Google & SEO

Adobe move to acquire Semrush

Adobe is acquiring Semrush for roughly £1.6 Billion, a move that signals where SEO and digital marketing tools are headed as AI-driven search reshapes discovery and visibility. 

The acquisition aims to combine Semrush’s search-intelligence capabilities with Adobe’s content and marketing stack which will help brands handle both traditional SEO and newer GEO channels.

As AI-powered search gains traction, the deal underscores a wider consolidation trend in the industry as marketers want unified platforms that track visibility across both classic search engines and AI-driven answer engines.

Google Search Console adds branded queries filter

Google has announced that they’re rolling out a new filter in Search Console which will allow us to filter branded queries in Performance reports. Allowing users to check the brand/generic ratio of search visibility and clicks without having to use regex.

This feature will give marketers clearer visibility into how much traffic is driven by brand, and also a sneaky insight into what Google believes is your branded terms (as you cannot specify them within the filter).

It is not yet rolled out into every property, and it cannot provide retrospective filtering, so we can only see this from the date it is available in the property.

Google’s AI Shopping Updates

Just ahead of peak shopping season, Google has released a number of handy AI updates to help you find the perfect gifts at the best prices online, although some features might only be available in the US at this point. Here’s a rundown of what’s new:

• Shop in AI mode: This latest upgrade allows you to search in a conversational way, like you’d say it to a friend, and Google will provide the goods. Think visuals, prices, reviews and even whether the item’s in stock, without the need to filter your results or try to come up with the correct variant of keyword to find exactly what you’re looking for. Results types are also tailored to the search, so it could be shopping images if you’ve asked for inspiration, or comparison tables if you are weighing up different products.
• Shop in the Gemini App: Go straight from brainstorming gifts to browsing with help from the Gemini App, with comparison tables to help decide between different products and real shopping results right within your chat, instead of just the usual text-based answers.
• Agentic AI’s call feature: If you’ve found the perfect gift but need to find it in-store, you can ask Google to call around and find it for you. This AI feature will even find out if there are any promos or discounts! When you search for certain products with a ‘near me’ search, you’ll see an option to ‘Let Google Call’. AI will do the hard work, saving you time during the busy festive season, and Google will send you an email or text with the answer.
• Price tracking with agentic AI: If the gift at the top of your child’s Christmas list is looking a bit pricey, use Google’s new price tracking technology to let you know when the price falls within your budget. Google will even buy the product on your behalf (with your permission) using Google Pay!

ChatGPT increases brand mentions as Google reduces their focus

Recent Semrush data is showing that over the last three months, ChatGPT is starting to reference more brands in their results, whilst Google’s AI Mode has narrowed its focus.

Results show that ChatGPT is experimenting with its answers, as 12% more brands were mentioned in September before declining in October.

Whereas Google’s AI Mode mentions fell by 4% during the same period. As well as this, the Semrush data shows that ChatGPT had an increase in the number of sources/sites used with figures showing an 80% increase in October, compared to just a 13% increase from Google’s AI Mode.

Whilst there is some consistency between the results of ChatGPT and Google’s AI Mode when it comes to referencing brands. It’s still important to keep testing to ensure you get the best results for each model.

Google introduces query groups in Search Console Insights

Google has announced Query Groups, a new AI-powered feature in Search Console Insights that clusters similar search queries to help marketers understand search performance more easily.

When reviewing search data, you often see dozens of slightly different versions of the same query, which can make it hard to spot trends and plan content. Query Groups solves this by automatically grouping related queries into clusters that represent main search interests, driving traffic to your site.

What it does

• Groups similar queries – see main search topics instead of long lists of variations
• Shows performance – track total clicks per group and spot rising or falling trends
• Drill down – click a group to view all individual queries in detail

Why it matters

Query Groups helps marketers identify emerging topics, optimise content strategy, and focus on high-level insights instead of sifting through hundreds of similar queries. Available over the coming weeks for high-volume accounts, this feature will make search analysis smarter and faster.

OpenAI launches ChatGPT Atlas browser

OpenAI has launched ChatGPT Atlas for macOS, a browser built around ChatGPT that has integrated search and assistant features.

A ChatGPT sidebar appears in any browser window to summarise content, compare products, or analyse data from the page you’re viewing, allowing you to find the additional information you need without leaving the current page.

One of the most significant features of ChatGPT Atlas is Agent Mode, which can open tabs and click through websites to complete tasks on the user’s behalf. According to OpenAI, it can research products, book appointments, or organise tasks inside your browser.

Why does it matter? Atlas blurs the line between browser and search engine by putting ChatGPT responses alongside traditional search results in the same view, allowing users to browse and ask questions simultaneously. It’s another major platform where AI-generated answers appear before organic links.

Versions for Windows, iOS, and Android are scheduled to launch in the coming months, but release dates have not been specified.

Find out more

Google expands AI Mode in Search to more languages and locations

Google has rolled out AI Mode to over 40 new regions and languages on 7th October 2025, making AI-assisted search answers more widely accessible globally.

Key updates:

• Broader availability: Users worldwide can now access AI Mode in their native languages.
• Enhanced answers: AI-powered search provides more comprehensive, refined responses to queries.
• Improved user experience: Tailored answers aim to surface the most relevant information efficiently.

Why this matters:
As AI becomes a bigger part of search, businesses need to ensure their content is structured and presented in a way that AI can easily interpret. Optimising for clear, authoritative, and well-organised content helps increase visibility in AI-powered results, while maintaining strong traditional SEO fundamentals ensures content remains discoverable in standard search listings.

What to expect:

• Broader reach: More users globally will interact with AI-generated search responses, increasing competition for clicks.
• Content clarity matters: Clear, structured, and authoritative content is more likely to be selected by AI as a source for answers.
• SEO evolution: Businesses may need to balance traditional SEO practices with AI-focused content strategies, ensuring content is both discoverable and easily understood by AI systems.
• Continuous monitoring: AI search results will evolve quickly, so keeping track of performance metrics and updating content will be critical for sustained visibility.

Microsoft outlines how to optimise content for AI search answers

Microsoft has just shared some new guidance for marketers on how to optimise content for inclusion in AI-powered search results.

Key tips:

• Prioritise content quality: Ensure your content is fresh, authoritative, structured, and semantically clear. AI systems favour content that is easy to understand.
• Enhance crawlability: Maintain traditional SEO practices like proper metadata, internal linking, and backlinks to ensure your content is discoverable.
• Structure for AI parsing: Organise content into clear, modular pieces that AI assistants can easily evaluate for relevance and authority.
• Focus on intent-based queries: Optimise for natural language and intent-focused search terms to align with AI’s understanding of user intent.

Why does this matter?

With Google disabling &num=100 parameters, visibility tracking just got trickier – but that only makes this guide more timely. With AI searches steadily on the rise, marketers need to focus on what really counts: clear, structured, high-quality content that AI can easily surface. It’s a reminder that while tools and interfaces change, content fundamentals still drive discoverability.

Google Disables &num=100

Google has disabled the &num=100 parameter in search URLs which many SEO tools use to pull ranking data. This has resulted in a significant change to Google Search Console data with impressions declining dramatically. 

Previously, when a full 100-result page was requested, Google would log every result on that page as an impression in Search Console, including the low-ranked listings, even if no real user ever saw them. This inflated impression counts by counting bot traffic as if it were a user seeing the listing on the SERP.

Why this matters

This change has raised questions about the idea of ‘The Great Decoupling’ where impressions were increasing but clicks weren’t following. It’s now looking like a lot of these impressions may have come from bots or SEO tools scraping search results as opposed to organic traffic. With the parameter gone, impression data should now reflect more accurately what users see in the search results. 

What to expect

• Impression baselines from before September 2025 might not be reliable
• Impression drop is more pronounced on desktop where scraping was most common
• Page rankings outside the top 10 are now less likely to show meaningful visibility

Next steps

• Review Google Search Console data from early September to account for the shift
• Reevaluate CTR and position metrics if reporting includes YoY comparisons
• Educate clients that recent impression drops are likely technical and not performance related

Google AI Mode set to become default search ‘soon’ 

Google leaders Sundar Pichai and Liz Reid have recently confirmed that AI Mode will soon replace the current search experience as the default. It’s already live in 180 countries, with ads due to roll in before Q4.

What to expect:

1. A combination of AI answers and traditional results, with the ability to ask further questions related to the topic within the SERPs.
2. Ads appearing as text and shopping placements with expansive product details.
3. Source pages shown before AI answers (giving more credit to the experts).
4. Websites and ads surfaced within AI responses to drive traffic to more information on the websites and sources.
5. An option to switch back to classic results for those who prefer it.

Why it matters:
This change will force brands to think less about keywords individually and more about becoming trusted voices in their subject matter. In a nutshell, if you’re not seen as a reliable source, AI won’t reward your content. Search is shifting from discovery to instant delivery, where visibility and value primarily lives inside the AI-generated answer itself. Google is no longer about “10 blue links”, its a conversation with Google, raising big questions about how smaller sites adapt. Likely, it will push businesses toward deeper, expert-led content to secure citations and remain visible in their sectors.

Google avoids breakup but must end exclusive search default deals

A U.S. judge has ruled that Google will not be broken up or forced to divest its Chrome browser or Android operating system, rejecting the Justice Department’s demand that Google sell Chrome. However, the tech giant will be barred from having exclusive default search deals and must share search data with their rivals.

These orders are temporary, lasting six years with potential oversight by a technical committee, and Google has the right to appeal the ruling.

Financial markets have reacted positively to the ruling, with shares in Alphabet, Google’s parent company, surging by more than 7%. However, the decision has also drawn criticism from some, including the founder of Google competitor DuckDuckGo, who argues that the order is not sufficient to curb Google’s dominant position.

Read more here

Google releases August 2025 Spam Update

Yesterday, Google announced and began rolling out its 2025 spam update at approximately 9:00 am PDT. This is the first spam-related algorithm update in about eight months. The rollout has been described as “normal”, which is applied globally across all languages and locations and is expected to take a few weeks to roll out.

Why this matters

This spam update follows Google’s June 2025 core update, marking Google’s continued efforts to refine search integrity. Historically, spam updates have brought volatility. Past updates in December 2024 and June 2024 had a significant impact, with December being particularly disruptive.

What to monitor

Search Engine Journal has reported that “Google hasn’t yet specified any targeted behaviours for this update, so it’s best to stay aligned with established quality and spam policies”.

“Expect ranking changes and traffic volatility while the update rolls out. Avoid reactive changes during the rollout window”.

What to watch

• Search Console trends: Track Impressions, clicks, average positions, and note any shifts from the start date of the update (August 26th).
• Pattern detection: Look for drops across clusters (e.g page types or query segments), not just individual URLS.
• Competitor movement: Identify if rank shifts are widespread across your industry or niche.

Recommended actions

As always, the likes of SEL & SEJ have advised to “Avoid Knee Jerk Reactions” within the rollout window, as rank changes are expected and can often be temporary.

AI Mode is officially rolling out in the UK

Google has finally confirmed that its AI mode is now rolling out in the UK, having already been live in the US for a few months.

Instead of a list of search results showing links to other websites in blue type, users who choose “AI Mode” will be given an answer written in a conversational style, containing far fewer links to other pages.

While AI Mode continues to be optional, users can still use traditional search.

What does this mean for search strategies?

It means:

• GSC doesn’t yet separate clicks from AI Mode from overall site clicks, making attribution harder than before
• Traditional position 1 or top 10 rankings will lose their impact
• Increased personalisation – users on AI mode will likely receive different results for the same query
• Informational content will see a decline in traffic as queries are answered on the SERPs
• Using PR and earned media will be key in showing up in LLMs and AI overviews

YouTube Shorts introduces AI tools for photo to video

YouTube Shorts has introduced new creation tools to enhance video making. The Photo to video feature turns pictures into animated videos, available now in select countries.

New generative effects transform doodles and selfies with AI-powered visuals. The AI playground offers tools for generating videos, images, and music, accessible in certain regions. All AI creations include SynthID watermarks and labels to ensure transparency.

Google has been testing a zoom functionality on images that feature in search results

Users in the industry have spotted that Google is testing a zoom effect on images that feature in the search result snippets. This looks to occur when you move your mouse cursor over the images.

Barry Schwartz has posted a screen recording example of this in action on X. Check out the video here.

Google announce second Core Update of 2025

Google announced its second Core Update of 2025. This follows the March Core Update, which we noticed that the sites that were hit most were those producing a lot of content programatically via AI.

SEO tracking tools are already detecting fluctuations, even if many site owners haven’t yet noticed changes. This update is expected to roll out over three weeks, longer than the typical two-week period. It’s still too early to see the full impact of this Core Update, however, it’s likely to negatively impact websites with low-quality content.

At Embryo, we’ll continue to monitor the impact on our clients and the wider industry.

Google updated guidelines for sites with explicit content

Google has released updated guidelines for sites with explicit content and what to do if your site is flagged incorrectly.

The update includes:

• Revised guidelines for how explicit content is handled in Search, and covers how SafeSearch works and best practices for websites who do host this type of content.
• Changes to ranking algorithms to target sites that host explicit videos but block Googlebot from accessing the files, and these sites may see a “significant drop in rankings.”

They’ve also updated guidelines around what to do if your site is incorrectly flagged, which can sometimes happen for sites that are subtly suggestive such as lingerie websites or sex education sites. 

Google Marketing Live: GA4 and performance monitoring

Google has just finished its ‘Google Marketing Live,’ and here is what we learned about GA4 and campaign measurements.

Cross-channel insights in GA4 are getting an upgrade, making it easier to track performance across platforms.

The minimum spend for incrementality testing will be reduced from $100K to $5k

A new Bayesian-based testing model means you can run tests in as little as 7 days.

Data Manager will consolidate first-party data from BigQuery, HubSpot, Salesforce, and Shopify, and a new API will facilitate smoother integrations.

They have introduced “Google Analytics experts” for in-depth analysis and strategic advice, and a “Marketing Advisor”, an AI agency assistant that offers step-by-step guidance.

And the big one, a unified view of impressions across both Google platforms and third-party data. This allows you to bridge the gap when understanding the full funnel.

Meet AI Mode – Google’s new expanded search feature

At the annual I/O developer conference, Google announced that it is upgrading its AI features, adding features like AI mode and AI overview.

What is AI Mode?
AI Mode is a new system powered by query fan-out. It breaks down big questions into smaller ones and runs multiple searches simultaneously to give more tailored, helpful results.

It’s designed to do more than answer questions. It can now help with shopping, too, surfacing personalised results like “Oasis tickets within 100 miles of Manchester” (we wish!). You’ll also see features like image results, visual try-ons, and sale alerts.

What does this mean for your website?
AI is quickly becoming the norm, which means optimising your website is more important than ever. Make sure it’s focused on your customers, aligned with search intent, and your content is delivering real value.

New study finds users only read a third of Google’s AI Overviews

A new UX study by Kevin Indig and Eric van Buskirk has revealed some interesting insights about how users interact with Google’s AI Overviews:

• Most users skim AI Overviews: The median scroll depth in AI Overviews is 30%, meaning most users don’t read more than the top third.
• Citations get few clicks: Only 19% of mobile users and 7.4% of desktop users clicked on a citation provided in the AI Overviews.
• Many turn to Reddit for social proof: When users leave the AI Overviews, a third go to Reddit, YouTube, or other forums.

This seems to confirm what many SEOs have feared: that AI Overviews mean a lower click through rate. However, the study’s sample size is small compared to Google Search’s massive scale, which means the insights can’t be used to make any definitive conclusions.

Still, the study’s insights are interesting and a reminder that online visibility is increasingly shifting from clicks to SERP visibility.