Google AI Overviews are killing our click-through rates - anyone else seeing this? Need strategies that actually work

I’ve been researching AI Overviews since they launched. Here’s the data that should change how you think about this:

The citation opportunity:

Yes, overall CTR drops when AI Overviews appear. BUT - and this is crucial - links included IN the AI Overview receive MORE clicks than if they appeared as traditional blue links.

The research shows:

• 54.14% of top-3 AI Overview citations rank in Google’s top 10
• But 45.86% DON’T - meaning AI Overviews pull from beyond traditional top rankings
• 67.82% of ALL AI Overview citations don’t rank in the top 10

What this means:

Being cited in the AI Overview is the new “position zero.” Getting pushed below it is the new “page 2.”

The strategy shift:

Stop optimizing to rank #1 below AI Overviews. Start optimizing to BE CITED IN AI Overviews.

Different optimization targets. Different content structure. Different success metrics.

We pivoted our entire content strategy 8 months ago when we saw AI Overviews expanding. Here’s what’s working:

The “AI-Citable Content” Framework:

1. Answer-first structure - Put the direct answer in the first 50-100 words. Don’t bury the lead.

2. Question-based headers - Use H2s that mirror how users phrase queries. AI Overviews often pull from sections with matching headers.

3. Supporting evidence - Include specific data, examples, and steps. AI likes content it can quote accurately.

4. Comprehensive depth - Cover related questions in the same piece. AI uses “query fan-out” to answer multiple related questions from single sources.

5. Fresh updates - We update key pages monthly with new data or examples.

Results:

Pages following this framework appear in AI Overviews 3x more often than our legacy content. When cited, they also get cited in more prominent positions.

Here’s the uncomfortable truth nobody wants to hear:

Some informational content simply isn’t worth creating anymore.

If AI can fully answer a query and users have no reason to click, competing for that traffic is futile. You’re fighting for crumbs.

Our strategic pivot:

We’ve shifted clients toward:

1. Transactional content - Content that requires action beyond reading (tools, comparisons, deals)

2. Unique data/research - AI can’t cite data that doesn’t exist. Original research creates citation opportunities.

3. Experience-based content - First-hand experiences AI can’t replicate or easily summarize

4. Community content - User-generated content, discussions, real experiences

5. Decision-support content - Help users make specific decisions, not just understand concepts

The 80/20 rule:

80% of informational queries might now be better served by AI. Focus on the 20% where human-created content adds genuine value beyond what AI can synthesize.

Technical SEO perspective on AI Overview optimization:

What I’ve learned from analyzing citation patterns:

Google AI Overviews use your content differently than traditional search. The technical factors that seem to matter:

1. Page speed - Faster pages get crawled more frequently, keeping content fresh in Google’s index

2. Structured data - FAQ schema, HowTo schema, and Article schema help AI understand content purpose

3. Clean HTML - Semantic markup (proper headers, lists, tables) helps AI extract information

4. Internal linking - Pages with strong internal link context tend to be cited for broader topic queries

5. Canonical clarity - Make sure Google knows which version of your content is authoritative

Checking your AI Overview visibility:

Use Google Search Console’s Performance report - AI Overview traffic is included in the “Web” search type. You can also use Am I Cited to monitor citations across AI platforms including Google AI Overviews.

Track which pages get cited, not just which pages rank.

E-commerce perspective - we’re seeing something different:

AI Overviews are actually HELPING our commercial queries.

For “best [product category]” queries, AI Overviews generate brand awareness even without clicks. Users see our brand recommended by AI, then search for us directly later.

What we track:

• AI Overview citations for product categories
• Branded search volume (has increased 15% since AI Overviews expanded)
• Assisted conversions from users who saw AI mentions

The insight:

AI Overviews are becoming a top-of-funnel discovery channel. Even if they don’t click from the SERP, they remember brands AI recommended.

Strategy:

Focus on being the brand AI recommends, not just the page that ranks. Brand mentions in AI responses have value beyond direct clicks.

Publishing industry perspective - this is existential for us.

Our traffic analysis:

• How-to content: -35% CTR
• News/current events: -15% CTR (less affected)
• Evergreen educational: -42% CTR
• Opinion/analysis: -8% CTR (least affected)

What we’re learning:

Original perspective and analysis are the “moats” that AI can’t easily replicate. When content is pure information transfer, AI can do that better and faster.

Our pivot:

Investing heavily in:

• Original reporting
• Expert interviews
• Data journalism
• Opinion and analysis

Scaling back:

• Generic how-to guides
• Definition/explainer content
• Basic listicles

The content that AI struggles to generate is the content worth creating.

Local SEO angle that’s often overlooked:

AI Overviews for local queries are different.

For local searches (“dentist near me,” “best pizza downtown”), AI Overviews often show Maps integrations and local pack results rather than just text summaries.

What I’m seeing:

• Local business content still drives clicks because AI can’t replace the need to contact/visit
• Service pages with clear location signals get cited
• Reviews and reputation heavily influence local AI Overview citations

For local businesses:

This is actually less scary than for publishers. AI can synthesize information, but it can’t perform the service. Local businesses need to ensure they’re cited as the recommended provider, but the business model isn’t under threat the same way.

Focus on being the business AI recommends for your area and services.

Here’s how we’re reporting on AI Overviews to clients:

The new metrics stack:

1. AI visibility rate - % of target queries where we appear in AI Overviews (use Am I Cited or manual tracking)

2. Citation position - Where in the AI Overview our content is cited (top-3 vs. “show more”)

3. Share of AI voice - How often we’re cited vs. competitors for category queries

4. Zero-click value - Estimated brand impressions from AI Overview citations (even without clicks)

5. Downstream branded search - Are AI mentions driving branded search volume?

The narrative shift:

We’ve stopped reporting pure traffic metrics for informational content. Instead, we report on “AI search visibility” as its own channel.

Clients are starting to understand that being cited by AI is valuable even if it doesn’t show up in Google Analytics the same way traditional clicks do.

Taking the long view here:

AI Overviews are just the beginning.

Google is testing even more integrated AI search experiences. ChatGPT has 400M+ monthly active users. Perplexity is growing 20% month-over-month.

The question isn’t “how do we rank in traditional search” - it’s “how do we remain visible in an AI-mediated discovery landscape.”

What I’m advising clients:

1. Diversify discovery channels - Don’t be 100% dependent on any single platform

2. Build direct audience relationships - Email lists, communities, direct traffic sources

3. Create un-summarizable value - Tools, experiences, services that AI can recommend but not replace

4. Monitor AI visibility across platforms - Not just Google AI Overviews, but ChatGPT, Perplexity, Claude

5. Accept the new reality - Informational content’s value proposition has fundamentally changed

The SEOs who adapt will thrive. The ones who fight it will struggle.

This thread has given me more actionable insight than months of industry reading. Thank you all.

My key takeaways:

1. Being cited IN AI Overviews > ranking below them - The new position zero

2. Content structure matters more than ever - Answer-first, question-based headers, comprehensive depth

3. Not all content is worth creating - If AI fully answers it, find different content angles

4. New metrics required - AI visibility, citation position, share of AI voice, downstream branded search

5. Original value is the moat - Research, opinions, experiences that AI can’t replicate

What we’re changing:

• Auditing all client content for “AI citability”
• Building AI visibility tracking into reporting dashboards
• Pivoting informational strategies toward original research and unique perspectives
• Setting up Am I Cited monitoring across platforms

The uncomfortable reframe:

We’re not “losing traffic to AI Overviews.” We’re entering a new era where AI visibility is a distinct channel alongside (not replacing) traditional SEO.

Time to adapt the playbook.