The numbers nobody wanted to publish, all in one place
If you've been watching your Google Search Console graphs flatten over the last 12 months while your impressions kept climbing, you weren't imagining it. Five independent research studies have now confirmed what every SEO practitioner has been quietly suspecting since Google rolled out AI Overview at scale.
The game changed. Most businesses haven't adjusted.
Here's exactly what the data says, what it means for your traffic, and the four plays that still work in 2026.
Five studies, one brutal pattern
Pew Research: 8% CTR when AI Overview is present
A Pew Research Center analysis of Google search behaviour found that when an AI Overview appears at the top of the search results page, only 8% of users click any blue link below it. When no AI Overview is present, the click rate is roughly 15%.
Clicks didn't shift. They evaporated. The user got their answer and closed the tab.
Ahrefs: Position 1 CTR down 58%
Ahrefs analysed click-through rates across 300,000+ keywords before and after AI Overview rollout. The position 1 organic result, which historically captured roughly 30% of clicks, is now seeing 58% lower CTR on queries where AI Overview appears.
For years, ranking #1 was the prize. The prize just got cut in half.
Seer Interactive: Organic traffic down 61%
Seer Interactive tracked 15 months of client organic traffic data through the AI Overview rollout. Across their portfolio of B2B and consumer brands, organic traffic from informational queries dropped by 61%.
Not impressions. Not rankings. Actual traffic to actual websites.
Daily Mail: From 25% CTR to 2.8%
The Daily Mail's organic traffic from Google fell from a 25% click-through rate to 2.8% after AI Overview began answering news queries directly. A nearly 90% collapse for what used to be a high-trust, high-relevance traffic source.
If you publish content that AI can summarise, you are now competing with the answer itself.
The hidden upside: Brand keywords +18% CTR
Buried in the same studies is one counterintuitive finding. Branded search queries with AI Overview present saw CTR increase by 18.68%.
Why? When someone searches for your brand by name, they have intent to find you, and AI Overview tends to surface your site as the authoritative answer. The AI doesn't compete with your branded SERP, it amplifies it.
This matters more than it sounds.
Why this is happening
Google's AI Overview is doing something fundamentally different from a traditional snippet or featured answer. It synthesises information from multiple sources, presents a complete answer, and gives the user no reason to click through.
This works for Google because users still get value, ad revenue still flows from the SERP, and Google's market share is preserved. It does not work for publishers and businesses who built their funnels on the assumption that ranking would deliver traffic.
A few realities are now permanent:
- Informational queries are largely gone as a traffic source. "How do I do X" or "What is Y" questions are answered by AI Overview before users scroll.
- Comparison and listicle content is being eaten alive. "Best X in 2026" type queries trigger AI Overview almost universally now.
- Local search still works, but with declining click rates. Map Pack results retain most of their CTR but the standard local 10-blue-links zone is shrinking.
- Transactional and branded queries still convert at high rates. When users know what they want and who they want it from, AI Overview helps rather than hurts.
What this means for SEO strategy in 2026
The four plays that still produce results, ranked by impact for an Australian small business or mid-market brand:
1. Optimise for citation, not for ranking
The new game is not "rank #1 on Google." The new game is "be cited by Google's AI Overview, ChatGPT, and Perplexity when your category gets queried."
This is a different optimisation problem. Citation favours:
- Specific, hard data that AI can extract and quote
- Original research that no other source has
- Clean structured Q&A formatting that maps to how AI generates responses
- Strong topical authority signals built across multiple platforms, not just your domain
- Schema markup (FAQ, HowTo, Article, Organization) that helps AI parse your content
Writing another generic 2,000-word "What is digital marketing?" article in 2026 does almost nothing. Writing a single piece of content with proprietary research, specific Australian numbers, and clear answers to specific questions can get cited dozens of times by AI Overview for years.
2. Double down on brand search
Branded queries are the only segment where CTR is increasing. That makes brand-building activities measurably more valuable than they were two years ago.
Ways to drive branded search:
- LinkedIn presence from founders and senior team, not just a company page
- YouTube videos with specific value people would search for by your name
- Podcast guest appearances in your industry's listening habits
- Public speaking, conferences, and panel discussions
- Original research and reports that get cited by other publications
Each of these creates a moment where someone learns your brand name, then later searches for it. That branded search is where AI Overview actually helps you rather than hurts you.
3. Track AI visibility, not just rankings
The SEO dashboards most agencies still send their clients (rank tracker results, position 1-10 keyword counts, monthly impressions) are increasingly meaningless. The dashboard that matters in 2026 looks more like:
- Are you mentioned in ChatGPT responses for your category? Test monthly.
- Does Perplexity cite your domain when users ask about your service? Test monthly.
- Does Google AI Overview pull from your content? Check the AI Overview source list.
- What is your branded search volume month over month? This is the new north star metric.
- What is your share of voice in industry-specific subreddits and LinkedIn discussions?
If your agency cannot answer these questions, your reporting is stuck in 2023.
4. Long-tail is the safe harbour
AI Overview is triggered most aggressively on high-volume, broad queries where users want fast answers. It triggers far less on:
- Hyper-specific long-tail queries with implicit local or commercial intent
- Multi-step buying decisions where users want to compare
- Niche professional queries in regulated industries
- Questions that require recent, time-sensitive information
For most Australian SMBs, this is actually good news. The traffic that converts at the highest rates was always long-tail commercial intent. Those queries still send traffic to websites. The high-volume informational queries that we lost were never going to convert anyway.
The brands that built their content strategy on listicles and "ultimate guide" content are getting hammered. The brands that focused on specific, useful, technical content for their actual customers are mostly fine.
What Australian businesses should do this week
Four concrete actions:
1. Pull your last 12 months of GSC data. Look at impressions vs clicks. If impressions are flat or up but clicks are down 30 to 60 percent, you are inside this trend. Confirm before you act.
2. Identify your top 20 informational pages and your top 20 commercial pages. The informational pages will not recover. Reposition them to capture downstream interest, not direct traffic. The commercial pages should still be defended.
3. Add proprietary data to your highest-effort content. Numbers from your own client base. Examples from your own market. Information that AI cannot synthesise from other sources because it does not exist anywhere else.
4. Audit your AI visibility manually. Open ChatGPT and Perplexity. Search the questions your customers actually ask. Note whether your brand appears. If it doesn't, you have a content gap that no amount of traditional SEO will fix.
The honest assessment
For most Australian SMBs, AI Overview is not a catastrophe. The pages losing the most traffic are usually the pages that produced the least revenue anyway. Top-of-funnel informational content was always a poor predictor of pipeline.
The businesses that will struggle in 2026 are publishers, content-driven affiliate sites, and agencies whose entire model was "write more articles to capture more keywords." The businesses that will thrive are the ones that treat their website as one of several places they show up, not the only one.
The rules of SEO didn't disappear. They got broader. "Search Engine Optimisation" is becoming "Search Everywhere Optimisation," which is just another way of saying brand-building has never been more measurable.
Need help adapting?
At DC Groups, we help Australian businesses make this shift: from rank-chasing to citation-chasing, from broad informational content to specific commercial intent, from single-channel SEO to multi-platform brand visibility.
If your traffic curves don't look right and you want an honest second opinion, send us a message. We'll look at your last 12 months of GSC data, identify what's recoverable and what's not, and tell you where to actually invest.
No pitch deck. No packaged proposals. Just a clear read on where you stand.