The evolution of SERPs: From 10 blue links to AI Overviews

Why SERPs never stand still

Twenty years ago, Google’s search results were famously simple: ten blue links, a handful of ads, and little else. Today, a single query can surface AI-generated summaries, maps, images, videos, shopping results, and forum discussions – often before a user clicks anything.

For CMOs, this shift can feel unsettling. But it shouldn’t be surprising. Search has always evolved in response to how humans seek information. AI Overviews are not a rupture – they are the latest chapter in a long story of search becoming faster, richer, and more efficient.

This article traces the evolution of SERPs over the last 20 years, places AI Search in its historical context, and outlines what leaders should focus on next.

Search before Search Engines: a human constant

Long before Google, humans relied on oral knowledge, libraries, encyclopaedias, and trusted experts. Each technological leap – the printing press, indexing systems, the internet – reduced friction between a question and its answer.

Search engines didn’t change why we search. They changed how quickly we can satisfy curiosity. From that perspective, AI-generated answers are not radical. 

They’re a continuation of the same trajectory: compressing effort, time, and cognitive load.

2005–2009: the era of 10 blue links

In the mid-2000s, SERPs were linear and predictable. Organic rankings were largely determined by links and keyword relevance, while paid ads appeared above or beside results.

Key developments:

• 2005: Introduction of sitelinks for navigational queries
• 2007: Universal Search blends news, images, video, and maps into organic results

Universal Search marked the first major break from uniform results, forcing brands to think beyond webpages and optimise for multiple content types.

2010–2013: richer results and the knowledge graph

As broadband improved and user expectations rose, Google accelerated SERP innovation.

Key milestones:

• 2010–2011: Autocomplete begins shaping user intent. Google’s predictive search suggestions started influencing how people phrased queries, subtly guiding demand and shortening the path to answers.
• 2012: Launch of the Knowledge Graph – “things, not strings”, representing a significant shift towards optimising for Entities.

Knowledge Panels allowed Google to answer factual queries directly, reducing dependency on clicks while increasing trust in search as an authority.

This shift laid the foundation for entity-based SEO, explored further in Hallam’s guide to entities in SEO.

2014–2015: featured snippets, local packs and mobile-first reality

The mid-2010s introduced the concept of zero-click search.

Key developments:

• 2014: Featured snippets (“Position Zero”) appear above organic results
• 2015: Mobilegeddon (or “Mobile-first”) prioritises mobile-friendly pages in SERPs
• 2015: Google reduced the number of local businesses shown in search results, concentrating visibility on fewer, more prominent listings – a move driven by mobile usage and user behaviour.

Featured snippets rewarded structured, authoritative content, and previewed a future where Google increasingly answers questions itself.

2015–2019: AI enters the algorithm

While SERP layouts stabilised, Google’s understanding of queries improved dramatically.

Key milestones:

• 2015–2016: RankBrain applies machine learning to ranking systems, allowing higher quality and relevancy in search results
• 2016: “People Also Ask” introduces expandable, question-led search journeys. Google began surfacing follow-up questions directly in the results, encouraging users to explore topics without reformulating searches or clicking through multiple pages.
• 2019: BERT enables contextual understanding of natural language

These updates reinforced a critical truth: optimisation was no longer about matching keywords, but about satisfying intent.

For brands, this meant investing in depth, clarity, and credibility – not shortcuts.

2020–2022: SERPs become experiences

The early 2020s saw Google reshape SERPs into immersive experiences:

• Continuous scroll replaces pagination
• Video and visual search gain prominence
• MUM expands multi-modal understanding

At the same time, user behaviour fragmented. Discovery increasingly happened across YouTube, TikTok, Reddit, and marketplaces – not just Google.

This period reinforced the need for a Total Search mindset, explored further in our article on navigating the era of AI-powered search.

2023–2025: AI Overviews and generative search

The launch of AI Overviews represents the most visible SERP shift in a decade.

AI Overviews:

• Generate original summaries from multiple sources
• Prioritise synthesis over listing
• Reduce clicks for informational queries

Early data suggests increased zero-click behaviour (60% of Google Searches now don’t result in a click), but also new opportunities for citation-level visibility.

Importantly, Google has been clear: AI Overviews still rely on traditional ranking systems and high-quality sources, and we do not need to reinvent the wheel by calling it “GEO”.

What this means for CMOs

AI Overviews feel disruptive because they’re visible. But strategically, they reward the same behaviours that have driven sustainable SEO for years.

John Mueller has repeatedly reinforced that there is no separate optimisation playbook for generative search – strong SEO foundations remain the prerequisite.

In other words: good GEO is good SEO. 

Actionable priorities for AI Search readiness

1. Invest in authority, not volume

AI systems preferentially cite trusted sources. Digital PR, expert-led content, and brand mentions matter more than ever – as explored in Hallam’s perspective on Digital PR in 2026.

2. Optimise for entities and understanding

Entity clarity helps AI contextualise your brand. This means:

• Clear topical ownership
• Structured data
• Consistent messaging across channels

3. Structure content for answers

Well-structured pages: clear headings, concise explanations, supporting evidence, are more likely to be surfaced in AI summaries.

4. Avoid chasing “GEO hacks”

There is no shortcut. Brands chasing AI-specific tricks risk undermining long-term performance. Instead, focus on:

• Technical excellence
• User-first content
• Credibility signals

Adaptation is the advantage

From encyclopaedias to AI summaries, the story of search is the story of human adaptation. Each evolution has reduced friction, and each has rewarded brands that prioritise usefulness, trust, and clarity.

AI Overviews are not the end of SEO. They are a reminder of what SEO has always been about.

For CMOs, the mandate is clear: invest in fundamentals, embrace change calmly, and build brands that deserve to be cited – by humans and machines alike.

To find out more about the work we do and how we can help you, get in touch here.