We have covered some essential points. For full info, check out the sources listed. Without further ado, let’s dive in.
1. The New Search Landscape
According to Ahrefs’ September 2025 data, Google still holds 90.4% of global search share.
But in AI assistants, ChatGPT commands 81.13%, followed by Perplexity (10.82 %), Copilot (4.05 %) and Gemini (2.82 %).
Stox quoted Ahrefs’ Ryan Law: “Google still sends far more traffic … AI search does not replace traditional search; it extends it.”
What it means:
The world is becoming dual-layered. Traditional search (Google, Bing, etc.) remains central, but a parallel layer of discovery is now emerging via AI assistants.
SEO impact:
You can no longer optimize only for Google. You also need to appear in AI assistant outputs and be citable by those models.
How to prepare (SEO pros):
Track both traditional organic traffic and AI referral/citation presence (if tools allow).
Maintain strong core SEO (technical, content, links) to retain Google visibility.
Simultaneously invest in making your content “AI-friendly” (clear, structured, referenceable) so it’s more likely to be surfaced by assistants.
2. AI Traffic Growth Is Explosive
Ahrefs’ Web Analytics (82,000 sites) shows AI-generated traffic has increased ~9.7× year-over-year.
The trendline in Ahrefs’ visualizations heavily favours referrals from ChatGPT over other assistants.
What it means:
Users are increasingly turning to AI assistants rather than classic search interfaces to get information, recommendations, or answers.
SEO impact:
A growing share of your potential audience may come via AI systems, meaning only following traditional tactics could yield diminishing returns.
How to prepare:
Audit your existing content: which pages are most likely to be cited (data, studies, definitive explainers).
Prioritize content that answers direct questions, provides unique insight, or can be cited as a reference.
Monitor and experiment with featured snippets, structured data, and FAQ blocks; they may help content be surfaced in AI summaries.
3. The “Great Decoupling”
In analyses of ~800,000 Google Search Console properties, impressions are increasing, but clicks are declining.
AI Overviews (generative summaries) now appear in ~19% of SERPs.
In March 2025, average CTR for position #1 dropped by 34.5%.
Stox projected that clicks could shrink to 25% of their former levels.
What it means:
As AI summaries become widespread, users often get answers directly without clicking through to sites. The traditional “clicks = engagement” model weakens.
SEO impact:
Relying purely on click volume (and CTR) as success metrics will become misleading. Impressions, visibility in AI contexts, and conversions matter more than ever.
How to prepare:
Begin tracking impressions, coverage, mentions in AI outputs/summaries, not just clicks.
Focus on conversion and engagement metrics (time on page, scroll depth, lead generation) as stronger indicators of value.
Ensure pages are optimized for snippet format, with clear answer blocks, structured data, and well-formatted content that’s easily “digestible” by AI.
4. AI Traffic tends to favour Desktop.
Device data from June 2025:
AI Traffic: 65 % desktop | 21 % mobile
Organic Search: 76 % mobile | 21 % desktop
What it means:
Many AI discovery sessions still originate on desktop platforms (possibly via web or desktop apps), whereas organic clickthroughs remain mobile-heavy.
SEO impact:
Mobile optimization is still necessary for organic search, but your strategy for AI will need to ensure desktop experience (load speed, readability) is strong, too.
How to prepare:
Test pages on desktop formats, especially for readability, speed, and layout.
Ensure content scales well between desktop and mobile views (responsive, modular sections).
Consider differences in how AI contexts render results, e.g., desktop snippets may favour richer content blocks (tables, infographics).
5. Brands dominate AI Overviews via mentions & entity signals
Ahrefs found branded web mentions (0.664 correlation), branded anchor text (0.527), and branded search volume (0.392) had the strongest correlation to being featured in AI Overviews.
In sum: having your brand name talked about across the web (not only your site) greatly increases your chance of being cited by AI systems.
What it means:
AI systems lean more on entity recognition and textual signals (mentions) than just backlinks or domain metrics when deciding what to cite.
SEO impact:
Traditional link building is still valuable, but entity-based visibility and brand mention authority now play a bigger role in AI visibility.
How to prepare:
Invest in digital PR, brand mentions, guest posts, authoritative coverage, and offline reputation to build your brand’s presence across the web.
Encourage publications to link with branded anchor text when possible.
Monitor brand mention growth, sentiment, and context across platforms.
Consider how your content can be referenced (data, quotes, case studies) to make it “citable.”
6. Content type & language matter; translation is essential
AI tools prefer to cite informational / research / data-heavy pages rather than product or homepage content.
International (non-English) content underperforms unless translated.
What it means:
AI systems are more likely to surface referenceable content (reports, studies, explainers) and favour content in languages well represented in their training corpus.
SEO impact:
If your content is deep and data-driven, it has better odds of being surfaced in AI answers. But if you only publish in less common languages, you may be excluded.
How to prepare:
Prioritize creating research, data studies, whitepapers, and tutorials that have content with depth and citation potential.
Localize and translate high-potential pages to regional languages to increase the chance of AI citation.
Use structured data, language tags, and hreflang properly to help AI systems interpret your multilingual content.
7. The “black hat era” is here
Stox warned of risks: parasite pages, expired domains, hacked content, spam from forums (Reddit, Quora), and data poisoning (false data entering models).
What it means:
As AI value becomes visible, some will try to game AI systems using low-quality or manipulative tactics to get surfaced.
SEO impact:
If AI systems begin to reward manipulated content, the signal-to-noise ratio increases. The credibility and reliability of your content become even more critical.
How to prepare:
Play ethically, don’t rely on spammy or black-hat shortcuts.
Monitor your content, site integrity, and backlinks, and ensure no unintended vulnerabilities (e.g., hacked pages).
Use fact-checking, maintain transparent citations, and encourage validation of data.
Stay informed about how AI systems evolve detection of manipulative tactics.
Summary & Strategy Checklist for SEO Pros
Strategy Pillar | Key Focus | Why It Matters |
|---|---|---|
Balanced Visibility | Optimize for both Google and AI assistant discovery | Dual-layer now; ignoring either is a risk |
Content Authority | Produce citable, data-rich, structured content | AI surfaces content as a reference |
Brand Mentions | Grow entity presence via PR, coverage, mentions | AI visibility correlates strongly with mention signals |
Metrics Evolution | Track impressions, citations, conversions, not just clicks | Clicks may decline due to AI Overviews |
Localization | Translate and localize high-impact content | AI models favour content in well-represented languages |
Ethical Practices | Avoid spam, manipulation, and low-quality shortcuts | Signals matter, credibility builds long-term trust |
In Conclusion
Patrick Stox’s presentation at Ahrefs Evolve 2025 made one thing crystal clear that search isn’t dying; it’s evolving.
We’re entering an era where discovery happens across two layers:
Traditional Search, driven by keywords, links, and rankings.
AI Search, powered by citations, mentions, and contextual authority.
The challenge for SEO professionals now isn’t to pick one, it’s to balance both.
Track how you rank and how you’re referenced. Optimise for visibility and credibility.
Stay curious. Experiment responsibly.
Sources:
