What Google’s latest AI search guidance really means for marketers

The search landscape is shifting quickly, with features, user behaviour and industry commentary updating almost every day.

Google is obviously central to that shift towards AI search, as it continues encouraging users towards generative AI experiences through features like AI Overviews and AI Mode.

That makes its “Guide to Optimising for Generative AI Features on Google Search” particularly interesting – not just because it shows how Google wants organisations to think about AI search, but because it offers a clearer view of what still matters if you want your website to be found in these experiences.

Google is trying to calm the AI search panic

First off, Google is clear that marketers don’t need to abandon SEO and learn an entirely new discipline. It says its generative AI features in Search are rooted in its core Search ranking and quality systems, and that from Google Search’s perspective, optimising for generative AI search is still SEO.

The best practices for SEO continue to be relevant because our generative AI features on Google Search are rooted in our core Search ranking and quality systems.

(Source: Google Search Central)

SEO still matters in generative AI search

There can be a temptation with AI search to assume that everything has changed overnight.

But Google’s guidance keeps pulling things back to the same core themes:

• Content that exists to help people, not just to manipulate visibility
• Clear technical structure
• Easy crawlability
• Indexability of pages
• Good user experience

While the core SEO principles remain, the landscape is certainly evolving. This reinforces the idea that marketers should be careful not to overreact to every new idea, file type or shortcut that promises immediate visibility in AI search experiences. 

If anything, AI search seems to be making the basics more important, not less.

Helpful, reliable, people-first content is still the standard

One phrase Google keeps returning to is “helpful, reliable, people-first content”. It has used that language for a while now, and it appears again throughout this guidance.

Helpful, reliable, people-first. This tells us what Google still wants from content, whether it’s for the traditional search engine or its generative AI features.

Not content for the sake of it. Not slight variations of the same page multiple times to match different keywords. Not AI-generated filler written to cover every possible version of a query.

Instead, the guide points marketers back towards content that is:

• Genuinely useful
• Organised clearly
• Distinctive
• Based on knowledge or experience
• Written for people first

One of the most useful parts of the guide is Google’s emphasis on “non-commodity content”. It explicitly talks about the value of unique points of view, first-hand reviews, original experience and content that goes beyond simply restating what already exists elsewhere.

This makes sense when thinking about how AI tools and LLMs work. If they’re increasingly able to synthesise common knowledge, then content based only on common knowledge becomes much easier to replicate. 

The stronger opportunity is in content that actually brings something to the table:

• Original experience
• A strong point of view
• Useful quotes
• Compelling evidence
• An author biography
• Statistics – especially proprietary ones 
• Images and video that support the experience

That is also where E-E-A-T still comes in. Experience is especially important here, because it is one of the clearest ways to create content that feels distinct, credible and hard to commoditise.

Technical clarity still matters

As with traditional SEO, pages still need to be indexed and eligible to appear in Search with a snippet if they are going to appear in Google’s generative AI search features. 

Google’s guidance points to crawling best practices, JavaScript SEO (making sure content isn’t being blocked by JavaScript), page experience and reducing duplicate content, which you can read more about here. 

From an on-page content point of view, a lot of the same recommendations still stand. Clear headlines, useful subheadings, short paragraphs, bullet points and tables still help people scan and understand information. They also make it easier for systems to interpret that information.

Structured data has seen a surge in interest alongside the rise of AI search, despite often being an underutilised SEO feature in the past. While Google’s guide says that “Structured data isn’t required for generative AI search”, it also states that maintaining structured data remains a logical and effective component of a comprehensive SEO approach.

What Google says you can stop worrying about

Google has officially shared a stance on many tactics and features that have become part of the AI search conversation this year, directly pushing back on several of them, including:

• llms.txt and other “special” AI files
• Chunking content purely for AI
• Rewriting content just for AI systems
• Inauthentic mentions across the web

However, there is some debate on even some of these points – even within Google. The llms.txt point is probably the best example of this. Google Search Central says: “You don’t need to create new machine readable files, AI text files, markup, or Markdown to appear in generative AI search.”

But, at the same time, Google’s Lighthouse documentation for developers acknowledges that llms.txt is an “emerging convention” for LLMs and agents, and provides instructions on how to create the file. Basically, the jury is still out on how useful it may become in broader agentic experiences. 

The llms.txt debate feels like a good example of where we as marketers are right now. This space is still evolving, what might matter today might not tomorrow (and vice versa), and that good, well structured, human-first content is still the priority. 

Key takeaway

Ultimately, I think the fundamentals haven’t changed. Focus on these pointers and your foundation for success is strong:

• Creating genuinely useful content
• Clearly demonstrating expertise and experience
• Building a logical website and page structure
• Prioritising a fast, intuitive user experience

If you want to sense-check how your brand is appearing in AI search, explore the AI Performance Analyser.