How the old rules of Search still sit within the new

Search Engine Optimisation (SEO): A Brief History and Key Actions in the New AI Search Era

Why the fundamentals still hold, and what the numbers now tell us.

Originally published on LinkedIn, September 2025. Updated May 2026 to reflect the accelerated adoption of AI search, significantly revised statistics, the emergence of Google AI Mode, and the new disciplines of Generative Engine Optimisation (GEO) and Answer Engine Optimisation (AEO).  Both different and clarified below.

Context in the form of a brief history

For two decades, and from first-hand experience, the only rule to ‘getting discovered’ online was “to get ranked on Google.” Initially, this was achievable through keywords and back-end technical work (meta tags, and the like) largely hidden from the website owner of the time.

Then marketers stepped in as the semantic web emerged and contextual content became better understood. Best practice was shaped by Google’s EAT parameters - Expertise, Authority and Trust - the premise being that if you write as a genuine expert in your field, authority and trust naturally follow. Google even provided Trust and Authority scores within their analytics. It made sense, was easy to remember, and had a built-in metric.

Searchers also began using longer phrases to find answers, known as ‘longtail’ search queries - the jewel in the crown. A well-written blog answering that exact question inched you up the rankings. Playing by these rules satisfied both users and Google’s bots.  And is this not rinse and repeat today?

Organic SEO is a long-haul job, and proving its value is not straightforward. The effort invested has to be balanced against the type of business you are and how important it is to drive traffic to your website. The long and the short of it is that Google no longer holds all the cards, and this has been building for some time.

The good practice fundamentals, however, are already in place.

The 2 clear shifts of SEO today

1. ‘Googling It’ is no longer the whole story

Google handles approximately 13.7 billion searches a day, but that accounts for only around 27% of all search activity on the internet. The remaining 73% is scattered across TikTok, YouTube, Reddit, Instagram, Amazon, and AI platforms including ChatGPT and Perplexity.

This fragmentation has driven the rise of Zero-Click Searches, where a user’s query is answered directly on the search results page without ever clicking through to a website. In 2024, approximately 58.5% of all Google searches ended this way. By 2026, that figure has reached approximately 65%, nearly two thirds of all Google searches. For searches where an AI Overview is triggered, the zero-click rate reaches 83%.

This evolution on Google’s own platform is a signal of a far broader shift - the rise of new AI gatekeepers that answer questions directly.

2. The new AI gatekeepers

Large Language Models (LLMs) such as ChatGPT and Google’s AI Overviews represent a clear structural shift. Instead of returning a list of links to research, these tools synthesise information from multiple sources and deliver a single, direct, conversational answer. As an analogy: traditional search handed you a pile of books. AI search hands you the expert’s summary.

In 2025, Google introduced AI Mode — a dedicated AI-powered search experience with a 93% zero-click rate. AI Overviews now appear on approximately 50% of all Google searches, up from just 18% in early 2025. When an AI Overview is present, organic click-through rates fall by an average of 58%.

That said, pragmatism applies. A user seeking serious answers to considered questions - around a B2B purchase decision, a professional service, a complex problem - will generally look beyond the AI summary to the source. As always, context matters.

What is undeniable is that this is a new model, a significant step in SEO transformation, and one that every business should now address in terms of their online presence, particularly if your business depends upon online visibility.  Of course, there are exceptions, and an industrial people to people business will find this less pressing.  However, unlikely as it may now seem, one has to imagine that another way of doing it is there.

New mindsets to adopt

• From keywords to context to conversations: Questions asked of AI appear to be more fully understood, with comprehensive answers provided. There has always been a need to answer questions in your content. Now, it is an absolute goal to know the questions your customers ask, and to deliver the answers clearly and directly.
  
  
• AI as the new gatekeeper: Google is no longer the only starting point for discovery. AI systems are increasingly influencing how users find information. You must now earn a place in the AI’s answer - which is the new front page.
  
  
• Visibility beyond Google: Discovery and decision-making are now happening across a wide array of platforms, from TikTok to Reddit. Your focus must shift from ‘search moments’ on Google to ‘decision moments’ happening everywhere else.  TikTok is not the exclusive playground of lifestyle creators and Reddit's subs cover, seemingly, every topic in the universe.  An engineering firm will know that his audience is no longer on the traditional trade content portals, so new avenues must be explored.  
  
  
• RETAIN trust signals as a priority: Reviews, comments, and discussions anywhere on the web are trust signals. AI models are trained to prioritise content that feels reliable, authoritative, and human; no doubt because their predecessor - let's take Gemini's 'grandad' Google Search (the one that delivered good old fashioned 'websites') managed to pass down some sound ethical advice . Building a web of trust signals across platforms is no longer a nice-to-have, it is the core of modern SEO, just as it was the core of 'traditional' SEO. Ironically, in a world of AI, the word ‘human’ is on overdrive.
  
  
• Human-centred content is king: Chasing algorithms unfortunately had a rebirth, noticeably on LinkedIn.  I can't help feeling the marketing profession was behind this - where 'posting several times a day' was the key to gaining impressions on LinkedIn. It felt wrong, well, particularly if you experienced this first time around. It remains imperative to create content that adds genuine value and solves real problems. Your content must be the human-centric source material that AI models prefer to cite -this may be ironic but let's take the good - it's a desired ethical standard.
  
  Specific platform characteristics:
  
  
• TikTok: Decisions driven by emotion and novelty.
• Reddit: Decisions based on raw authenticity and unfiltered user opinion. Reddit now accounts for 46.4% of AI social media citations.
• YouTube: Decisions based on perceived expertise and depth. YouTube accounts for a further 31.8% of AI social media citations.
• Amazon: Decisions driven by social proof and reviews.
• ChatGPT / LLMs: Decisions based on semantic clarity and citations from authoritative sources.
• LinkedIn: While the most-cited domain for professional queries across AI Overviews, AI Mode, ChatGPT, Microsoft Copilot, and Perplexity, cynics may argue that it is the only platform for B2B business and networking (in the UK). Linked in is not the platform for every business, when one considers who, exactly, is active on it.  

Are the old SEO rules obsolete?

This article shows how the rules have evolved, rather than changed. The fundamentals of traditional SEO still apply which is good news for any business that has invested in its organic presence. Critically, 99.5% of URLs cited in AI Overviews overlap with pages already in Google’s top 10 organic results. In other words, traditional SEO performance and AI visibility are not competing priorities, they're reinforcing each other.

Google’s E-E-A-T framework - Experience, Expertise, Authoritativeness, Trustworthiness - remains vital. The additional ‘E’ for Experience was added in 2022. These are precisely the signals that AI uses to assess whether your content is reliable enough to be included in a 'synthesised' answer. Well-structured, authoritative content remains the raw material for AI.

The ‘Search Everywhere Optimisation’ roadmap — now with a name

The industry has now formalised the approach outlined in the September 2025 version of this article: it is called Generative Engine Optimisation (GEO), alongside Answer Engine Optimisation (AEO). The goal is to be cited, referenced, or summarised by AI-powered answer engines - not just indexed by Google. GEO sits alongside traditional SEO, not in its place.

The strategy is ‘Search Everywhere Optimisation’: being present and chosen on the various platforms where your customers make decisions. This requires a mental shift from optimising for ‘search moments’ to optimising for ‘decision moments.’ Each platform carries its own decision logic.

The goal is not to be everywhere at once. It is to understand where your customers’ most critical decision moments happen and to build a strategic, trusted presence on those specific platforms.

The reality:  attaining the above, and this is where strategy on paper barely resembles the reality once you have to apply it in often very complex businesses.  See my article called 'What marketing techniques do not make commercial sense for startup or established, yet niche, b2b.

3 action steps to adapt to this new landscape

1. Think ‘Quotable,’ not just ‘Clickable’

The new position one is not a blue link, it is being cited directly in an AI answer, mentioned in a Reddit thread, or referenced in a TikTok video. Your goal is for your expertise to become part of the conversation, making your brand a trusted source that both AI and people rely upon.

The data now tells us precisely how to improve your chances. Structured content significantly increases AI citation rates: comparison pages with tables, list-based content, and short sentences (under ten words on average) all perform measurably better in AI systems. Original data like surveys, case studies, proprietary statistics makes you the primary source AI cites.

2. Become a detective

Perform a simple audit. Go to Claude, ChatGPT, Perplexity, or use Google’s AI Overview or AI Mode feature. Ask the questions your customers would ask about your industry, your products, or the problems you solve. Does your brand get cited? Do your competitors? The answers will reveal your current visibility gaps in the AI ecosystem.  It works.  

This audit is now more urgent than it was eight months ago. AI Overviews appear in roughly 50% of searches. AI Mode is live. Nearly 50% of UK adults now use AI-assisted tools to find information - up 22 percentage points year on year. The gap between businesses visible in AI answers and those that are not is widening.

3. Shift from chasing visibility to earning validation

AI summarises based on trust signals from across the web. Visibility is showing up (for example, your own LinkedIn page). Validation is being mentioned by others (another user referencing your brand, a forum discussion citing your expertise, a positive review on a relevant platform). AI prioritises validation - making it the most valuable currency in this new era.

Being mentioned in Reddit communities, cited in expert articles, reviewed positively, and discussed in group forums are now mission-critical SEO activities. This really does extend far beyond traditional SEO. The practical implication: get your front-line staff engaging helpfully on group forums and being genuine ambassadors for your business. Authenticity, at scale, is the strategy.

Conclusion: the opportunity ahead

The end of the ‘ten blue links’ as the primary method of discovery represents a fundamental shift in strategy and thinking but it also presents a significant opportunity for businesses willing to adapt. The fundamentals that drove traditional SEO success are the same fundamentals that drive AI visibility. That is not a coincidence; it is the point.

By focusing on genuine trust, real value, and a strategic presence where decisions are actually made, you can build a stronger, more resilient, and more visible brand than ever before. The game has changed but the rules, at their core, have not.

Are you already exploring what your AI audit reveals? For businesses across Hampshire and Wiltshire, understanding where you currently appear — or do not appear — in AI-generated answers is a practical first step that costs nothing and reveals a great deal. It would be interesting to hear how businesses in the region are approaching this.

Useful sources

Semrush: Zero-Click Searches Study 2025

Ahrefs: AI Overviews CTR Impact, February 2026

Profound: LinkedIn Most-Cited Domain in AI, March 2026

BrightEdge: AI Overviews Prevalence, February 2026

OtterlyAI: AI Social Media Citations Report, April 2026

SEO Works: 75 AI SEO Statistics for 2026

Google: E-E-A-T and Search Quality Guidelines

© The Real Marketing Company 2026. This article is provided for informational purposes. For specific guidance on your digital marketing strategy, visit www.therealmarketingcompany.co.uk.

www.therealmarketingcompany.co.uk