 hours long, but it's the best primer I have ever seen on the topic. Invest in yourself and make the time. 🤩

Don't bother with vanity metrics for your links

You should not be judging the quality of your links with metrics like DA, TF, or DR ❌

These are third-party (and easily exploitable) metrics that Google doesn't care about.

​Tamara Novitović in her talk at Zagreb SEO Summit suggests using an IMPACT score based on:

🖱️ Traffic: Does it actually get clicks (this was the #1 metric I have previously suggested!)

📈 Engagement: Are users actually getting the value they are expecting?

🎯 Placement: Is it related to the user's current information forage?

💰 Conversion: The gold standard, can you get converting traffic directly from that link?

​

Good talks on link building are hard to find, I'd suggest giving Tamara Novitović a follow, and the chat Jack Chambers-Ward had with her on the Search with Candour podcast, about whether links are worth it in 2025!

Screaming Frog's semantic similarity scrutiniser

As of yesterday, you can analyse the semantic similarity of pages in a crawl to help detect duplicate, similar and potentially off-topic, less relevant content on a site with Screaming Frog v22.

Using their existing AI provider integrations via ‘Config > API Access > AI’ (including OpenAI, Gemini & Ollama) you can capture vector embeddings of pages.

Lots of great releases from the team. 👊

Pull AlsoAsked data into your AI assistants

"The Model Context Protocol (MCP) helps large language models "talk and work" with external systems (but in a standardized way)" - Myriam Jessier​

The hugely talented Metehan Yeşilyurt (follow him) has built something that even a human would describe as "game changing": An AlsoAsked MCP server that brings real-time PAA data directly into AI assistants like Claude.

A fantastic community effort; and his post is a great place to start getting your head around MCP.​

There is no such thing as a "national" search in Google.

Since we added the City option to AlsoAsked I've had people email saying they don't want to specify a city and they want to perform the research on a "national level", here is what I tell them:

​

💡Google is always passing a longitude and latitude with any standard search. We need to (and always have) passed location information.

💡Whether a search is "localised" or "national" depends on the intent of the query, which Google determines.

💡Knowing whether a query has local intent is important. It means you need to localise your content!

💡For example, searching the US for "food trucks" (thanks for the example Arnout) triggers a bunch of local questions; you need to know this!

💡Searching for something like "best laptops" does not. All the results are kind of the same because the exact location is not important.

💡Previously, AlsoAsked would just silently pass the largest city location in the background when you specified a region (which is now the default option).

So if you want "AlsoAsked classic" experience, just leave it as the default!

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)